Improving the Quality of Accident Investigation...of quality as consistency, depth, and time...

266
! Improving the Quality of Accident Investigation National Transportation Safety Board Office of Marine Safety Submitted to: Project Liaison: Michael Rosecrans Primary WPI Advisor: Natalie Mello WPI Advisors: David DiBiasio, Constance Clark Submitted by: Dane Mellan, Mason Ruffing, Victor Zeng Date: 12/18/2009

Transcript of Improving the Quality of Accident Investigation...of quality as consistency, depth, and time...

Page 1: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!""

Improving the Quality of Accident Investigation

National Transportation Safety Board Office of Marine Safety

Submitted to: Project Liaison: Michael Rosecrans Primary WPI Advisor: Natalie Mello

WPI Advisors: David DiBiasio, Constance Clark Submitted by:

Dane Mellan, Mason Ruffing, Victor Zeng Date: 12/18/2009

Page 2: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

#""

ABSTRACT The NTSB Office of Marine Safety (OMS) conducts accident investigations

to prevent the reoccurrence of accidents. We compared the OMS and other

agenciesʼ investigations through interviews and analysis of reports. We found

inconsistency of definitions and methods used in investigation is an area of

improvement for the OMS. We made recommendations to adopt the IMO causal

factors definition as probable cause definition in conjunction with E&CF analysis

to improve the quality of OMSʼ investigations by increasing focus on safety

issues.

Page 3: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

$""

ACKNOWLEDGEMENTS There are many individuals that we would like to thank and acknowledge, as without their support this project would not have been possible. We would like to express our deepest appreciation to Deputy Director Mike Rosecrans, of the National Transportation Safety Board Office of Marine Safety for sponsoring our IQP. His continual support and guidance throughout the eight-week process made this project a reality. Secondly, we would like to thank Director Jack Spencer, Office of Marine Safety for his support and encouragement throughout our time in Washington, DC. We would also like to express our gratitude to the members of the Board for their invested interest and encouragement throughout the completion of our project. Deborah Hersman Chairman National Transportation Safety Board Christopher Hart Vice Chairman National Transportation Safety Board

Robert L. Sumwalt Member National Transportation Safety Boar

We would like to express our deepest thanks to our advisors who continually provided us with guidance and support throughout the entire IQP process. Without these three individuals there was no possibility of this project reaching completion. We are very thankful for their help and very much obliged for their helping our project reach its fullest potential.

Natalie Mello Advisor Worcester Polytechnic Institute

Constance Clark Advisor Worcester Polytechnic Institute

David DiBiasio Advisor Worcester Polytechnic Institute

Additionally, we would like to thank all the Marine Investigators from the Office of Marine Safety that worked closely with us throughout our time at the National Transportation Safety Board. A special thanks to:

Page 4: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

%""

Rob Henry Chief of Marine Investigations Division A Larry Bowling Sr. Accident Investigator Thomas Roth-Roffy Sr. Accident Investigator William Woody Sr. Accident Investigator

Barry Strauch Chief of Marine Investigations Division B Liam LaRue Sr. Accident Investigator Eric Stolzenberg Sr. Accident Investigator Jim Scheffer Associate Director of Investigation Quality

In addition, we would like to thank these individuals from other modes of the National Transportation Safety Board for their assistance while we looked at alternative methods for Accident Investigation: John DeLisi Deputy Director Office of Aviation Safety-NTSB

Mark Bagnard Chief of Investigations Division Office of Highway Safety-NTS

Lastly, we would like to thank those representatives from international accident investigation agencies that took the time to speak with us regarding our project. We would like to express our gratitude to the following: Mike Walker Sr. Transport Safety Investigator Australian Transport Safety Bureau Ken Potter Manager of Head Office-Marine Investigation Operations Transportation Safety Board-Canada Wei Wang Marine Accident Investigator Maritime Safety Administration Peoplesʼ Republic of China

Captain Mabito Hamada Marine Accident Investigator Japan Transport Safety Board Alan L. Blume Deputy Commissioner of Maritime Affairs Republic of the Marshall Islands Sanny Shamoun Human Factors Specialist Statens Haverikommission-Sweden

Mike Travis Principal Inspector Marine Accident Investigation Branch United Kingdom

Page 5: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

&""

TABLE OF CONTENTS

!

ABSTRACT......................................................................................................................2 ACKNOWLEDGEMENTS................................................................................................3 TABLE OF CONTENTS...................................................................................................5 EXECUTIVE SUMMARY .................................................................................................7 1 INTRODUCTION.........................................................................................................10 2 BACKGROUND ..........................................................................................................12

2.1 FACTUAL COLLECTION.............................................................................................12 2.1.1 Witness Evidence ..........................................................................................13 2.1.2 Physical Evidence..........................................................................................14 2.1.3 Documentary Evidence..................................................................................15 2.1.4 Factual Collection Techniques.......................................................................16

2.2 CAUSAL ANALYSIS...................................................................................................18 2.2.1 Accident Causal Theories ..............................................................................18 2.2.2 Causal Analysis Techniques..........................................................................20

2.3 SAFETY RECOMMENDATIONS ...................................................................................25 2.4 HUMAN AND TECHNOLOGICAL RESOURCES ...............................................................26

2.4.1 Technological Resources...............................................................................27 2.4.2 Human Resources .........................................................................................28

3 METHODOLOGY........................................................................................................30 3.1 INTERVIEW PROTOCOL.............................................................................................30

3.1.1 Factual Collection ..........................................................................................32 3.1.2 Causal Identification.......................................................................................32 3.1.3 Safety Recommendation................................................................................33

3.1.4 HUMAN AND TECHNOLOGICAL RESOURCES ............................................................33 3.2 IDENTIFY ACCIDENT AGENCIESʼ INVESTIGATION PROCESSES ......................................34 3.2.1 NTSB .................................................................................................................34 3.2.2 INTERNATIONAL AGENCIES....................................................................................35 3.3 IDENTIFY THE HIGHEST QUALITY PROCESS FOR THE OMS .........................................35 3.3.1 COMPARISON OF INVESTIGATION PROCESSES ........................................................36 3.3.2 CONSIDERING THE DIFFERENCES BETWEEN AGENCIES............................................37 3.3.3 COMPARING INVESTIGATION PROCESS LENGTH ......................................................37

4 RESULTS....................................................................................................................39 4.1 ACCIDENT INVESTIGATION PROCESS AND APPROACH.................................................39 4.2 ACCIDENT INVESTIGATION METHODS ........................................................................43

Page 6: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

'""

4.2.1 Factual Collection ..........................................................................................46 4.2.2 Causal Analysis .............................................................................................50 4.2.3 Safety Recommendation................................................................................52 4.2.4 Human and Technological Resources ...........................................................61

5 CONCLUSIONS..........................................................................................................66 6 RECOMMENDATIONS...............................................................................................69 REFERENCES...............................................................................................................70 GLOSSARY ...................................................................................................................75 APPENDIX A .................................................................................................................78 APPENDIX B .................................................................................................................80 APPENDIX C ...............................................................................................................125 APPENDIX D, E, F, G..................................................................................................190 "

Page 7: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

(""

EXECUTIVE SUMMARY Transportation accidents can result in the loss of human lives and material

goods, along with the possibility of environmental damage. The United States congress

created the National Transportation Safety Board (NTSB) as an independent accident

investigation agency to prevent the reoccurrence of accidents. The NTSB conducts

accident investigations to determine safety issues. Using the knowledge of safety

issues, the NTSB can make recommendations to improve transportation safety. The

NTSBʼs accident investigation quality directly influences the effectiveness of the

recommendations made.

The NTSB Office of Marine Safety (OMS) strives to constantly improve their

accident investigation quality so the most effective recommendations can be made to

improve maritime safety. The NTSB OMS accident investigators conduct accident

investigations through the steps of factual collection, causal analysis, and safety

recommendations. To conduct accident investigations of the highest quality, accident

investigators must utilize the most effective methods for identifying safety issues.

The goal of our project is to identify the highest quality methods that are practical

to be implemented by the NTSB OMS. To do this we compared investigative methods

that are available to accident investigators. We found that there are many accident

investigation methods developed through academic research. The actually usability of

these methods based on field experience of accident investigators is rarely published.

To assess the effectiveness of accident investigation methods we chose three aspects

Page 8: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

)""

of quality as consistency, depth, and time efficiency. To compare the quality of accident

investigation methods, we conducted interviews with experienced accident investigators

from multiple renowned investigative agencies. In addition to our interviews, we

compared the accident investigation reports of multiple accident investigation agencies

to identify the strength of different agencies.

We identified many methods that are utilized by multiple accident investigation

agencies, which are yet to be implemented at the NTSB OMS. We found that both

Canadian transportation safety board and the Australian transportation safety bureau

have a standard definition of probable cause and causal analysis method amongst their

accident investigators. Both of these agencies use a definition similar to IMOʼs causal

factor definition, which define probable causes as safety issues of an accident. These

agencies also utilize a version of event & causal factor for their investigations. We found

that there is not a standard definition of probable cause and causal analysis method in

the NTSB OMS. During investigations the NTSB OMS assign a single safety issue as

the probable cause; this result in a difference of opinions amongst the NTSB OMS

investigators during causal analysis step of the accident investigation.

To improve the causal analysis step of the NTSB OMSʼ accident investigation.

We made a recommendation for the NTSB OMS to adopt a standard definition of

probable cause. We determine by using the IMO definition of causal factor from chapter

2.2 of MSC-MEPC.3/Circ.2 as a standard definition of probable cause will improve the

quality of the NTSB OMSʼ accident investigation. We believe by having the standard

Page 9: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

*""

definition of probable cause as all safety issues of an accident, will allow the

investigators to focus more in the development and implementation of

recommendations. We believe the NTSB OMS should also use a standard causal

analysis method such as the event & causal factors. The use of a stand causal analysis

method will allow the investigators to improve their ability to compare their objective

conclusions during casual analysis. We believe the adoption of a standard probable

cause in conjunction with standard causal analysis method will improve the quality of

the NTSB OMSʼ accident investigations.

Page 10: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!+""

1 INTRODUCTION Investigations of transportation accidents are necessary to identify the safety

issues of an accident and provide safety recommendations. The goal of these

recommendations is to prevent the reoccurrence of accidents and increase

transportation safety. One agency that conducts accident investigations in the United

States is the National Transportation Safety Board (NTSB). The United States Congress

created the NTSB as an independent agency to prevent the reoccurrence of

transportation accidents, and to increase safety in all modes of transportation.

International Marketing and Statistical confirmed that the United States

possesses 10,697,300 gross tons of marine shipping capacity (2004). Due to the large

amount of maritime transport that occurs in the U.S. and abroad, marine accidents do

occur. Therefore, the NTSB has an Office of Marine Safety (OMS), which conducts

investigations for maritime transportation accidents. As an independent agency, the

NTSB needs to look into all safety issues of an accident. The advancements in marine

technology, management, and oversight have increased the complexity of maritime

transportation. The OMS needs to utilize methods that account for these advancements.

In an effort to improve accident investigation methodology, experts of accident

investigation publish books and articles that provide guides to the accident investigation

process. These experts develop in-depth methods for factual collection, causal analysis,

and recommendation production. These experts create an extensive amount of

Page 11: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!!""

theoretical literature on accident investigation. However, they have not conducted many

studies on which methods investigators prefer to use.

In addition, investigators rarely publish books or articles on their investigative

experience. Unlike academic researchers, marine investigative agencies such as the

NTSB, the Transportation Safety Board (TSB) of Canada, and the Marine Accident

Investigation Branch (MAIB) of the United Kingdom do not make information on

improvements to their investigative methods readily available.

The goal of our project was to improve the quality of accident investigation for the

NTSB OMS. To do this, we looked into three aspects of quality that we identified as

consistency, depth, and time efficiency. Along with our research, we interviewed expert

accident investigators from several agencies on their accident investigation processes.

We also conducted analysis on published reports from several investigation agencies.

We then analyzed the data from our interviews and research to identify areas of

improvement for the OMS. We then created recommendations on specific methods to

help the NTSB OMS improve effective accident investigation according to consistency,

depth, and time efficiency.

Page 12: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!#""

2 BACKGROUND The Center for Chemical Process Safety defines an “accident” as “an unplanned

event or sequence of events that results in undesirable consequences” (1992, p.327).

Accident investigators conduct accident investigations with the goal of preventing

accidents from reoccurring. During these investigations, investigators gather facts about

the accident to determine safety issues. Unlike criminal investigations where the causes

are determined for assignment of blame, accident investigators at the NTSB strive to

identify safety issues within an accident using the knowledge of causes (Leveson, 2004,

P. 1). Accident investigation specialists have developed many theories and methods to

aid accident investigators in factual collection, causal analysis, and safety

recommendations. In addition, accident investigators utilize human and technological

resources to further aid their investigations.

2.1 Factual Collection Accident investigators cannot make any conclusions regarding an incident

without the collection of accident facts. Investigators gather witness evidence, physical

evidence, and documentary evidence as three forms of facts for accident investigations

(Brown, et al., 1995, p. 162; Hendrick & Benner, 1987, p. 140; Greenspan, et al., 1989,

p. 29). The accuracy of factual collection directly affects the validity of an investigatorʼs

causal determination. Investigators must use proper techniques for each form of

evidence to ensure the accuracy of their facts (Beitman, 2005, p. 147). Any modification

of accident sites or incorrect evidence may mislead an investigation. For example,

Page 13: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!$""

NTSB investigators noted several witnesses who believed a missile caused the 1996

aviation accident of Trans World Airline Flight 800. These witnesses claimed they saw a

missile hit the airplane prior to the explosion. The reliability and accuracy of their

statements were later unsubstantiated; investigators determined from physical evidence

that the explosion of a fuel pump caused the accident. The witnessesʼ statements

misled the investigation, which caused the NTSB to waste resources.

2.1.1 Witness Evidence Witnesses can be an unreliable source of evidence for an accident investigation

as previously shown; however, they are often useful for investigators when trying to

reconstruct the scene of an accident (Brown, et al., 1995, p. 125). Witnesses provide

their interpretations of the events from prior, during, and after an accident. Investigators

obtain clues from the witnessesʼ descriptions of the accident to identify safety issues.

For example, in the 2008 marine accident involving the sinking of the fish-processing

vessel the Alaska Ranger, witnesses stated the crew were all asleep during the

accident. These witnessesʼ statements led accident investigators to further study the

safety issue of the crewʼs sleep schedule during an expedition.

To ensure the credibility and usefulness of witness interviews, accident

investigators consider multiple factors prior to interviewing witnesses. Investigators

consider that witnesses might have a lack of knowledge of the events during an

accident, which would cause a witness to produce false conclusions that lack the

accuracy needed by accident investigators. Accident investigators also need to indicate

Page 14: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!%""

the date and time they interview each witness, because psychological factors can alter

witnessesʼ interpretations of accidents over time (Lloyd-Bostock et al, 1983, P. 3;

McAuliff & Kovera). Post-traumatic stress disorder is an example of a psychological

factor that affects witnesses after accidents. In addition to psychological factors, a

witnessʼ responsibility and relationship to an accident can affect their response during

an interview. A witnessʼ responsibility refers to whether a witness has ties to the

accident cause. A witnessʼ relationship to an accident refers to any connections a

witness may have to an accident whether financial or familial. Though accident

investigations conducted by the NTSB are not criminal investigations, nor intended to

assign blame to individuals or organizations, the nature of investigations may result in

witnesses being dishonest during interviews. With these factors in mind, accident

investigators determine the credibility of witnesses to assess the accuracy of their

statements as evidence. Investigators must keep in mind the subjectivity and bias

behind witness statements when using interviews as a form of evidence, and use other

forms of evidence to support the details collected from witnesses.

2.1.2 Physical Evidence Investigators often support the statements made by witnesses with physical

evidence, which is the most reliable source of information (Brown, et al., 1995, p. 167;

Greenspan, et al. p.30). Investigators look for any physical evidence such as video or

audio recordings from prior, during, or after the time of the accident to support

identification of safety issues. Audio and video recordings provide accident investigators

Page 15: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!&""

with vivid and credible details of the scene of an accident. Investigators also collect

pertinent information from voyage data recorders (VDR). Although a VDR is a type of

physical evidence, we further discuss the VDR as a technological resource in section

2.4.1. Using physical evidence along with witness statements, accident investigators

reconstruct an event timeline including events leading up to an accident through the

rescue and recovery operations of first responders.

Separate from video and audio recordings, physical evidence also includes

measureable data(Brown, et al., 1995, p. 167; Greenspan, et al. p.30).Investigators can

analyze this physical evidence using parametric equations. For example in the 2007

marine accident involving the passenger vessel the Empress of the North, investigators

measured the physical damage done to the vessel due to the allision with ice.

Investigators then analyzed the measurements using parametric equations to determine

the exact movement of the vessel upon contact. This helped the investigators identify

the navigational error of the pilot. Measurable physical evidence rarely degrades over

time due to human factors, but environmental factors may render physical evidence

perishable. These environmental factors require the timely gathering of some physical

evidence. By considering physical evidenceʼs integrity, investigators ensure its

credibility.

2.1.3 Documentary Evidence Unlike witness statements and physical evidence, documentary evidence is both

non-perishable and reliable. Accident investigators often gather documentary evidence

Page 16: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!'""

as supplementary information for improving accident investigation quality (Brown, 1995,

p. 126). Documentary evidence includes manufacturersʼ specifications and maintenance

histories. Investigators also use public and private records such as codes, standards,

and case histories as documentary evidence. Investigators gather some documentary

evidence from informational databases that organizations create and maintain to aid

investigations. For example in the 2008 marine accident involving the containership the

Cosco Busan, the medical history of the vesselʼs pilot allowed accident investigators to

determine the possible safety issue of granting an individual a license to operate a large

vessel while taking prescription medications that impair judgment. In this case,

documentary evidence led accident investigators to identify an additional safety issue

that influenced the accident scenario.

2.1.4 Factual Collection Techniques Accident investigators have adopted and developed many documentation and

storage techniques to improve the accuracy and credibility of factual collection. One

technique that is often used to document witness statements is the use of a storyboard.

Storyboards allow accident investigators to organize and summarize witness statements

(Walton, 2008). The use of a storyboard helps accident investigators to determine

witnessesʼ credibility through cross-referencing of all witness statements. Accident

investigators may identify convergences and corroborations of different statements,

increasing the statementsʼ credibility. Storyboards also help determine where

statements do not converge and investigators can then gather extra evidence to support

Page 17: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!(""

which statements true. The documentation of witness statements does improve the

credibility of witness evidence. However, investigators need credible physical evidence

to support these statements to ensure the accuracy of an investigation.

Accident investigators perform specialized techniques to maintain the credibility

of physical evidence. The Marine Accident Investigatorsʼ International Forum (MAIIF)

developed thorough procedures to follow in the documenting and storing of physical and

documentary evidence (Appendix E). These procedures stress the importance of the

mapping, photographing, and tagging of all evidence prior to removal from the accident

site. MAIIF also suggests a storage technique requiring careful packing of fragile

evidence components and the enclosure of evidence that may contain chemical residue

in non-absorbent material for future examination. Similar to physical evidence,

investigators must preserve the integrity of documentary evidence. The MAIIF states

documentary evidence should be securely stored for each investigation to avoid the

alteration of documents. In addition, paper duplicates of all documentary evidence

should be stored as backup in the case of an electronic failure.

Although investigators utilize techniques to help preserve the accuracy and

credibility of witness, physical, and documentary evidence, investigators also need to

organize all facts collected during an investigation. The use of an evidence matrix is

another technique used to document all forms of evidence gathered. The MAIIF also

has a proposed evidence matrix. An evidence matrix prompts accident investigators to

record information on the origin, purpose, and other details pertaining to evidence. The

Page 18: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!)""

organization of the evidence gathered makes the determination of probable cause an

easier process because the evidence is more accessible when in an easily recognizable

format.

2.2 Causal Analysis Accident Investigators conduct many methods of causal identification using the

facts collected during investigations (Brown, et al., 1995, p. 137). These methods of

causal identification allow accident investigators to identify causal factors of an accident

and assess safety issues relating to these factors. Many accident investigation

specialists have come up with accident causal theories to aid the understanding of the

occurrence of accidents.

2.2.1 Accident Causal Theories There are many accident causal theories created by accident investigation

specialists. Stellman writes that multiple theories such as pure chance theory, biased

liability theory, and accident proneness theory do not consider that causes of

investigations can be determined (1998, p. 56.6). These theories state that accidents

are inevitable. If accidents are inevitable, then their prevention is not possible; therefore,

accident investigators do not find these theories useful because they do not help with

the prevention of accidents. However, there are theories that do support investigatorsʼ

beliefs that safety recommendations can prevent accidents.

Page 19: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!*""

In 1931, H.W. Heinrich developed the first accident causal theory, the domino

theory, sometimes termed linear system models (Lundberg, et al, 2009). Domino theory

states that a series of failures is the cause of an accident (Ridley & Channing, 2003, p.

286; Thygerson, 1977, p.44). Heinrich believed the removal of a single failure in the

chain may terminate an accident. One accident model that is similar to domino theory is

the “Swiss cheese” model. This model is a representation of a chain of failed or absent

defensive regulations and safety mechanisms that line up to cause an accident

(Reason, 1997). An example of this model is shown below in figure 2.1.

Figure 2.1 Swiss Cheese Model

Figure 2.1 Reasonʼs “Swiss cheese” model (Kauffield and Luxhoj, 2003)

This model is similar to domino theory because one of the slices of cheese can

be modified to stop the chain of events from happening. However, the Swiss cheese

model is different because it specifically considers failures in safety regulations and

management systems whereas the domino theory considers other failures as well. The

Page 20: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

#+""

Swiss cheese model is extremely applicable for investigation agencies that make

recommendations on how to improve safety regulations. Another causal theory that

helps determine the failure of safety measures is the energy transfer theory, which

investigators also refer to as the complex linear system model (Karmis, 2001, p. 44;

Kjellen, 2000, p. 32). Energy transfer theory states that accidents occur from an initial

hazard that exceeds the barrier of safety measures or accident prevention.

While these theories can help for most accidents, for large-scale accidents,

investigators often use multiple causation theory as a platform for casual identification

(Petersen, 1971, p. 13). Investigators who support the multiple causation theory believe

the convergence of multiple failures causes accidents, which is why investigators also

refer to it as complex interaction theory (Lundberg, et al, 2009). Although these theories

are all useful in accident investigation, investigators often combine the beliefs behind

these theories with the utilization of causal identification techniques when identifying

causal factors.

2.2.2 Causal Analysis Techniques Investigators often use causal identification techniques in conjunction with causal

theories. In some cases, investigation experts develop these techniques based on

causal theories. Causal identification techniques aid accident investigators with the

organization of the facts collected and the determination of causal factors. Some

examples of causal identification techniques that accident investigators use are

Page 21: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

#!""

Management Oversight & Risk Tree (MORT), Why-Because analysis, Hazard Barrier

and Target analysis (HBT), and Events and Causal Factors analysis (E&CF).

Management Oversight & Risk Tree (MORT) is a causal identification technique

developed to identify safety issues within accidents (Johnson, 1973). Johnson created

MORT based on multi-causal theory and uses logic gates in a tree format of analysis.

MORT is a complex system accident investigators can utilize to identify root causes of

an accident. Rather than explaining the intricacies of this intensive method, a power

point presentation explaining the MORT method in detail has been included in Appendix

F.

Peter Ladkin used why-because analysis to identify causal factors of the

Glenbrook, NSW rail accident that occurred on December 2, 1999 (Ladkin, 2005). Why-

Because analysis is an approach to accident investigations where you pick an event

and ask, “why” did this occur? From the resulting “because”, or reason, the question of

“why” is asked again. The why-because diagram created during the Glenbrook accident

analysis is shown in figure 2.2 below.

Page 22: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

##""

Fig 2.2 Why-Because Analysis

Figure 2.2 Why-Because analysis from Glenbrook NSW rail accident

In this diagram, each box is labeled with a number that corresponds to the order

investigators analyzed each box. Investigators asked the initial “why” for box 0 and the

resulting “because” was box 1. Investigators continued this why-because analysis until

they felt that all major issues were identified.

Page 23: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

#$""

Hazard, Barrier, and Target analysis (HBT) is another causal identification

technique. HBT is modeled after the energy transfer theory, which differentiate it from

E&CF. Mechanical, kinetic, thermal, chemical, electrical, radiant are the means of

energy source analyzed in HBT (Goldberg, et al, 1994). Using HBT, accident

investigators may determine when the prevention barrier was breached during the

accident. Barrier can take forms as physical, management, and oversight. Accident

investigator may find the problem within a specific barrier to raise possible safety issue

for future recommendation.

Event and Causal Factors analysis (E&CF) is a cause identification technique

used by accident investigators to verify causal relationships and the event timeline

sequence (SCIENTECH, Inc., 1995). Investigators also use E&CF analysis to organize

the accident facts collected. A generic example is included in figure 2.3 on the following

page (Integrated Publishing). This example shows a timeline of events that leads up to a

fireball burning John while he is changing a fuse. The E&CF analysis organizes the

events into the correct order and for significant events, the causal factors leading to

these events are assessed. For example, event 11 was caused by factors 11.1 and

11.2.

Page 24: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

#%""

2.3 E&CF Example

Page 25: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

#&""

Accident investigators utilize the causal theories and causal identification

techniques above with the goal of creating safety recommendations that can effectively

prevent the reoccurrence of accidents.

2.3 Safety Recommendations The final step of the investigation process for most investigative agencies is the

creation of recommendations for the accident investigation report. This step is an

important part of the investigation process because the recommendations can prevent

the reoccurrence of accidents. Agencies such as the NTSB issue recommendations

based on the safety issues identified in the causal analysis. Wood summarized that for

these recommendations to be of any use the recommendations must be feasible (as

stated in Ferry, 1988, p.236). Whether the regulatory agency can practically implement

and maintain the changes made from a recommendation determines the usefulness of

the recommendation.

The first determining factor, the practical implementation of the recommendation,

breaks down into the cost and time involved. For a recommendation to produce positive

results, it must be affordable to the majority of the interest groups it directly affects.

Should the recommendation yield a costly change, some interest groups may be unable

to meet the financial requirement. The intended outcome of any safety recommendation

is the inclusion of all interest groups; therefore, investigators must consider cost when

issuing such recommendations. In addition to the cost, investigators must take into

consideration the time necessary to implement the changes when creating a

Page 26: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

#'""

recommendation. If a recommendation is going to take multiple years to implement,

then investigators may consider adding restrictions or adaptations to the

recommendation. Investigators determine a compromise between cost and time to

create the most practical recommendations possible.

In addition to creating a practical recommendation, investigators must take into

consideration the ability to maintain the changes implemented by the recommendation.

“Recommendations could include review of current policy, new policy, re-training

personnel on existing requirements, or additional training needs” (OSHAcademy, n.d.).

Investigators must take into account the practicality of the maintenance necessary when

determining the recommendation to issue.

The purpose of investigative agencies such as the NTSB is to produce

recommendations based on the safety issues identified through the investigation. “The

finest report fails if it merely states facts and draws conclusions. Corrective actions are

needed, and the report should identify them (Ferry, 1988, p. 244).” Therefore, an

accident investigatorsʼ goal is to issue practical recommendations that can effectively

prevent the reoccurrence of accidents.

2.4 Human and Technological Resources Investigators use a variety of tools when conducting an investigation. These tools

range from photogrammetry techniques to the analysis of Voyage Data Recorders

(VDR). These tools may also include individuals who are experts in specific fields such

as human factors, law, or naval architecture. Investigators use these tools to create a

Page 27: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

#(""

complete picture of the entire accident in order to determine the safety issues that lead

to the accident occurring.

2.4.1 Technological Resources During the late 1990ʼs, there was a surge of investigation agencies that wanted

the IMO to require large ships to carry voyage data recorders (Brown, 1999). A voyage

data recorder (VDR), also called the black box of a ship, records information regarding a

shipʼs position, heading, radar, bridge audio, and other information. VDRs can record

this information from up to 12 hours prior to an accident. These VDRs are especially

helpful in investigations because of the data that they provide. There are limitations,

though, because these VDRs were only required to have on passenger ships and ships

with larger than 3000 gross tonnage that were built on or after July 1, 2002

(International Maritime Organization, 2002).

Along with VDRs, advancements in organizational software have made it

possible for accident investigation agencies to create accident databases. These

databases contain categorized information regarding all accidents that the agency

investigates. One agency that has such a database is the United States Coast Guard

(USCG). Their “marine accident database” contains information about all accidents

concerning their geographical distributions and also other related and important

information about each accident (Hottendorff & Hullmeine, 1992). Other agencies also

have databases similar to the USCGʼs such as the European Maritime Safety Agency.

This database will be part of the European Marine Casualty Information Platform

Page 28: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

#)""

(EMCIP) (European Maritime Safety Agency, 2008). This database is a joint effort from

the countries with marine investigation agencies in the European Union. This database,

which EMSA is still constructing, will allow the exchange of information regarding the

facts collected in each accident investigation.

While EMSAʼs database will only include this information, the International

Maritime Organization (IMO) has created a database known as the Global Integrated

Shipping Information System (GISIS). This system includes information ranging from

marine casualties and incidents to condition assessment schemes and simulator

information (International Maritime Organization, 2005). This database is important to

the advancement of marine investigation because it provides information regarding

marine investigation techniques and marine accidents to all member agencies. The

investigators who access the GISIS can even search for accidents by region and vessel

name.

2.4.2 Human Resources Although these technological advancements have created a more

collective marine accident investigation community, investigation agencies still need to

have organization of human resources to produce timely and accurate evaluations of

accidents to make databases credible. To utilize human resources, agencies should hire

investigators with specialties in fields of law, naval architecture, navigation, human

factors, and engineering with experience on the sea (Marine Accident Investigators

International Forum, 2003). The Marine Accident Investigation Branch (MAIB) of the

Page 29: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

#*""

United Kingdom organizes their investigators into teams with four specialists, one from

each of four categories similar to the ones listed above. The MAIB can launch one of the

teams on any investigation and each team will have a specialist from each category.

This greatly reduces the need for the investigators to collaborate with experts from

outside of the investigation team. Although this may be true, investigators must also

keep up-to-date within their specialty fields. Investigators often attend workshops and

seminars to keep current on the latest trends in special areas of accident investigation.

The Marine Accident Investigators International Forum (MAIIF) and IMO sometimes

organize these workshops through the agencies that have experience to get newly

formed agencies up to date on current investigation procedures.

Page 30: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

$+""

3 METHODOLOGY The goal of our project is to improve the quality of the NTSBʼs current method of

marine accident investigation. We identified the different steps of the accident

investigation process to determine the protocol shown in Appendix A. We created this

protocol to collect information regarding the depth of investigations, the amount of time

taken to do investigations, and the consistency of each step of the process during an

investigation. We interviewed investigators from all modes of transportation at the NTSB

and other marine investigative agencies such as the Australian Transportation Safety

Board (ATSB), the Canadian Transportation Safety Board (TSB), the Marine Accident

Investigation Branch of the United Kingdom (MAIB), the Marshall Islands Maritime

Authority (MIMA), and the Swedish Accident Investigation Board (SHK). We then

compared and analyzed the data from these investigation agencies to identify the

highest quality marine accident investigation process specifically for the NTSB Office of

Marine Safety (OMS).

3.1 Interview Protocol The first step of our methodology was to identify the steps of the accident

investigation process and determine factors to describe the quality of an accident

investigation. We determined that the three aspects that contribute to the quality of an

accident investigation that we would research are consistency, depth, and time

efficiency. The consistency of an accident investigation refers to how often the

investigators at an agency use the same method for each step of the investigation

Page 31: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

$!""

process each time they perform an accident investigation. If an agency does not use a

consistent method it is possible for investigators to identify different causes of an

accident. For example during the causal identification step one method might lead an

investigator to discover one cause of an accident; however, if the investigator used a

different method, a different cause may have been found. A consistent method guides

investigators to reach conclusions through an objective process. Although consistency

helps to raise the quality of an accident investigation, without depth, consistency does

not make an investigation worthwhile. The depth of an investigation refers to the amount

of detail that investigators cover during an investigation. The amount of depth that an

investigation covers is important in determining the quality of the investigation. An in-

depth investigation allows investigators to determine safety issues related to system

errors, rather than just human factors and electro-mechanical failures. Although depth is

important, investigators can investigate too in-depth and take too long to complete and

issue their recommendations. If investigators go too in-depth they waste resources such

as time. Time is an important factor because a time efficient investigation provides

recommendations to the public as soon as possible and can help prevent the

reoccurrence of accidents. We collected information regarding the following steps of the

accident investigation process so that we could identify the highest quality accident

investigation for the NTSB OMS according to these three aspects: consistency, depth,

and time efficiency.

Page 32: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

$#""

It was important for us to determine the steps of the accident investigation

process so that we could create the protocol we used to interview investigators from the

agencies mentioned in section 3. The steps of the investigation process that we

identified are factual collection, causal identification, and safety recommendations. In

addition to these steps, we looked into the utilization human and technological

resources. We developed the questions in our protocol to identify any techniques and

methods that the agencies we interviewed currently use.

3.1.1 Factual Collection Gathering accident facts is an important part of the accident investigation. The

questions for our protocol that targeted this step of the investigation process are 2-6 of

our protocol in Appendix A. The questions involved first asking about the cooperation

between the investigation agency and the first responders to the accident. Second, we

asked questions to find out if any of the agencies we interviewed had specific methods

or techniques for collecting evidence. Lastly, we asked if any of the agencies we

interviewed had methods for identifying when the investigators gathered enough

evidence to determine safety issues.

3.1.2 Causal Identification The second step of the accident investigation is causal identification. The

questions we asked regarding this step of the investigation are numbers 7-10 of our

protocol in Appendix A. We created the first of these questions to identify if each of the

agencies we interviewed had a set definition for probable cause. Second, we needed to

Page 33: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

$$""

identify if there are any methods used by each agency to determine the probable

causes of an accident and how long the analysis of the accident facts takes in an actual

investigation. Lastly, we asked questions to try to understand how each agency knows

when investigators have identified the causal factors. Investigators determine causal

factors to identify safety issues and produce recommendations that can prevent causes

from reoccurring and help prevent future accidents.

3.1.3 Safety Recommendation The third step of the accident investigation process is the production of

recommendations by the investigators to help prevent the reoccurrence of accidents.

For investigators to provide these recommendations to the public, they must construct

accident investigation reports that support their reasoning for regulatory agencies to

implement the recommendations. Question number 14 from our protocol in Appendix A

pertains to the production of these accident reports. We asked this question to

determine if investigators feel that the report-writing step of the investigation process

takes too long. If there is redundancy in this process, then the regulatory agencies can

activate the recommendations sooner, which helps prevent the reoccurrence of

accidents. While these three main steps of the investigation process are important, we

felt that we should also investigate how the NTSB is currently utilizing human and

technological resources to provide additional suggestions on the improvement of the

OMS accident investigation process.

3.1.4 Human and Technological Resources

Page 34: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

$%""

We directed questions 11-13 from our protocol in Appendix A towards

understanding how the agencies we interviewed utilize human and technological

resources. We asked the first question to understand how investigators keep up with the

technological advances over time. We asked the second question to determine if any of

the agencies we interviewed use any software programs to help manage or aid their

investigations. We also investigated database programs to help organize the NTSBʼs

accident data. We developed the interview protocol in Appendix A to interview the

agencies discussed in section 3.

3.2 Identify Accident Agenciesʼ Investigation Processes We interviewed investigators using the protocol discussed in sections 3.1.1-3.1.4

to identify accident investigation processes currently used by agencies such as the

NTSB, the Canadian TSB, the Chinese Maritime Authority, the United Kingdom MAIB,

the Marshall Islands (MIMA), and the Swedish SHK. We conducted email, face-to-face,

and phone interviews with investigators from these agencies to identify the investigation

processes that these agencies currently use. We conducted these interviews to analyze

the consistency, depth, and time efficiency of the investigative processes that each

agency uses.

3.2.1 NTSB During our interviews, we considered three different modes of transportation

accidents that the NTSB investigates: OMS, Office of Aviation Safety (OAS), and Office

of Highway Safety (OHS). Our objective was to identify the accident investigation

Page 35: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

$&""

processes used by each of these three offices within the NTSB. The office of aviation at

the NTSB has existed since 1967 and has performed over 124,000 aviation accident

investigations (National Transportation Safety Board [NTSB], 2004, History and Mission,

para. 6). The experience and success of the aviation branch has helped to create a high

quality process of investigation for aviation accidents. The number of successful

recommendations that the office of aviation safety has created sets this quality.

Identifying the accident investigation processes that the NTSB uses for each mode of

transportation was important to research so that we could compare many different

possible results.

3.2.2 International Agencies Along with identifying the accident investigation processes that the different

offices of the NTSB use, we also identified accident investigation processes that

international agencies use. The international agencies that we interviewed are the

Canadian TSB, the Chinese Maritime Authority, the United Kingdom MAIB, the Marshall

Islands (MIMA), and the Swedish SHK. We included brief descriptions of these agencies

and our interview outlines in Appendix B. It was important for us to identify the

processes that other international agencies use because the NTSB recognizes these

agencies for their excellence in accident investigation.

3.3 Identify the Highest Quality Process for the OMS

Page 36: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

$'""

We identified the highest quality marine accident investigation process for the

NTSB OMS according to the three components of quality we determined we would

consider: consistency, depth, and time efficiency. To achieve our goal of identifying the

highest quality process for the NTSB we decided to provide recommendations to the

NTSB on the areas of their investigation process that they could improve to increase the

effectiveness of their investigations. To provide these recommendations we first

compared and analyzed the investigation processes that we identified from our

interviews. We performed a simple content analysis to organize the responses from

these interviews. We then analyzed the content by comparing the answers to each other

using consistency, depth, and time efficiency as our basis for defining the highest quality

techniques and methods from the investigation processes. To determine which of these

identified techniques and methods were usable by the NTSB OMS we had to perform

extra research outside of the interview data that we collected. We considered the

differences between the agencies that we interviewed and also the time efficiency of the

report issuing for the agencies that we interviewed.

3.3.1 Comparison of Investigation Processes The first analysis that we conducted was a content analysis of the data that we

collected during our interviews. We separated the data from each interview into the

topics discussed in sections 3.1.1-3.1.4. We then created excel spreadsheets with x-

coordinates that represented each interviewee and y-coordinates that represented each

response that the investigator could have responded with. We created an excel

Page 37: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

$(""

spreadsheet for each of the topics/questions that we asked. We show and discuss

these spreadsheets in our Results section. The content analysis that we performed was

important to organize the data that we obtained through our interviews. We then

analyzed the data by comparing the methods and techniques from each step of the

investigation processes based on which methods would fit the NTSBʼs OMS.

3.3.2 Considering the Differences between Agencies To determine the highest quality process for accident investigation, we had to

consider the differences between agencies to determine which techniques and methods

are applicable to the NTSB OMS. We considered differences such as size of the

agency, the number of cooperators involved in the on-site evidence gathering step, and

the approach of accident investigations. The size of the agency affects the applicability

of techniques because the NTSB does not have the number of personnel to sustain

techniques that larger agencies employ. While considering the number of personnel

involved in an accident investigation, we also considered the number of personnel that

the NTSB sends to an accident investigation scene compared to the number that other

agencies send to the scene. Apart from the size of the investigation agencies, we also

considered the approaches behind the investigations that each agency completes. We

compared the recommendations created by the Chinese MSA, which is a regulatory

agency, with the OMS recommendations using keywords that relate to compliance with

regulations.

3.3.3 Comparing Investigation Process Length

Page 38: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

$)""

We obtained data regarding the time efficiency of the total report writing times for

six different agencies. The six different agencies that we considered include the ASTB,

three branches of the NTSB, the MAIB, the SHK, the TSB, and the New Zealand

Transportation Accident Investigation Commission (TAIC). We looked up the amount of

time taken to write each of the past ten reports from each agencyʼs website. Then we

entered this data into an excel spreadsheet to calculate the average report time for each

agency. We then considered several different factors that might affect the investigation

length such as the number of investigators in each agency, the number of reports

produced in the past three years, the number of recommendations each report made,

and the number of recommendations that have been implemented. We did this special

comparison between the NTSB OMS and MAIB because the MAIB has a significantly

shorter investigation process.

Page 39: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

$*""

4 RESULTS We conducted research in areas of accident investigation and interviewed

several expert accident investigators to improve the quality of investigation at the NTSB

OMS. Through our research and interviews, we identified the process and approach of

accident investigation used by different offices of the NTSB and foreign agencies.

During our interviews with accident investigators from these agencies, we also identified

specific methods used for each step of the investigations. In addition to understanding

how these agencies conduct accident investigations, our interviewee provided us with

opinions on effectiveness of certain accident investigation methods along with insights

on possible areas of improvement. Using all of the information gathered through our

research and interviews, we were able to determine some areas of improvement for the

NTSB OMS to achieve a more effective accident investigation according to consistency,

depth, and time efficiency.

4.1 Accident Investigation Process and Approach Based on our research, we identified two types of accident investigation

agencies. There are authoritative agencies that have power to regulate marine

transportation such as the MSA, and the MAMI. The others are Independent agencies

that do not have power to regulate marine transportation, such as the NTSB, the MAIB,

the SHK, the TAIC, and TSB. Both authoritative and independent agencies conduct

accident investigation with the goal of improving marine safety. The distinct difference

between authoritative and independent agencies is their accident investigation

Page 40: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

%+""

approach. Accident investigation approach defines accident investigatorsʼ mindset

during investigation. Independent agencies clearly define their approach through

disclaimers within their reports, as shown below.

“The Independent Safety Board Act, As codified at 49 U.S.C Section 1154(b),

precludes the admission into evidence or use of Board reports related to an incident or

accident in a civil action for damages resulting from a matter mentioned in the report.” –

the NTSB

“The Transportation Safety Board of Canada (TSB) investigated this occurrence

for the purpose of advancing transportation safety. It is not the function of the Board to

assign fault or determine civil or criminal liability.” - TSB

“SHK investigates accidents and incidents with regard to safety. The sole

objective of the investigations is the prevention of similar occurrences in the future. It is

not the purpose of this activity to apportion blame or liability.” – the SHK

“Regulations 2005 shall be the prevention of future accidents through the

ascertainment of its causes and circumstances. It shall not be the purpose of an

investigation to determine liability nor, except so far as is necessary to achieve its

objective, to apportion blame.” – the MAIB

“It is not a function of the ATSB to apportion blame or determine liability.

However, an investigation report must include factual material of sufficient weight to

support the analysis and findings. At all times the ATSB endeavors to balance the use

Page 41: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

%!""

of material that could imply adverse comment with the need to properly explain what

happened, and why, in a fair and unbiased manner.” – the ATSB

“The Transport Accident Investigation Commission is an independent Crown

entity established to determine the circumstances and causes of accidents and

incidents with a view to avoiding similar occurrences in the future. Accordingly, it is

inappropriate that reports should be used to assign fault or blame or determine liability,

since neither the investigation nor the reporting process has been undertaken for that

purpose.” –the TAIC

Those disclaimers above show the independent agenciesʼ approach to accident

investigation, which is not to assign faults or blames. Authoritative agencies does not

follow the same approach because their executive duties. Authoritative agencies also

affect each accidentʼs outcome due to their duties of execution of regulation and the

responsibility of search and rescue. The approach of not assigning faults and blames

allow accident investigators of independent agencies to be able to evaluate all safety

issues of an accident without biases.

Page 42: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

%#""

Table 4.1 Accident Report NTSB OMS

Number of Key Word

Length of Report

Athena 106 5 68 Ethan Ellan 9 54 Cosco Busan 13 147 Axel Spirit 5 37 Kition 0 36 Shuttle II 4 57 Empress of the North 20 113 Alaska Ranger 2 83 Crown Princess 3 95 Queen of the West 1 21 Average 6.2 71.1 Median 4.5 68

Table 4.1 shows the number of time key words (violate, violation, comply, compliance) appeared in 10 of the latest major NTSB OMS accident reports. This table also includes the length of each respective report in pages.

Table 4.2 Accident Report MSA

Number of Key Word

Length of Report

Dubai World and Zhelinyu 2261 4 10 CSCL NingBo and JinHaiDa 18 15 Jin Sheng and Golden Rose 11 18 Han Jin Gothenburg and Chang Tong 6 24 Afflatus and Wen Yue 5 23 CMB Biwa and Lu Ri Yu 1608 8 23 An Jin 1 14 New Ferry 85 and Dong Qu No.1 7 21 Harvest and Jin Hai Kun 8 21 BBC Ontario and Zhexiangyu 48038 2 10 Average 7 17.9 Median 6.5 19.5

Table 4.2 shows the number of time key words (violate, violation, comply, compliance) appeared in 10 of the latest major MSA accident reports. This table also includes the length of each respective report in pages.

Page 43: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

%$""

Independent agencies conduct accident investigations without bias, which allows

them to achieve more depth in their investigations compared to authoritative agencies.

The MAMI publishes the results of their accident reports in the form of safety advisories.

The MAMIʼs safety advisories do not include information on factual collection and cause

analysis; the lack of this information does not allow the public to scrutinize their work.

However, some authoritative agencies do publish their accident investigations such as

the MSA. According to the last 10 major reports provided by both the NTSB OMS and

the MSA, the NTSB produced in average a longer report in pages. We do not believe

the length of an accident investigation report directly influences the depth. Based on our

analysis of accident investigation reports, the keys words violate, violation, comply, and

compliance are used in context with regulations. Within the shorter MSA reports, there

is in average more use of those key words described. This data shows the MSA reports

focus more on the enforcement of regulation compared to the content of NTSB OMS

reports. The longer length of NTSB OMS report shows the NTSB OMS discusses safety

issues other than the area of regulation compliance within their report. In summary, the

analysis between the reports of the NTSB OMS and MSA provided support that

independent agencies achieve more depth in accident investigation compared to

authoritative agencies.

4.2 Accident Investigation Methods As described previously, the independence of accident investigation agencies

greatly improves the depth of their accident investigations by providing accident

Page 44: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

%%""

investigators an unbiased investigation approach. To further improve the quality of the

OMS accident investigation according to the three aspects of quality that we considered,

we have interviewed expert investigators from many agencies. We asked questions

regarding the factual collection, causal analysis, and safety recommendations steps of

the overall accident investigation process. We have conducted a content analysis on

these investigatorsʼ responses to questions of our interview protocol (Appendix A).

Table 4.3 Content Analysis

NTSB International Agencies 1 Rob Henry OMS 2 Barry Strauch OMS 3 Tom Roth-Roffy OMS 4 Larry Bowling OMS 5 John Delisi OAS 6 Mark Bagnard OHS 7 Ken Potter TSB 8 Sanny Shamoun SHK 9 Alan Blume MIMA 10 Mike Travis MAIB 11 Mike Walker ATSB 12 Mabito Hamada JTSB

Gathering Evidence Evidence Checklist 1 2 3 4 5 6 7 8 9 10 11 12 TOTAL Uses Evidence Checklist X X X X 4 No Evidence Checklist X X X X X X 6 N/A X X 2 Checklist (Designation) 1 2 3 4 5 6 7 8 9 10 11 12 TOTAL Agency Designated Checklist X X 2

No Designated Checklist X X X X X X X 6 N/A X X X 3

Page 45: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

%&""

Evidence Priority (First) 1 2 3 4 5 6 7 8 9 10 11 12 TOTAL Witness X X X X X 5 Physical X X X X 4 Documentary 0 N/A X X X 3 Determining Probable Cause

Definition of Probable Cause 1 2 3 4 5 6 7 8 9 10 11 12 TOTAL

Uses a Probable Cause Definition X X X X 4

Does not Use a Probable Cause Def. X X X X X X X X 8

Probable Cause= P.C. 1 2 3 4 5 6 7 8 9 10 11 12 TOTAL Agency Has definition for P.C. X X 2

Agency does not have P.C. Def. X X X X X X X X 8

N/A X X 2 Causal Analysis 1 2 3 4 5 6 7 8 9 10 11 12 TOTAL Designated Method X X X 3 Non-designated Method X X X X X X X 7 N/A X X 2 Causal Analysis Method 1 2 3 4 5 6 7 8 9 10 11 12 TOTAL Event Cause Timeline X X X X X X 6 Producing Recommendations

Report Writing 1 2 3 4 5 6 7 8 9 10 11 12 TOTAL Redundancy in Report Writing X X X X X 5

No redundancy X X X X X X 6 N/A X 1 Utilizing Human and Technological Resources

Human 1 2 3 4 5 6 7 8 9 10 11 12 TOTAL Cooperative learning within the agency X X 2

Workshops/seminars X X X X X X X X X X 10

Page 46: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

%'""

Table 4.3 shows the content analysis of our interview. Where existence of key content with responses are mark with checks according to respective interviewee.

4.2.1 Factual Collection We have identified factual collection of accident investigation is done initially

through gathering of evidence in the forms of documentary, physical, and witness. To

evaluate the consistency of evidence gathering, we interviewed investigators on

whether their agencies provide a documented methodology for factual collection. From

our responses, we determined evidence checklists are the only form of documented

methodology for factual collection. Evidence checklists consist of items that

investigators need to gather during the factual collection step of accident investigations.

We found that the TSB is the only agency that provides a checklist for their

investigators, and the TSB modeled their checklists specifically for different accident

types. (Potter, personal communication, 11/17, 2009) 30% of the interviewees

On the Job Training 4 N/A X 3 Software 1 2 3 4 5 6 7 8 9 10 11 12 TOTAL Utilizes software X X X X X 5 Does not Utilize Software X X X X X X 6 N/A X 1 Software is Useful X X X X X 5 Software is Not Useful X X 2 N/A X X X X X 5 Kind of Software Used 1 2 3 4 5 6 7 8 9 10 11 12 TOTAL Scene reconstruction Software X 1

Simulation Software X 1 Causal Event Software X X 2 Organizational Software X X X X 4 N/A X X X X X 5

Page 47: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

%(""

responded with the usage of evidence checklists. These accident investigators utilize

either personalize checklists or adaptations of other organizationsʼ checklists. (Bowling,

personal communication, 11/19, 2009; Shamoun, personal communication, 11/19,

2009)

Other than the 30% of the accident investigators who utilize checklists, the rest of

the accident investigators responded that they do not use checklists during their

investigations, or do not care for their usage. For those investigators who do not utilize

checklists, some stated the usage of evidence checklists might be constraining

(Strauch, personal communication, 11/2, 2009). These accident investigators believe

the usage of evidence checklists is constraining because the possibility of overlooking

important evidence that are not within the checklists. There are other statements of

aversion regarding evidence checklist related to its time efficiency (Travis, personal

communication, 12/3, 2009). These accident investigators oppose the usage of

evidence checklists because if all evidence of checklists is gathered, accident

investigators may dramatically lengthen the duration of factual collection. As shown in

appendix C the TSBʼs evidence checklists contain approximately 1500 items in total.

These statements provided insight into why some accident investigators dislike the

usage of evidence checklist.

Although some investigators dislike the use of a checklist, some accident

investigators do utilize evidence checklists for factual collection (Potter, personal

communication, 11/17, 2009). These interviewees stated accident investigators could

Page 48: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

%)""

use evidence checklist as a guideline for their factual collection. In addition, their

responses stated that evidence checklists could prevent accident investigators from

overlooking simplistic evidence, which can cause accident investigators to revisit

accident site after the initial on-site investigation. Although the investigators we

interviewed did not agree on if a checklist is useful, we did find that all investigators,

whether they dislike and support the use of evidence checklists, converged in one

common conclusion. This conclusion is evidence checklists can serve as an aid

especially for inexperienced accident investigators.

Along with our questions regarding the use of an evidence checklist, we have

also analyzed investigatorʼs opinions on which form of evidence should receive priority

during accident investigations. From the intervieweesʼ responses, we found 50% of the

accident investigators prioritize witness evidence. These investigators stated multiple

field situations that correspond to witness evidenceʼs degradability. Multiple accident

investigators who are human factor specialists stated, even though accident

investigation of independent agencies is not interested in determining blame, over time,

witnesses may modify their responses to associate less involvement in the accident with

themselves (Strauch, personal communication, 11/2, 2009). Other human psychological

factors accident investigators mentioned are as described in section 2.1.1. Accident

investigators did state a non-psychological factor that results in the degradability of

witness evidence. The majority of the crewmembers and other personnel who are

involved in accidents may serve as witnesses and have duties after the occurrence of

Page 49: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

%*""

an accident. These witnessesʼ obligations may cause accident investigators to have

difficulties in approaching them (Blume, personal communication, 11/19, 2009). If

accident investigators do not conduct interviews in a timely manner, they may not be

able to make contact with the desired witnesses.

Second to the support of witness evidence, 25% of the accident investigators

prioritize physical evidence. Some accident investigators stated in the case of marine

accidents, vessels typically maintain their location after the accident (Roth-Roffy,

personal communication 11/10, 2009). In addition, environmental factors do not harm

majority of physical evidence on vessels, thus timeliness of physical evidence gathering

does not take the same priority like other forms of investigations. Those accident

investigators who prioritize physical evidence stated that the factors mentioned in

section 2.1.2 can perish physical evidence (Henry, personal communication, 10/30,

2009) Accident investigators also mentioned, due to the current regulations, physical

evidence such as VDRs are not in the sole jurisdiction of the accident investigation

agencies. There is a possibility the vessel owners may modify devices such as the VDR

prior to when accident investigators are able to arrive on scene.

Accident investigators have full jurisdiction over documentary evidence due to

their subpoenas. Thus, no accident investigators chose documentary evidence as their

priority. Some investigators stated, in some cases documentary evidence may be the

only form of evidence that accident investigators can gather in accident investigations

due to the complete loss of vessels. These accident investigators raised the point that

Page 50: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

&+""

oftentimes, the prioritization of evidence gathering is situational. Because accidents vary

case by case, 25% of the accident investigators did not provide a response to their

preference of evidence priority.

4.2.2 Causal Analysis Causal analysis is a step of accident investigation, where accident investigators

use the factual data of the accidents to determine safety issues through probable

causes. To determine the safety issues of accidents, accident investigator must have a

definition of probable cause. Without a definition of probable cause, accident

investigators will not have a defined goal while conducting causal analysis. From our

interviews, we have determined that TSB and ATSB are the only two agencies with a

clear definition of probable cause that their whole agency utilizes. Both the TSB and

ATSB use a definition that is similar to the IMO definition of causal factors (Potter,

personal communication 11/17, 2009; Walker, personal communication, 12/10, 2009).

The IMO definition was recently changed to a definition where all safety issues within an

accident are considered as causal factors.

From our interviews, we have determined that the NTSB OMS does not have an

organizational wide accepted definition for probable cause. Some accident investigators

of the NTSB OMS have adopted their own personal definitions. One investigator uses

the definition of probable cause adopted from the USCG (Bowling, personal

communication, 2009). The USCG defines their definition of probable cause as the root

cause of an accident.

Page 51: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

&!""

With a set definition of probable cause, accident investigators may use causal

analysis methods to determine safety issues of accidents. Fifty percent of the

investigators we interviewed stated that the usage of a causal analysis method is not

necessary. These investigators believe they are able to determine the safety issues

using purely their expertise (Strauch, personal communication, 11/2, 2009). Accident

investigators stated that MORT is a time consuming method of causal analysis. From

our research we have found that MORT has around 1500 items that are arranged into a

fault tree as shown in Appendix F. The fifty percent of investigators who do use causal

analysis methods each stated a different method. However, through further research we

determined each different method is an adaptation of the Events & Causal Factors

method (E&CF). From our interviews, we have gathered that the ASTB, the MAIB, and

the TSB all highly recommend their accident investigators to use a form of E & CF

analysis. E & CF is explained in detail in section 2.2.2. An example of the MAIBʼs usage

of E & CF is shown in Appendix G. In addition to the use of the E & CF method, the TSB

and the SHK recommended investigator to use Why-Because analysis throughout the

investigation (Potter, personal communication, 11/17, 2009; Shamoun, personal

communication, 11/14, 2009). These agencies believe investigators should maintain a

“why-because mindset”. From our research, why-because analysis is a method

developed to ensure accident investigation objectivity (Ladkin, 2005). Why-because

analysis uses factual data to determine causal factors in a very similar manner as the

E&CF method.

Page 52: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

&#""

During our research, we found that HBT method is documented in all of

the accident investigation guidebooks. However, we did not find any accident

investigators who utilize HBT through our interviews. HBT is a causal analysis method

use to analyze prevention barriers within accidents. One investigator of the NTSB stated

during the Exxon Valdez accident investigation, the NTSB board utilized HBT for the

creation of a recommendation (Woody, personal communication, 12/2, 2009). In this

case, the NTSB board utilized HBT to determine the capability of a prevention barrier in

reaching the desired level of safety.

4.2.3 Safety Recommendation Independent agencies notify the public of safety issues, and provide

recommendations to improve safety through accident investigation reports. The NTSB,

ATSB, MAIB, SHK, and TAIC are all independent agencies and have identical purposes

for their accident investigations. They strive to use unbiased facts to determine accident

causes. We were unable to fully assess the depth of each agencyʼs investigation reports

due to the scope of this project. Therefore, we identified each agencyʼs depth as what

they strive to achieve based on their mission and investigative approach.

Based on our determination that each independent agency strives for an

identical depth, we were able to compare the time efficiency of the whole accident

investigation process. This time efficiency is the total amount of time it takes an agency

to publish an accident report with recommendations from the day of the accident. We

have compiled the time in weeks for the last 10 major accident investigations by marine,

Page 53: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

&$""

aviation, and highway office of the NTSB along with the ASTB, the MAIB, the SHK, the

TAIC, and the TSB.

Page 54: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

&%""

Table 4.4 The NTSB Office of Marine Safety Report Time

Brief Accident Description Report Time (In Weeks) Athena 106 35 Ethan Ellan 42 Cosco Busan 67 Axel Spirit 70 Kition 76 Shuttle II 76 Empress of the North 77 Alaska Ranger 79 Crown Princess 80 Queen of the West 84 Average 69 Median 76

Table 4.4 shows the amount of time in weeks it took the NTSB OMS to publish an accident report after the accident. There is a large variance of the total time. The median is higher than the average due to few accidents that took exceptionally shorter time to complete.

Table 4.5 The NTSB Office of Aviation Safety Report Time

Brief Accident Description Report Time (In Weeks) Boeing 767-200, N799AX 52 Aerospatiale SA365N1, N92MD 56 Embraer ERJ-170, N862RW 60 Jet CL600-2B19, N8905F 62 Cessna 500, N113SH 73 Eurocopter AS350B2, N613TV 79 McDonnell Douglas DC-9-82, N454AA 80 Cessna 310R, N501N 81 DHC-6-100, N203E 111 Cessna Citation 550, N550BP 119 Average 77.3 Median 76

Table 2 shows the amount of time in weeks it took the NTSB OAS to publish an accident report after the accident. There is a large variance of the total time.

Page 55: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

&&""

Table 4.6 The NTSB Office of Highway Safety Report Time

Brief Accident Description Report Time (In Weeks) Collision, Upper Pittsgrove Township, New Jersey 19 Run-Off-The-Bridge and Rollover, Sherman, Texas 64 Rollover, Mexican Hat, Utah 67 Collapse Highway Bridge Minneapolis, Minnesota 67 Override of Elevated Exit Ramp Atlanta, Georgia 71 Railroad Crossing Accident in Elmwood Park, Illinois 133 Chain-Reaction Collision, Lake Butler, Florida 134 Rollover Following Collision, Osseo, Wisconsin 152 Override Following Collision, Huntsville, Alabama 156 Run-Off-the-Road and Rollover Turrell, Arkansas 202 Average 106.5 Median 102

Table 3 shows the amount of time in weeks it took the NTSB OHS to publish an accident report after the accident. There is a large variance of the total time.

Table 4.7 The MAIB Report Time

Brief Accident Description Report Time (In Weeks) Pacific Sun 15 Jo Eik 29 Vallermosa 39 Maggie Ann 40 Stena Voyager 42 Ville de Mars 42 Riverdance 42 HMS Westminster & Princess Rose 51 Eurovoyager 53 Abigail H 54 Average 40.7 Median 42

Table 4 shows the amount of time in weeks it took the MAIB to publish an accident report after the accident. There is a large variance of the total time.

Page 56: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

&'""

Table 4.8 The TSB Report Time

Brief Accident Description Report Time (In Weeks) L'Acadien II 32 Algomarine 39 Fireboat 08-448B 55 Big Sister 59 Sea Urchin 60 Skalva 68 Fair Jean 89 Queen of the North 97 Sichem Aneline 102 Kometik 135 Average 73.6 Median 64

Table 5 shows the amount of time in weeks it took the TSB to publish an accident report after the accident. There is a large variance of the total time. The median is lower than the average due to few accidents that took exceptionally longer time to complete.

Table 4.9 The SHK Report Time

Brief Accident Description Report Time (In Weeks) Astral 47 Tinto 47 SFC-7153 48 Listerland 49 Team Joker 52 MT Prospero 53 St. Erik 64 Atlantis Alvarado 83 Finnbirch 109 MT Bervik 146 Average 69.8 Median 52.5

Table 6 shows the amount of time in weeks it took the SHK to publish an accident report after the accident. There is a large variance of the total time. The median is lower than the average due to few accidents that took exceptionally longer time to complete.

Page 57: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

&(""

Table 4.10 The ATSB Report Time

Brief Accident Description Report Time (In Weeks) Van Gogh 37 Atlantic Eagle 44 MSC Lugano 44 Saldanha 48 Great Majesty 48 Malu Sara 49 Francoise Gilot 54 Iron King 62 Northen Fortune 67 Enterprise 90 Average 54.3 Median 48.5

Table 7 shows the amount of time in weeks it took the ATSB to publish an accident report after the accident. There is a large variance of the total time. The median is lower than the average due to few accidents that took exceptionally longer time to complete.

Table 4.11 The TAIC Report Time

Brief Accident Description Report Time (In Weeks) Pursuit 40 Cruise Cat 41 Santa Maria II 51 Kimihia 52 Anatoki and Lodestar Forest 60 Shikari 67 Taharoa Express 101 Kotuku 101 Mildford Sovereign 111 Walara K 119 Average 74.3 Median 63.5

Table 8 shows the amount of time in weeks it took the ATSB to publish an accident report after the accident. There is a large variance of the total time. The median is lower than the average due to few accidents that took exceptionally longer time to complete.

Page 58: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

&)""

Figure 4.1 Comparison of Report Time.

Figure 1 shows the comparison between the average and median of time in weeks it took each respective agency to produce an accident report after the accident. The MAIB has an overall lowest time for producing reports compare to the rest of the agencies. The NTSB OHS has the highest overall time in producing reports. Though the ATSBʼs time in producing reports is longer than MIAB, it also has a lower overall time compare to some of the other agencies.

The NTSB OMS takes approximately 6 months longer in average to produce an

accident investigation report compare to the MAIB. As determined, both NTSB and the

MAIB strive for the same depth during their accident investigations. Because the MAIBʼs

excellent time efficiency in producing accident investigation reports, we were especially

interested in comparing those reports of the NTSB OMS and the MAIB.

Page 59: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

&*""

Table 4.12 The NTSB OMS vs. MAIB Comparison

The NTSB OMS The MAIB Number of Investigators 10 20 Number of Investigations Between 2007-09

10 70

Average Investigation Time 69 40.7 Average Number of Findings 15.3 12.2 Average Number of Recommendations 4.3 3.7 Average Report Length 71.1 48.3 Recommendation Acceptance Rate 74.6% 92.7% Table 4.12 shows the comparison between the NTSB and the MAIB regarding the number of investigators, the total number of investigation conducted between the years of 2007 to 2009, and the average time in weeks it takes for 10 of the latest major accident reports to be published. This table also shows the average number of recommendations and findings for 10 of the latest major accident reports for both the NTSB OMS and the MAIB. The recommendation acceptance rate shown for the NTSB is gathered form the total acceptance rate of recommendations to date. The MAIB acceptance rate is base on the MAIB annual reports from 2003-08.

We have found that the MAIB has two times the amount of accident

investigator compare to the NTSB OMS. The MAIB produce 7 times the amount of

accident investigations compare to the NTSB OMS. The MAIB produce slightly fewer

findings and recommendations compare to the NTSB OMS per accident investigation.

The average length of the MAIBʼs accident investigation report is shorter than those of

the NTSB OMS. As stated previously, the length of accident investigation reports do not

directly influence their depth. We also compared the acceptance rate of

recommendations between the NTSB and the MAIB. The NTSB conducted 3 accident

investigations in the year of 2008, with the issuance of 8 recommendations in total. We

were unable to determine an acceptance rate for the 8 recommendations. We gathered

the acceptance rate for all 2342 of the NTSB OMSʼ recommendations, which is 74.6%.

Page 60: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

'+""

We have found the MAIB produce 117 recommendations in 2008, and 635 from 2003 to

2008. The acceptance rate of the 635 recommendations is 92.7%. Base on these data,

we believe further assessment of the difference between the methods used by the

NTSB OMS and the MAIB should be conducted.

We believe there is a possibility for improvements on the report writing and

recommendation step of the NTSB OMSʼ accident investigations. During our interviews,

we collected data regarding whether or not each investigator believes there is a

redundancy during the report-writing step of their accident investigations. All of the

investigators of the NTSB OMS responded during their interview that they believe there

is a redundancy in their report writing process. One accident investigator stated the

accident investigation board contributes to 1/3 of the overall report writing time (Strauch,

personal communication, 11/2, 2009). Accident Investigators stated that though the

report revision process of the NTSB is long, but it ensures the quality of the NTSBʼs

reports (Henry, personal communication, 10/24,2009). Two investigators stated there is

a lack of agreement between accident investigators and management regarding report

content (Roth-Roffy, personal communication, 11/10, 2009; Bowling, personal

communication, 11/19, 2009).

Base on our case study of the cosco busan, difference of opinion on report

content amongst accident investigators resulted from their determination of probable

cause. Multiple accident investigators believe the pilot was the probable cause of the

accident. In the final report the NTSB OMS stated the master was the probable cause.

Page 61: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

'!""

One of the board member during this investigation believed the VTS and USCG should

be included as probable causes. We believe there is a considerable difference in the

determination of probable cause amongst accident investigators during the NTSB OMSʼ

accident investigation.

Through our comparison of accident investigation reports of the independent

agencies, we have found a difference in report content. Those reports produce by the

MAIB, the TSB, and the ATSB have an additional section termed “action taken”.

Through our interviews we have found that the MAIB, the TSB, and the ATSB discuss

and implement some recommendations of their accident investigation prior to the

publication of the accident investigation report (Travis, personal communication, 12/3,

2009). The “action taken” section of these agenciesʼ reports specifically addresses the

implementation of recommendations. The MAIB accident investigator we interviewed

stated the MAIB is able to tailor their recommendation to be better suited for the

recipients by working closely with them before the publication of the accident

investigation report.

4.2.4 Human and Technological Resources

In addition to methods of accident investigation, we conducted interview in

utilization of human resources and technology by the accident investigation agencies.

Keeping investigatorsʼ knowledge of accident investigation techniques, marine

engineering, and safety systems up to date is an important responsibility of accident

Page 62: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

'#""

investigation agencies. Investigators must be knowledgeable about technological

advances to perform accurate accident investigations. We asked about accident

investigator training to determine effectiveness of the NTSB OMSʼs current methods are

for educating investigators. Currently the NTSB OMS accident investigators stated the

training they receive is “pretty good” (Bowling, Personal Communication, 11/19, 2009).

From our interviews, we determined that all but one agency we have interviewed utilize

workshops and seminars to train investigators in specific areas of development. Some

accident investigators believe cooperative learning between investigators could help to

improve the effects of workshops and seminars (Roth-Roffy, personal communication,

11/10, 2009). Through cooperative learning, accident investigators that attend

workshops and seminars could share beneficial information with the other accident

investigators. One problem with workshops is the limitations of budget provided for

training. Some accident investigators stated that the problem with budget is due to

timeliness of dispersal. Some workshops that the investigators would like to attend are

already passed, when the available funding arrives (Bowling, personal communication,

11/19, 2009).

While workshops and seminars are one way that investigators can stay current

with technology, other methods of trainings are available. From our interviews with the

Page 63: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

'$""

foreign agencies, we have found many methods of training for accident investigators.

Both TSB and MAMI utilize the method of “on the job training” (Potter, personal

communication 11/17, 2009; Blume, personal communication, 11/19, 2009). Both

agencies send their investigators on observational trips on liquid natural gas carriers.

Other than on the job training, an accident investigator of the JTSB noted that MAIIF is a

very useful conference for accident investigators to exchange their skills (Hamada,

personal communication, 12/11, 2009). This Japanese investigator is attending English

training seminar just so he may attend MAIIF in the future.

Other than training of accident investigators, we have gathered other forms of

human resource utilization through our interviews. Those investigator of the NTSB OMS

stated that the NTSB office of engineering help perform forensic testing for accident

investigations (Bowling, personal communication, 11/19, 2009). The NTSB OHS stated

during the investigation of Minnesota bridge collapse, the FBI aided majority of the

engineering study that dramatically speed up the accident investigation (Bagnard,

personal communication, 11/19, 2009). The NTSB AHS also mentioned the support of

FBI, specifically in the area of factual collection (Delisi, personal communication, 11/4,

2009). Investigators of TSB utilize third party firms to support their analysis of physical

evidence during accident investigations (Potter, personal communication, 11/17, 2009).

Page 64: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

'%""

Accident investigation agencies may utilize technology to synergistically aid

accident investigatorʼs work. Computer software is a form of technology that can help

accident investigators with file management and analysis of accident investigations.

From our interview we gather the NTSB OAS and OHS utilize simulation software for

scene reconstruction of accidents (Delisi, personal communication, 11/4, 2009;

Bagnard, personal communication, 11/19, 2009). As stated previously the NTSB office

of engineering conducts a portion of the forensics analysis for the NTSB OMS. The

NTSB offices of engineering specifically conduct analysis of VDRs using MADAS.

MADAS is the only simulation software use for marine accident reconstruction. From our

interviews, we have determined all foreign marine accident investigation agencies utilize

MADAS.

There is also software to help accident investigators with causal analysis step of

their investigations. TSB provide software for accident investigators to create evidence

diagrams (Potter, personal communication, 11/17, 2009). This software TSB utilize

ultimately allow the investigators to conduct the entirety of why-because analysis on the

computer. Although the TSB utilizes this software, majority of accident investigators that

we interviewed do not find this technology useful. Some accident investigators from the

NTSB OMS do believe creating event causal chats on the computer isnʼt different

compare to using pen and paper (Strauch, personal communication, 11/2, 2009; Roth-

Roffy, personal communication, 11/10, 2009; Henry, Personal Communication, 10/30,

Page 65: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

'&""

2009). Accident Investigators from the MAIB also does not find causal identification

software useful (Travis, personal communication, 12/3, 2009). One investigator also

commented that causal analysis software is expensive, being around 300 dollars per

license (Bowling, Personal Communication, Nov. 19, 2009).

We also looked into other computer technological resources such as

informational database for accident investigation. Through our interviews with NTSB

marine investigators, we found that some investigators believe the NTSB lacks a

database for them to see trends within accidents (Stolzenberg, personal

communication, 12/9, 2009). Accident investigators explained the NTSB never built a

database for accident trends analysis due to the low numbers of accident investigation

conducted yearly. Currently, with the globalization of communication between marine

accident investigation agencies, the IMO has established the GISIS. The USCG and

NTSB provide their reports to the GISIS along with all marine investigative agencies that

are members of the IMO (Scheffer, personal communication, 12/11, 2009). Although

this database is comprehensive of all accidents submitted by IMO member agencies,

data mining to find trends in accidents is difficult. When searching the database, an

investigator must search either by region or vessel name. The database does not have

the capability to search by accident type. Further utilization of GISIS may help the NTSB

accident investigators in their cause analysis.

Page 66: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

''""

5 CONCLUSIONS Through the analysis of our interviews and research, we identified some areas of

improvement for the NTSB OMS. We considered the methods for each steps of the

accident investigation process and the approach of accident investigation agencies. In

addition, we looked into the utilization of human and technological resources. Though

each of these areas has possible improvements, we believe improvements of the cause

analysis step of the accident investigation would be the most practical for the NTSB

OMS. To solve the inconsistency of causal analysis, we recommend the NTSB OMS

adopt a standard probable cause definition. By using a standard definition, the NTSB

OMS can reduce discrepancies amongst accident investigators in their determination of

probable cause. We believe the NTSB OMS should adopt the IMO causal factor

definition as their definition of probable cause. The IMO causal factor definition is shown

in chapter 2.2 of Appendix D.

The IMO definition defines a causal factor as any causes within an accident that

are safety issues pertaining to the occurrence of the accident. The NTSBʼs mission is to

prevent future reoccurrences of accidents through the issuance of safety

recommendations. To accomplish the NTSB OMSʼ mission, accident investigators must

make recommendations to prevent reoccurrences of pertaining safety issues. Thus, the

identification of all causal factors, as defined by the IMO, allows the accident

investigators to create effective safety recommendations.

Page 67: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

'(""

Using the IMO causal factors definition as the standard probable cause definition

will allow the accident investigators of the NTSB OMS to improve the consistency of

their accident investigation. A goal of NTSB OMS accident investigators is to define the

probable cause of an accident. By having a standard definition, the NTSB OMS accident

investigators will have a common goal in investigation. Currently the NTSB OMS

accident investigators traditional believe probable cause as a single safety issue, and

the rest of the safety issues are contributing causes. Accident investigators of the NTSB

OMS have different background and expertise, which cause them to each safety issues

differently. We believe the determination of a single safety issue as probable cause over

others is subjective, and is one of the causes of difference in opinions amongst

investigators. We believe adopting IMO causal factor definition will also further ensure

the independency of the NTSB OMS. By not assigning a single safety issue as probable

cause eliminates the possibility of third parties from inferring the assignment of probable

cause as blame.

Maritime safety can only be achieve with the contribution of all those involved in

marine transport. It is important for the NTSB OMS to cover all levels of depth in their

accident investigation, in effort to make helpful recommendation to everyone involved in

marine transport. The use of IMO definition of causal factor as the standard definition of

probable cause will assure all those involved in accident and those who are capable in

aiding the prevention of accident to be addressed as probable. Overall, this will allow

Page 68: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

')""

the NTSB OMS to have the capability of achieving the highest depth required to make

effective recommendations.

The usage of the IMO causal factor definition as the standard probable cause

definition will also increase the time efficiency of accident investigation. As explain

previously, the difference of opinion amongst accident investigators resulted in

redundancy of report-writing step of accident investigation. We believe the usage of the

IMO causal factor definition will eliminate the redundancy, and preventing the elongation

of time in report production. With all safety issue addressed as probable cause, accident

investigators may spend more time on making the most effective recommendation.

Determination of which recommendation to make is an objective process as explained

in section 2.3.

To further ensure the consistency, depth, and time efficiency of accident

investigation, we recommend that all NTSB OMS accident investigators to use event &

causal factor method in conjunction with the IMO causal factor definition as the probable

cause definition. From our interview, we have determined that many investigators are

already using event & causal factor method for cause analysis. Event & causal factor

method ensures objectivity of cause analysis, because accident investigators must draw

conclusion from facts. By using this method accident investigators may be able share

logics of their causal analysis, thus improve cooperation.

Page 69: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

'*""

6 RECOMMENDATIONS To the NTSB OMS:

R-1. Adopt MSC-MEPC.3/Circ.2 Chapter 2.2 as standard probable cause

definition

R-2. Utilize events & causal factors analysis to determine probable causes for all

accident investigations.

Page 70: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

(+""

REFERENCES

Bagnard, M. (2009, November 19). Chief of Investigations Division OHS. (D. Mellan, Interviewer)

Blume, A. L. (2009, November 19). Deputy Commissioner of Maritime Affairs. (D. Mellan, Interviewer)

Bowling, L. (2009, November 19). Sr. Accident Investigator. (V. Zeng, Interviewer)

Brown, M. T. (1999, May 5). Marine Voyage Data Recorders - International Symposium on Transportation Recorders. Retrieved December 13, 2009, from National Transportation Safety Board: http://www.ntsb.gov/Events/symp_rec/proceedings/authors/brown.pdf

Brown, S., Fromson, D., LeMay, I., Lyons, J., Svalbonas, V., Sweet, J., et al. (1995). Forensic Engineering, Part I an introduction to the investigation, analysis, reconstruction, causality, prevention, risk, consequence, and legal aspects of the failure of engineered products. Humble: ISI Publications.

Center for Chemical Process Safety. (n.d.). Retrieved October 2009, from Center for Chemical Process Safety CCPS: http://www.aiche.org/ccps/

DeLisi, J. (2009, November 4). Deputy Director OAS. (V. Zeng, Interviewer)

European Maritime Safety Agency. (2009, 3 19). European Marine Casualty Information Platform. Retrieved December 13, 2009, from European Maritime Safety Agency:

Page 71: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

(!""

http://www.emsa.europa.eu/end185d007d003d002d008.html

Ferry, T. S. (1988). Modern Accident Investigation and Analysis. Canada: John Wiley & Sons, Inc.

Goldberg, B., Everhart, K., Stevens, R., Babbitt, N., Clemens, P., & Stout, L. (1994). System Engineering "Toolbox for Design-Oriented Engineers. NASA.

Gould, J. (2000). Review of Hazard Identification Techniques. Sheffield, South Yorkshire, U.K.: Crown.

Hamada, M. (2009, December 11). Marine Accident Investigator.

Henry, R. (2009, October 30). Chief of Marine Investigations Division A. (V. Zeng, Interviewer)

Hottendorff, T., & Hullmeine, M. (1992, August). U.S. Coast Guard Marine Accident Database. Retrieved December 13, 2009, from Geodesy and Geomatics Engineering UNB: http://gge.unb.ca/Pubs/TR159.pdf

Integrated Publishing. (n.d.). Figure C-1. Events and Causal Factors Chart. Retrieved December 13, 2009, from Integrated Publishing: http://www.tpub.com/content/doe2/nst1004/nst10040039.htm

International Maritime Organization. (2002). Voyage Data Recorders. Retrieved December 13, 2009, from International Maritime Organization: http://www.imo.org/Safety/mainframe.asp?topic_id=768

Johnson, W. (1973). Management Oversight and Risk Tree -

Page 72: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

(#""

MORT. Scoville, ID: Aerojet Nuclear Co.

Karmis, M. (2001). Mine health safety management. SME.

Kauffield, K., & Luxhoj, J. T. (2003). Evaluating the Effect of Technology Insertion into the National Airspace System. Retrieved December 17, 2009, from The Rutgers Scholar: http://rutgersscholar.rutgers.edu/volume05/luxhoj-kauffeld/luxhoj-kauffeld.htm

Kumar, V., Gavrilova, M. L., Tan., C. J., & L'Ecuyer, P. (2003). Computational science and its applications. New York, New York: Springer.

Kjellen, U. (2000). Prevention of accidents through experience feedback, volume 55. CRC Press.

Ladkin, P. B. (2005, December 19). Why-Because Analysis of the Glenbrook, NSW Rail Accident. Retrieved December 17, 2009, from Bielefeld University - Faculty of Technology: http://www.rvs.uni-bielefeld.de/publications/Papers/Ladkin-Glenbrook.pdf

Marine Accident Investigators' International Forum. (2003, November 7). Meeting 12. Retrieved December 13, 2009, from Marine Accident Investigators' International Forum: http://www.maiif.org/meetings/maiif12_minutes.htm

National Transportation Safety Board. (2009). Special Publication. Retrieved 10 9, 2009, from www.ntsb.gov/publictn/gen_pub.htm

OSHAcademy. (n.d.). Establishing Findings and Developing Recommendations. Retrieved December 2009, from

Page 73: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

($""

OSHAcademy Occupational Safety & Health Training Network: http://www.oshatrain.us/notes/findingsrecommendations.html

Petersen, D. (2008). Techniques of safety management. Madison: McGraw-Hill.

Potter, K. (2009, November 17). Manager of Head Office-Marine. (D. Mellan, Interviewer)

Rollenhagen, C., Hollnagel, E., & Lundberg, J. (2009). What-You-Look-For-Is-What-You-Find – The consequences of underlying accident models in eight accident investigation manuals. Safety Science , 47 (10), 1297-1311.

Roth-Roffy, T. (2009, November 10). Sr. Accident Investigator. (V. Zeng, Interviewer)

Scheffer, J. (2009, December 11). Associate Director of Investigation Quality. (M. Ruffing, Interviewer)

SCIENTECH, Inc. (1995, August). Events and Causal Factors Analysis. Retrieved December 13, 2009, from College of Engineering & Engineering Technology at NIU: http://www.ceet.niu.edu/tech/asse/tech482/trac14.pdf

Shamoun, S. (2009, November 19). Humans Factors Specialist. (V. Zeng, Interviewer)

Stolzenberg, E. (2009, December 9). Sr. Accident Investigator. (V. Zeng, Interviewer)

Strauch, B. (2009, November 2). Chief of Marine Investigations Division B. (D. Mellan, Interviewer)

Page 74: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

(%""

Travis, M. (2009, December 3). Principal Inspector. (D. Mellan, Interviewer)

Walker, M. (2009, December 10). Sr. Transport Safety Investigator. (D. Mellan, Interviewer)

Woody, W. (2009, December 12). Sr. Accident Investigator. (M. Ruffing, Interviewer)

Page 75: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

(&""

GLOSSARY Allision: when a moving object strikes a stationary object

AMSA: Australian Maritime Safety Authority

AS: Aviation Safety

ATC: Air Traffic Control

ATSB: Australian Transportation Safety Board

AVDR: Audio/Video Data Recorder

CG: Coast Guard

E&CF: Events and Casual Factors

EMCIP: European Marine Casualty Information Platform

EMSA: European Maritime Safety Agency

FAA: Federal Aviation Administration

FBI: Federal Bureau of Investigation

FDR: Flight Data Recorder

FMEA: Failure Mode and Effect Analysis

GAO: Government Accounting Office

GISIS: Global Integrated Shipping Information System

HAZOP: Hazard Operability Analysis

HBT: Hazard, Barrier, and Target Analysis

HQ: Headquarters

Page 76: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

('""

HS: Highway Safety

IIC: Investigator In Charge

IMO: International Maritime Organization

ISIM: Integrated Safety Investigation Methodology

JTSB: Japan Transport Safety Board

LNG: Liquid Natural Gas

LNGC: Liquid Natural Gas Carrier

MADAS: Marine Accident Data Analysis System

MAIB: Marine Accident Investigation Branch

MAIIF: Marine Accident Investigators International Forum

MCARMI: Maritime and Corporate Administrator of the Republic of the Marshall Islands

MIMA: Marshall Islands Maritime Authority

MOU: Memorandum Of Understanding

MS: Marine Safety

MSA: Maritime Safety Administration

MORT: Management Oversight & Risk Tree

MTO: Man-Technology-Organization

NASA: National Aeronautics and Space Administration

NTSB: National Transportation Safety Board

OAS: Office of Aviation Safety

Page 77: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

((""

OHS: Office of Highway Safety

OJT: On the Job Training

OMS: Office of Marine Safety

PI: Principle Issue

PIMM: Principle Issue Management Model

PMI: Principle Maintenance Inspector

POI: Principle Operations Inspector

PRC: Peoplesʼ Republic of China

RADAR: Radio Detection And Ranging

RPM: Revolutions Per Minute

SATB: Swedish Accident Investigation Board

SHK: Statens Haverikommmission

TAIC: Transport Accident Investigation Commission

TIMS: Total Information Management System

TSB: Transportation Safety Board

US: United States

USCG: United States Coast Guard

VDR: Voyage Data Recorder

VTS: Vessel Traffic Service

Page 78: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

()""

APPENDIX A Interview Protocol

1. Gather specific information of the interviewee. (Name, Agency, Job Description, Years in current position, Contact information) Confirm previously gathered information with interviewee.

a. Please give further detail of your position?

2. What agency conducts the initial observation of the accident site?

a. What is your opinion of your agencyʼs relationship with the agency identified by the previous response?

b. Does the agency of initial observation provide adequate information prior to your agenciesʼ investigation?

3. What is the typical routine of your on-site investigation?

4. How often are accident sites revisited for gathering of additional facts?

5. How do you make sure all the facts are gathered?

6. Does your agency have a documented methodology for collecting accident facts?

7. How do you define a probable cause?

8. How are probable causes determined using the collected facts?

a. Is there a method that you particularly prefer? Why?

b. How many weeks does it typically take for the analysis identified by the previous response to be done?

Page 79: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

(*""

9. How do you determine if you have found all the probable causes of accident investigations?

10. Does your agency have a documented methodology for probable cause analysis?

11. How do investigators keep themselves up to date with technological advances in investigation, engineering, and system dynamics?

12. What areas of training are you personally interested in receiving?

13. Does your agency use any software to manage or aid your investigations?

14. Do you find that there is redundancy in the report writing part of your investigations?

15. Do you have any information regarding accident investigations such as methods developed by your agency to improve the thoroughness of investigations that would help our project?

16. May we contact you for additional questions?

Page 80: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

)+""

APPENDIX B "

INTERVIEW TRANSCRIPTS

NATIONAL TRANSPORTATION SAFETY BOARD (UNITED STATES) "

OFFICE OF MARINE SAFETY ROB HENRY !+,$+,+*"

-./01"21"345/60"788/906:"/6;0<:/=4:/26">04?@7"

A>BC"!*"D045<"

EB-F"##"D045<"

"

G3B".4<"46"49;46:4=0"2;05"2:.05"?290<"H084I<0"-F"920<"J50K/?/645D"469"826<IK:<"L/:."I<M"

"

B2?0"21":.0"9/11/8IK:/0<"125"I<"450":.0"/<2K4:/26"469"50?2:0"K284:/26<"21"<2?0"488/906:<"<I8."4<"NI:8.".45H25"L./8.":4O0<"#"94D<":2"=0:":2M"

"

P250/=6";0<<0K<@";05D"</?/K45"I<I4KKD"/6"EB":055/:25/4K"L4:05<";05D"10L"/6;0<:/=4:/26<"288I5"/6"2:.05":055/:25/4K"L4:05<M"

@"

-22J054:/26Q"

F20<"IJ"469"92L6M"N0J069<"26"-24<:"FI459"J05<26"/6"82??469"469"./<"J05<264K"2J/6/26<"469"50K4:/26<./J<M"

Page 81: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

)!""

R<<I0<"82?0"IJ"26"./=."J521/K0"488/906:<"<I8."4<"85I/<0"<./J<"L.050"H2:."L46:":2"K049M"

3GE@"#++#".0K9"IJ"J50::D"L0KKS":.0D"2H;/2I<KD"846T:"/6;0<:/=4:0"L.050":.0DT;0"H006"J45:"21":.0"84I<0"

R6":.0"3GE":.050"450"901/6/6="UI0<:/26<":2"90:05?/60"/1"-F"90105<":2"A>BC@"D0<S"/1"<82509"2;05"!++"J:<"

V="852L6"J5/680<<S"50=459K0<<"9/96T:":4O0"46D"K26=05":.46"625?4K"

@"

W50K/?/645DQ"X62:"504KKDY"X062I=.":2"K0:"I<"O62L"L0"L46:":2"=0:"/6;2K;09Y"ZI<:"H4</8<@"L.050[S"1/50S"</6O/6=S"0\JK2</26S"0:8]"

0=Q"G546=0"BI6@"25/=/64KKD"<2I6909"K/O0"<./J"J52HK0?"0/:.05"?08.46/84K"25"<./J"14/KI50"=2:"26"<8060"469"L4<".I?46"05525"

@"

G6"B8060Q"

G5=46/^4:/264K"?:=S"069"21":.0"94D"?:=S""

3464=0?06:"21":.0"26@<8060"/6;0<:/=4:/26@"12KK2L<"901/6/:/26"21"X/6:050<:09"J45:/0<Y"K4LD05"50K4:/26<S"/6;0<:/=4:/26<"450"125?4K"H2459"?/K/:45D"<:DK0"

B_G`"L.06"-F"K049<"/6;0<:/=4:/26"HD":.0/5"5IK0<"469"50=<"<K2L@"15I<:54:/6="/0"_49D"345D"

@"

799/:/264K";/</:<":2"<8060Q"X90J069<Y"

>DJ/84KKD"26"<8060"125":L2"L00O<"

R1"<I6O"469"6009<":2"H0"<4K;4=09Q"_4O0"F025=0"V:.46"7KK06"/<<I0"L/:."<84K/6=".49":2"=2"H48O"469"=0:"K4<05"JK2::/6="125"L25O4HK0"K/60<"469".IKK"125?"

@"

W52HK0?<Q"

BIHJ2064"4I:.25/:D"62:"/6:0564:/264K"

R6:05;/0L<"450"8261/906:/4K"?2<:"2:.05"JK480<"<I8."4<"Ea"L.050"4<"A>BC"/<"JIHK/8"125I?"846"8504:0"/<<I0<"<I8."4<"/1"Ea"546"/6;0<:/=4:/26"

@"

_4:0"*+<Q":.0"J5/68/JK0<"21"488/906:"/6;0<:/=4:/26"

Page 82: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

)#""

P4IK:":500S"3Gb>"509"J4=0":4="

c/?"B8.01105"L/KK":4KO"4H2I:"9/110506:"<8.22K<""

V;06:<"469"84<I4K"148:25<@"I<09":294D"

@"

>DJ/84K":54/6/6="469"L25O<.2J<Q"

32<:"50806:";05</26"21"WR33"?490"/[email protected]<0"469"/:T<"62:"9260"

>/?0K/60@"<:/8O/0<"26"4"L4KK"/0"A25L4D"

>2?"b2:.@b211D@"H2/K05"0\JK2</26"4<O"125"0;06:"469"84<I4K"<:I11".I=0":/?0K/60"

@"

N0J:.Q"

N01/609"HD"/1"846"H0"I<09"/6"J548:/84K"469"4JJK/84HK0"<410:D"5082??0694:/26""

7"K2:"21":./6=<"L/KK"ZI<:"=0:"62:09@"0="14:/=I0"dL/KK"8.08O"<K00J"8D8K0@"(#".5<"625?4K"IJ":2"#"L00O<eS"95I=<S"J50<85/J:/26"

325:@"<J480"<.I::K0"469">.500"3/K0"R<K469"HI:"2:.05L/<0":22"?I8."L25O"6"4":DJ/84K"A>BC"/6;0<:/=4:/26M"

@"

3IK:/64:/264KQ"O/694"

37RRP@"=4:.0509"./?<0K1""

R3G@"X62:":.4:"I<01IKY"

7CB@"c48O"BJ06805"

3Gb>@:/?0"826<I?/6="90:4/K09"

"

C/6905<Q"

HK48O@"/6:05;/0L/6="L/:60<<0<S"H4</8"K2=/8":22K<S"<25:"21"J508I5<25":2"1I:I50"H22O<"

J/6O@"2K9"8290"125"R3G"

L.:@"045KD"L25O"A>BC"045KD"J52:282K<S"4I:.25/:/0<"I<09":2"90;0K2J"?464=0?06:"26"<8060]"5010506809"045K/05"

Page 83: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

)$""

L.:@"I6/;05<4K"82I5<0@"J/J0K/60":54/6/6="

L.:"d4H2;0e@"2K9":./6=]"<2?0:./6="

<J/54K@"<04K"26"1526:@"L25O<.2J<":I56"21"806:I5D"=229"2;05;/0L"

L./:0"<J/54K@"L25O<.2J":2"06.4680"'"

L.:@"37RRP"#++#"I<09":2"50L5/:0"<2?0"A>BC"K/10"<4;/6="J52:282K<"

509@"522:"84I<0"464KD</<"7CB"

HKI0"<J/54K@"90J45:?06:"21"0605=D"

L.:"<J/54K@"82I5<0":2"4JJKD"3Gb>":I5609"/6:2"B37b>""

"

"

Page 84: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

)%""

BARRY STRAUCH !!,#,+*"

-./01"21"345/60"788/906:"/6;0<:/=4:/26">04?@C"

$"D5<"G3B"!&D5<"G7B"

#'"EB-F"

"

G;05<00<"/6;0<:/=4:/26<"21":04?@C"

"

R<".I?46"148:25<"<J08/4K/<:"125"G3B"

@"

-22J054:/26Q"

EB-F@"1/0K9"K0;0K"26@<8060"=229S"fg"XK0<<"=229"HI:"2OY"

A>BC,EB-F@"-2?JK/?06:45D"1I68:/26"HI:"62:"/906:/84K"

346D"5082??0694:/26<"=2":2"-F"469":.0D"926T:"4=500"L/:.":.0?"469":.0D"9/0"211"

@"

>546</:/26":2"G3B"dH/==0<:eQ"

I6905<:469/6=":.0"8IK:I50"469":.0"2;05<00/6="4=068/0<"/6"?45/60"L25K9"

@"

W50K/?/645DQ"

EB-F@"L.050S"1/50S"14:4K/:/0<S"<J08<S".2?0"<:4:0h"26KD"62:/1/0<"I<"/1"?00:<"A>BC"<:469459<"

@"

G6"B8060Q"

G5=46/^4:/264K"?00:/6=S"8K2<02I:"?00:/6=S"1/0K9"62:0<"!"RR-"

@"

799/:/264K";/</:<":2"<8060Q"

Page 85: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

)&""

"X?2<:"21":.0":/?0Y"

@"

V;/90680"=4:.05/6="d?4O0"<I50"4KK"/<"82KK08:09eQ"

b0<8I0<".4;0"./=.0<:"J5/25/:D@":.0D"?4D"90<:52D":.0"J.D</84K"0;/90680"

`/:60<<"<:4:0?06:<"?2<:"/?J25:46:":2"82KK08:"L.06"1/5<:"26"<8060"9I0":2":.0"J<D8.2K2=/84K"90=5494:/26"21"50K/4HK0":0<:/?26D""

"

A4;/=4:/26"L/<0"fI?46"V5525"25"308.46/84K"V5525"

X/6"?D"0\J05/0680"/1"D2I"926T:"=2"H48O"469"=4:.05"?250"0;/90680"D2I".4;06T:"9260"4"=229"Z2HY"

"

`0"I<I4KKD"?4:8.":.0"L/:60<<"<:4:0?06:<":2"J.D</84K"469"0K08:526/8"94:4"469"82?J450":2"2:.05"<:4:0?06:<":2"8261/5?""

"

XR"82IK9"=/;0"4"#%".2I5"82I5<0"26"/6:05;/0L/6="469"J<D8.2K2=/84K"01108:<Y"

@"

>22K<Q"

:.0DT50"4KK"I<0K0<<S"?490"HD"48490?/4"469"62:"/6;0<:/=4:25<h"62"?0:.29":.4:"1/:<"4KK"</:I4:/26<"D2I"L2IK9"069"IJ":5D/6=":2"12580":./6=<"/6:2":.0"0UI4:/26"

"

-05:4/6"UI0<:/26<"D2I".4;0":2"4<OQ"

I<I4KKD":.0"</?JK0<:"46<L05"/<"I<I4KKD":.0"5/=.:"260"

"

XL0"60;05"4880J:":.0"62:/26":.4:"H49":./6=<".4JJ06S"L0"4KL4D<"K22O"125"4"84I<0"469".4;0"46"/6;0<:/=4:/26".2L0;05"/:"?4D"H0"</?JK0Y"

V5525<"288I5":.06"L0"L25O"H48OL459"152?":.0"05525<":2":.0"<2?0:./6=":.4:"82IK9"84I<0":./<"469"L0"<:2JM"B2?0"21":.0<0":22K<"12580":./<"469"926T:"1/:M"

"

Page 86: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

)'""

-2<82"CI<26"

-4J:4/6"<:4D09"2I:"21"2J054:/26"21"<./J"L./K0"J/K2:"L4<"26"H2459h":.0"J/K2:".469K09"4KK"64;/=4:/26"50<J26</H/K/:/0<M"X82K9"1480Y"

"

XR"8.4KK06=0"D2I":2"1/69"46D"464KD:/84K":22K":.4:":4O0<"/6:2"4882I6:"8IK:I54K"148:25<S":.050"450"AGAVY":.0"/6:0548:/26"469"X82K9"1480Y"?4D".4;0"K049":2":.0"84J:4/6"9/<:468/6="./?<0K1"152?":.0"2J054:/26"21":.0";0<<0K"469":.I<"4KK2L/6=":.0"488/906:":2".4JJ06"

@"

`5/:/6="21"50J25:<@"X82IJK0"?26:.<Y":./<"90J069<"26":.0"82?JK0\/:D"21":.0"488/906:"i":.0"L25OK249"21":.0"/6;0<:/=4:25<"4:":.4:":/?0"d2:.05"/6;0<:/=4:/26<"I6905L4De"

@"

N0J:."d?4O0"<I50":.0"84I<0<"450"/906:/1/09"900J"062I=.S"J2:06:/4KKD"82IK9"=2"1250;05e"

b22:"84I<0@"L0"926T:"I<0"X522:"84I<0Y"/:T<"46"48490?/8":05?":.050"?/=.:"H0"?IK:/JK0"84I<0<"

N0J:.@"=229"UI0<:/26S"62"1/5?"46<L05h"</?JK0"K2=/8"=I/90<"UI0<:/26/6="L.050"92"D2I"<:2J"d:.0"-2??46946:"/<":22"145"50?2;09":2"H5/6=":.0"?4::05":2"./?"26"L.D":.0"J/K2:"L4<"<:/KK"4KK2L09"4"K/806<0e"

@"

-I55068DQ"

a00J"K/806<0<"8I5506:S"<069"J02JK0":2"<8.22K<,L25O<.2J<S"L25O"L/:."J45:/0<"469":.0D"826<IK:"

j/8:25@"L.4:"4H2I:"4"Z2I564K"8KIH"L.050"/6;0<:/=4:25<"=0:":2=0:.05"469"<.450"L.4:":.0DT;0"<006,5049"/6"50J25:<["

C5/:<@=229"50J25:<h"62:"=504:"I6/:"148:25<":.0D"<4D"J/K2:"05525"L0"<4D"L.4:":.0"148:25<"450"26"J/K2:":.4:"84I<09"./?":2"05525"

"

78490?/8"06;/526?06:"846"=/;0"4<</=6?06:<"504K"L25K9"4:"L25O"62:"J548:/84K"d62"Z2I564K"8KIHe"

@"

76D:./6=":2"499S"L.4:"L2IK9"D2I"8.46=0S"J45:":.4:"<.2IK9"H0"/?J52;09,"

Page 87: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

)(""

N/110506:"9541:<@"4<O"9/508:25<"L.4:":.0/5"52K0"/<"/6"9541:<S".2L"K26="920<"/:":4O0":2"50:I56"9541:<"L/:."82??06:<"/6"06:/50:D"62:"J45:/4KKDS"4<O"9/508:25"469"90JI:D"9/508:25Q"48:/;/:/0<"HI:"?4/6KD"9I54:/26S"4K<2"?464=0?06:"9/508:25"469"L.4:".0"0\J08:<"152?"9/508:25<"/6"50;/0L"J5280<<"=0:"/904"152?"./?"

"

C2459Q":.0"J2K/:/84K"H2459"/<"!,$"21":.0":/?0""

Page 88: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

))""

TOM ROTH-ROFFY !!,!+,+*"

345/60"V6=/6005@"788/906:"R6;0<:/=4:25"

"

A25L4D@"=52IJ"8.4/5?46"21"06=/6005"=52IJ"

!!"D5<"A>BC"G3B"

"

W50K/?/645D"7=068DQ"

EB-F@":.0DT50"/6"0;05D"J25:S"3/4?/"WN"469"2:.05<"H084I<0"21"0\JK2</26":.0D"L0506T:"<I50"21"85/?/64K"/?JK/84:/26<"/1"46D"

@"

7=068D"b0K4:/26<./JQ"

b0K4:/26<./J@">`7)++"X:DJ/84KKD"=229"./<:25/84KKD"L0T;0".49"<2?0":52IHK0<Y"L/:."fg"

PCR@"62:"I<I4KKD"H49@"3GE"

EB-F@"<2?0:/?0"J52HK0?4:/8S"J50;/2I<"G3B"N/508:25"12I=.:"L/:."-F"<2"<2?0":52IHK0"4:"fg"K0;0K<Q"B:4:06"/<K469"0\JK2</26")@!+"D5<["W5/?48D"/<<I0<"L.2<0"=2/6=":2"K049"469"L.2<0"=2/6=":2"Z2/6S"c48O".4<"?490"/:"H0::05"

"

W2<</HKD"/6"H09"L/:."-F":22"8K2<0"49;05<45/4K"XHI99D"HI99DY"=2"95/6O/6=":2=0:.05"

"

-2<82"CI<26@"-F"8.46=09":.0/5"50J25:":2"?4:8."2I5<]"<:546=0]["

@"

W50K/?/645D"7<<0<<?06:Q"

XR6"=06054K":.0D"=/;0"I<"L.4:":.0D".4;0Y"

@"

V;/90680"W5/25/:D"

Page 89: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

)*""

-261K/8:09"26"L.4:":2"92"1/5<:M"`/:60<<"0;/90680"/<"J05/<.4HK0"<.2IK9"=0:"1/5<:"4:":.0"<4?0":/?0"D2I"L46:":2"=0:"26":.0"<./J"469"14?/K/45/^0"D2I5<0K1"L/:.":.0"<8060"?4DH0"K22O"4:"<2?0"928I?06:<"90J069<"26"488/906:S"82?JK0\/:DS"J02JK0"/6;2K;09S"0:8]"=06054KKD":.0D":048."I<":2"82KK08:"J05/<.4HK0"0;/90680"1/5<:"

"

W2</:/26"0;/90680@0\JK2</26"25"?48./60"82?0<"4J45:"D2I"6009":2"928I?06:"25"J.2:2=54J.":./<"H01250"/:"=0:<"?2;09"

"

-2?JK0:0"K2<<Q"XL.2K0"/6":.0"L4:05YS"/6":./<"</:I4:/26":.0D"L2IK9"J52H4HKD"=2":2":.0"82?J46D"1/5<:"469":5D"469"50;/0L"<J08<"25"508259<"0:8]"

GH;/2I<KD"?250"8.4KK06=/6=".2L0;05S"90J069<"26"</:I4:/26Q";0<<0K"</^0"469"L4:05"90J:.S"?4D"H0"4HK0":2"54/<0";0<<0KS"9/;0S"25"bGj"

"

>.0"64;D"<IH"i"c4J460<0";0<<0K"211"f4L4//Q"54/<09";0<<0K"i"0\:06</;0"<0458."9I0":2"8IK:I54K"469"i"/6:0564:/264K"50K4:/26<"

@"

799/:/264K"j/</:<Q"

b0:I56":2"<8060"125"499/:/264K"0;/90680Q"J50;/2I<"9/508:25@AGS"HI9=0:"826<:54/6:<"

c48O@"kVBS"62L":.0D"<00?":2"826<:46:KD"=2/6="H48O"

3I<:".4;0"L45546:09"504<26"25"<2?0:./6="90;0K2J/6="/6"25905":2"=0:"4I:.25/^4:/26"

@"

N0J:.Q"

A2"504K":22KM"F0:"H48O".4;0"?00:/6=S":4KO"4H2I:"<410:D"/<<I0"<00"L.4:":.0D"L46:":2"JI5<I0"

>.52I=."82KK4H254:/;0"908/</26"469"L.4:":.0"908/90":2"12KK2L"

"

G3B@"62"J45:/8IK45"?0:.29"25":22KS";45/0<"HD"8/58I?<:4680<"21":.0"84<0"

"

W50;/2I<"9/508:25@":5/09":2"JI<."0;06:":/?0K/60"84I<4K":.025/0<S"14IK:":500"R".4;0"J05<264KKD"I<09S"3Gb>"L0"<.2IK9"I<0":.4:"?250"HI:"926T:"I<0"/6"?250"<:5I8:I54K"469"5/=/9"L4D"

Page 90: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

*+""

"

>.0"?250"0\J05/06809"=ID<"XL0"O62L"L.4:":2"K22O"125"469".2L":2"92"/:Y""

"

E<0":22K<"125":.0"60L"=ID<[@"X/6":.0"H0=/66/6=Y""

"

345=45/60"K/O09"H084I<0"?250"04</KD";/<I4KKD"9/<JK4D09"469":.0"826608:/26<M"

"

A25L4D@":22O"%"D5<"H084I<0"?26:.<"<:2JJ09"469"<:45:09"4=4/6"0:8]M"2;05"469"2;05"

@"

W52H4HK0"-4I<0Q"

W52H4HK0"84I<0@"82KK4H254:/;0"4=500?06:""

" " W52H4HK0"84I<0"\D^"

" " -26:5/HI:/6="84I<0"\D^"

"

-2<82"CI<26@"<J06:"4"K2:"21":/?0"90:05?/6/6="52K0<"21"J/K2:"i"?4<:05"

@"

3/<8M"W52HK0?<,"-2??06:<Q"

b7AN"50J25:@"8I55068D@?464=0?06:"8.46=0<S"F7GS":.0"J52HK0?<".4;0"H006"RNT9"62:"<I50"/1":.0DT;0"H006"49950<<09"

"

CI9=0:/6=@"0K08:526/8"J52JIK</26S"85/:/84K":./6O/6="L25O<.2J"

"

-22J054:/;0"K0456/6="/6"G3B"H0:L006"/6;0<:/=4:25<"82IK9".0KJ"

"

V5/8"P/0K9<@"90JI:D"21"<:54:0=/8"?464=0?06:S"L52:0"./?<0K1":.0"Z2HS"/:"60;05"=20<"46DL.050"

"

Page 91: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

*!""

b0;/</26"$++H"H2459"25905"

"

3464=0?06:""!"":04?"50K4:/26<"4L1IK"

"

3464=0?06:"6009<":2".4;0"4"H0::05"4JJ508/4:/26"125"26@<8060"/6;0<:/=4:/26"469"L.4:":.0D"=2":.52I=.M"

Page 92: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

*#""

LARRY BOWLING !!,!*,+*"

B5M"788/906:"R6;0<:/=4:25"

#"D5<"A>BC"

#!"D5<"EB-F"788/906:"R6;0<:/=4:25"

"

W50K/?/645D"7<<0<<?06:Q"

EB-F@"/6"J2:06:/4K"85/?/64K"<2?0:./6="2:.05h"1/50S"=52I69/6=@"A>BC"920<"H0::05"Z2H"21"928I?06:/6="

"

-2<82"CI<26@"-F"L4<"8K2<05":2"X:5I0"</:0"928I?06:4:/26"=2:"26":.0"<./JS":/?0"K01:S"0:8]"

"

7"K2:"21"/6;0<:/=4:25<":.050".2L0;05"26KD"/6:050<:09"/6"J2KKI:/26"4<<0<<?06:"469"50<J26<0"J48O4=0"

"

)@*@!+"L.2K0"HI68."21"J02JK0"

"

X`0"546"K049".2L0;05"-F"546":.0/5"2L6"J454KK0K"/6;0<:/=4:/26Y"

@"

b0K4:/26<./JQ"

XBD?H/2:/8"260"D2I"6009"H2:."4=068/0<"/:"/<"4"<D<:0?"21"8.08O<"469"H4K4680<Y"

"

Xb0K4:/26<./J<"=2::06"4"K2:"H0::05Y"

X846":.0D"H0"H0::05"805:4/6KDS"26KD"H2:."</90<Y"HI:"125":.0"?2<:"J45:":.0DT50"4"K2:"H0::05":.46":.0D"L050Y"Cl"

@""

G6"<8060Q"

Page 93: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

*$""

XR".4;0"62:"546"K049"RR-"26"A>BCY"

RR-"25=46/^4:/264K"?:=S"=0:"/6"=52IJ<"

>./6=<".0"846,.4<"9260Q"

V6=/6005/6=@<?4KK"

N08O@<?4KK,K45=0"

BI5;/;4K@4KK"

fI?46"148:25<@"926T:"K/O0":2"=0:"/6;2K;09"

@"

V;/90680"W5/25/:DQ"

XR:T<"4"H/="90J069<Y"X90J069/6="26":.0"<80645/2Y"

N28I?06:45D@K/806<0S"0:8]"

V;/90680"=4:.0509"HD"-F"d8.08OK/<:e"

"

>.4:"K04;0<"J.D</84K"469"L/:60<<"0;/90680@"90J069<"26":/?0"4KK284:09"

@"

799/:/264K"j/</:<Q"

A>BC@X?2<:"84<0<"R".4;0":2"=2"H48O":2":.0"48:I4K"</:0"4:"K04<:"2680Y"

"

j>B@";0<<0K":5411/8"<05;/80"d7>-"125"<./J<".2L0;05"I6K/O0"7>-"62:"?4694:25D"125"4KK";0<<0K<e"

@"

N0J:.Q"

`.4:".4;0"L0"4882?JK/<.09"L.4:"92"L0"L46:":2"4882?JK/<.@"94/KD"L/:./6":.0"=52IJ"

"

`0":.06":0KK"RR-"L0".4;0"0;05D:./6="L0"6009"

"

Page 94: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

*%""

-2<:Q"B46"P5468/<82@"46D:./6=":.4:"82IK9"H0"9260"2680"L0"=2:"H48O"L0"9/9"9I0":2"82<:<Q"m&+"845"J05"94D"62"/6,2I:"m%&+".2:0K"J05"94D"

"

R3G@"=I/94680"8.08OK/<:S"0\J05/0680S"<:04K"469"J/K105"152?"-FS":.050"/<"62"</:I4:/264K"8.08OK/<:"125"048."<80645/2"

@"

W52H4HK0"-4I<0Q"

XR"I<0"R3G"901/6/:/26Y"/:T<"4"H/:"21"4"<250"<J2:"R".4;0"4<O09"125":./<"211/80T<"901/6/:/26"21"J52H4HK0"84I<0"

"

NAj@8K4<<"<28/0:D@"B-7>"<D<:0?"

-F@"I<09":2".4;0"260"84KK09"522:"84I<0"

"

Xf050"D2I"52I69":4HK0"/:S"<2?0:/?0<"/:"L25O<"<2?0:/?0<"/:"920<6T:Y"

"

f2L"K26="920<"B-7>":4O0["

XN0J069<"26":.0"<82J0"21":.0"/68/906:Y"

7K4<O46"b46=05@"&"94D<Q"/<2K4:09"2I:"J/K2:S"J52JIK</26S"0:8"469":.0D"L050"4KK"/690J06906:KD"464KD^09"""""""""""""""""""""""""!!"

"

XN20<6T:".4;0":2"H0"260"J52H4HK0"84I<0"846"H0"?IK:/JK0"L/:."4":26"21"826:5/HI:/6="84I<0<Y"

"

X>./<"211/80".4<"H2552L09"152?Y"2:.05"<D<:0?<"

@"

-I55068DQ"

A0:L25O/6=@"[email protected]@2:.05"82I6:5/0<"

`0H</:0<S"826;06:/26<S"/6:0548:/26"L/:."2:.05"?45/605<"

Page 95: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

*&""

XA2"62:"504KKD"L0"=2:"J50::D"=229":54/6/6=".050Y"

@"

3/<8M"

`4D<":2"/?J52;0":.0"J5280<<Q"

>2J"$"125":/?0Q"

B:504?K/60"50J25:"J5280<<"

3/6/?/^0"L/692L"125"09/:/6="8D8K/6="

G3B"901/6/:/26"21"XW52H4HK0"-4I<0Y"

WR33Q"X=229"?290K"469"R":./6O"/:TKK"L25OY"

@"

B21:L450Q"

f0KJ1IK".2L0;05S"<21:L450@"m$++,K/806<0"

Page 96: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

*'""

OFFICE OF AVIATION SAFETY JOHN DELISI N0JI:D"N/508:25"G11/80"21"7;/4:/26"B410:D"

d#+#e"%!%@"'$%%"

b?"&%+%"

!!,&,+*"

"

P48:I4K"50J25:"152?"/6;0<:/=4:/;0"=52IJ<Q"

>.06"2K9"?290K"<4D<"D2I"L2IK9"464KD^0"

%"J/K2:<"6260":DJ0L5/::06"/6"(+("

N0:05?/64:/26"21"J52H4HK0"84I<0"

A0L"?290K@"X826<:46:KD"464KD^/6="148:<Y"

@"

V;/90680"W5/25/:DQ"

7jNbQ"4I9/2,;/902"94:4"50825905S"82KK08:09"

7>-"82??I6/84:/26"

j05D"045KD"90;0K2J"4":/?0K/60"

@"

N0J:.Q"

WR33Q"

PI0K":5I8O"62"K26=05"4"WR"

B:526=05".049L/69<""

A2"908K454:/26"21"46"0?05=068D"

"

WRn7>-"i"K46=I4=0"H0:L006"J/K2:"469"cPaQ"<4/9"X0\J09/:09Y"50:I56"/6<:049"21"0?05=068D"

Page 97: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

*(""

"

799"WRT<"4<"L0"=2"4K26=S"/6;0<:/=4:/26"/<"60;05"8K2<09":.0D"846"=2"H48O"469"502J06"/1"60L"8/58I?<:4680<"25"0;/90680<"/<"12I69"

"

P77@"50=IK4:/6="4=068D"/6"0;05D"4<J08:S"HD"K4L"4"J45:D"/6"0;05D"4;/4:/26"/6;0<:/=4:/26"

"

PCR"L/KK"=2":2"488/906:<"4<"K26="4<"6080<<45D"I6:/K"/:"/<"90:05?/609":.4:":.050"/<"62"85/?/64K"48:/;/:D":2"84I<0":.0"488/906:"

"

A7B7"148/K/:D"/6"fI6:<;/KK0"7_"

PCR"/6;2K;0?06:@"920<":.4:"<K2L"D2I"92L6["X<K/=.:KDY"

"

V=DJ:"4/5"!***"

@"

V;/90680"W5/25/:DQ"

G6@<8060"4H2I:"260"L00O"928I?06:/6="4<"/:"K/0<"469"O0D":./6=<"

N28I?06:/6="J05/<.4HK0"0;/90680Q"K2=<S"0:8]"

"

o!"1K/=.:"94:4"50825905"i"1K/=.:"4I9/2"50825905"

@"

799/:/264K"j/</:<Q"

`508O4=0"<:2509"/6".46=45"469":.0D"?4D"=2"H48O"$@%]":/?0<"2;05"60\:"82IJK0"?26:.<"

"

`/69"<.005"L456/6="<D<:0?":.4:"9/9"62:"=2"211S"L0"600909"8./J"152?"82?JI:05"<D<:0?"60;05"9/9"1/69"/:"

`/:60<<"/6:05;/0L<"450":DJ/84KKD":.0"K04<:";4KI4HK0"0;/90680":.0D"=0:"I6K0<<"152?":.0"850L"/6"84<0<"L.050"4/58541:"/<"62:"4":2:4K"K2<<"50<IK:/6="/6":2:4K"14:4K/:/0<M"

Page 98: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

*)""

@"

-4I<0"464KD</<Q"

-26;05:"PNb"J454?0:05"94:4"/6:2"46/?4:/26"469"<00"L/:."1K/=.:"</?IK4:25h""

8K2<09Q"ZI<:"94:4"50<IK:"

2J06Q"L/:."J/K2:"825508:/26"25"82?J06<4:/;0"50<J26<0<""

"

j2/80"50825905@62:"26KD";2/80<"HI:"2:.05"<2I69<"<I8."4<"06=/60"bW3B":.52I=."<2I69"<J08:5I?"<21:L450":.0"82?JI:05"846"90:05?/60"bW3B""

@"

3/<8M"

77"&)(@"cPa"7$++"1/6/:0"0K0?06:"464KD</<"

"

W3R" " " " " " WGR"

W5/68/JK0"?4/6:064680"/6<J08:25" " W5/68/JK0"2J054:/26<"/6<J08:25"

"

N0J:.Q"X2I5"8.45:05"/<"90:05?/60":.0"J52H4HK0"84I<0Y"

#"6/=.:<"045K/05"9I5/6="?4/6:064680"8.08O""

"

A2:":.0"48:I4K"84I<0Q"0K0;4:25"<:I8O"/6"J2</:/26"

>.0D"450"K22O/6="125":.4:"8.4/6"21"0;06:<""

"

b0K04<0"5082??0694:/26<"9I5/6="/6;0<:/=4:/26"

"

R69I<:5D"**p"<0K1"50=IK4:09"HD":.0"/69/;/9I4K"4/5K/60<"L4:8./6=":.0?<0K;0<"

"

W/K2:<"450":.0"K4<:"K/60"21"90106<0S"62"<:4:/<:/8<"26".2L"?46D"488/906:<"450"J50;06:09"HD"850L,J/K2:<"

Page 99: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

**""

@"

BI==0<:/26<Q"

R?J52;0?06:"125"G3BQ"=0:"?250"82?125:4HK0"L/:."/906:/1D/6=":.0"J52H4HK0"84I<0"

"

B64="dG7BeQ"=0::/6="/6;0<:/=4:25<":2"464KD^0"148:<S"J52HK0?<"826;0D/6="94:4"/6:2"I6905<:4694HK0"</:I4:/264K"<80645/2"

BR3W_V"764KD</<"i"76<L05"

"

B:411"9541:@"<:411"50;/0L<"

7B@!"q"c2.6"N0_/</"50;/0L<"d!"L00Oe"

F24KQ"'"?4Z25"!+"H5/01"50J25:<"4"D045"

"

>04?<Q"54692?"K2::05D"90J069/6="26"52:4:/26"L.06"488/906:<"288I5"

)@*"<J08/4K"148:25<"

Page 100: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!++""

OFFICE OF HIGHWAY SAFETY MARK BAGNARD !!,#$,+*"

GfB"

N/;/</26"8./01"21"/6;0<:/=4:/26"9/;/</26""

%"D5<"

!#"D5<"./=.L4D"/6;0<:/=4:25"

"

W50K/?/645D"7<<0<<?06:Q"

>DJ/84KKD"/:"L/KK"H0"L.4:0;05"K284K"K4L"0612580?06:".4<"ZI5/<9/8:/26"

@"

b0K4:/26<./JQ"

/:"/<"=06054KKD";05D"=229M"`0T50"125:I64:0"/6"GfB":.4:"?46D"J02JK0".4;0"4"K4L"0612580?06:"H48O=52I69M"

@"

V;/90680"W5/25/:DQ"

`0KK"/:"90J069<"26"/1":.0D".4;0"46"/6:06:":2"1/K0"85/?/64K"8.45=0<M"R1"/:T<"4"84<0"L.050":.0D"L/KK":4O0"/:"92L6":.0"5249":2"J52<08I:/26":.0D"L/KK"J52;/90"0\:54259/645D"<IJJ25:"125"I<M"

@"

A2:/1/84:/26"21"488/906:Q"

34Z25/:D"21":.0?"450":.52I=.":.0"82??I6/84:/26"806:05"2884</264KKD"L.06"/:T<"/6"46"4504"L.050"L0T;0"L25O09"H01250":.0D"?4D"826:48:"I<"9/508:KD"HI:"/:T<"I<I4KKD":.52I=.":.0"82??I6/84:/26"806:05M"

@"

G6"<8060Q"

`0".4;0":L2"/6;0<:/=4:/;0":04?<":.4:"L0"52:4:0"L.2"450"26"84KKM"V48.":04?".4<"46"RR-"469"1/;0"=52IJ"8.4/5?46Q"./=.L4DS"?2:25"8455/05S"<I5;/;4KS".I?46"J05125?4680S"469";0./8K0M"`.06"L0"=0:"26"<8060":.2<0"1/;0"/69/;/9I4K<"L/KK"=2"2I:"469"K22O"4:":.0/5"<J08/1/8"<08:/26"21"

Page 101: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!+!""

:.0"/6;0<:/=4:/26M"`.4:"L0"92":2"=0:":.0"H/="J/8:I50"/<"4:":.0"069"21":.0"94D"L0TKK".4;0"4"?00:/6="469"0;05D260"L/KK"9/<8I<<":.0/5"1/69<"469"0;05D260"L0/=.<"/6"26"L.4:".4JJ0609"/6":.0"488/906:M"

@"

799/:/264K";/</:<Q"

R:"L2IK9"H0";05D"K2LM"`0"26KD"=2"H48O"L.06"L0"1/69"<2?0:./6="60L"25"L0".4;0"46"/<<I0"L/:.":.0"J2K/80"469":.0/5"25/=/64K"928I?06:4:/26M"

@"

N0J:.Q"

`0".4;0".49"260"/6":.0"J4<:"HI:"R"82IK96T:":0KK"D2I":.0"K4<:":/?0"/:"L4<"I<09M"`0T50"8I5506:KD"/6":.0"J5280<<"21"50;/</6=":.4:M"`0"92".4;0"4"<0:"21"J52:282K"125"048."=52IJ"8.4/5?46"4<"145"4<"L.4:"/<":.0"?/6/?I?":2"H0"0\4?/609M"`.4:"L0"926T:T"L46:":2"92"/<"=0:"/6":.0".4H/:"21"I</6="4"8.08OK/<:"L.050"D2I"?4D"2;05K22O"<2?0:./6="HD"12KK2L/6=":.0"8.08OK/<:":22"8K2<0KDM"

@"

W52H4HK0"84I<0Q"

A2:":.4:"RT?"4L450"21M"`0"K22O"125":.0"J52H4HK0"84I<0"H4<09"26"L.4:"L0".4;0"12I69"L0"K22O"125":.0"L.D"469":.0".2L":.0"488/906:"288I5509M"

@"

"

30:.29<,464KD:/84K":22KQ"

`0"92"?250"L/:.":.0"L25O"JK466/6="?00:/6="469"L0TKK"</:"92L6"L/:.":.0"J52Z08:"?464=05"21":.4:"84<0"469"K0:"048."=52IJ"8.4/5?46"=2"2;05":.0/5"1/69/6=<"469"I</6=":./<"L0TKK"<:45:"901/6/6="4"J52H4HK0"84I<0M"

@"

764KD</<Q"

`0".4;0":L2"</90<":2":.0"211/80"fB@#+"L./8."/<":.0"/6;0<:/=4:/26"</90"469":.06"fB@$+"L./8."/<":.0"L5/:/6="</90M"

"

>/?0"464KD</<":/?0Q"L25O"JK466/6="?00:/6="L00O"'"?4Z25/:D"21"464KD</<"/<"82?JK0:09S"?/=.:"62:"H0"/6"4"1/64K"50J25:"HI:"?2<:KD"82KK08:09M"

Page 102: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!+#""

G60"D045"HI:"R":./6O"/:T<":08.6/84KKD!&"?26:.<"

@"

b09I69468DQ"

>04?"9541:"50;/0L@211/80"9541:"50;/0L@"?464=0?06:"211/80"50;/0L"

V;05D"<:0J"/<"6080<<45D"4:"K04<:"L/:.":.0"L4D"L0"450"92/6="/:"L/:.":.0"<:411":2".4;0":.0"1/5<:"8548O"4:"/:"469":.06"/:"=20<":2"?464=0?06:M"R":./6O"048."<:0J"/<"/?J25:46:M"

@"

-I55068DQ"

34/6KD":.4:".4<":2"92"L/:.":.0"466I4K":54/6/6="J52=54?M"`0"L/KK"52I:/60KD"<069"/6;0<:/=4:25<":2":.0"?46I148:I505<":2"<00"L.4:".4JJ06<"/6":.0"J5280<<M"

@"

B21:L450Q"

`0"I<0"4"K2:"21"-7N"<21:L450M"`.06"L0"928I?06:"4"<8060"L0"I<0"4":2:4K"<:4:/26":2"826<:5I8:"4"$N";/0L"21":.0"48:I4K"488/906:"<8060"/6"504K":/?0M"

@"

W50K/?/645D"7<<0<<?06:Q"

)&@*+p"/<"J52;/909"HD":.0"J2K/80"25":.0"1/5<:"50<J26905M"f2L0;05S"L0"ZI<:"6009":2"L4:8.":.0/5"K0;0K"21"/6:050<:"/6":.0"84<0M"R1":.0D"JK46"D2I"1/K0"85/?/64K"48:/26<":.0/5"/6;2K;0?06:"L/KK"H0":.4:"?I8."?250":.46"/1"/:T<"ZI<:"46"488/906:"L.050":.0D"=4:.05":.0"90:4/K<"HI:"=2"62"145:.05M"

Page 103: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!+$""

TRANSPORTATION SAFETY BOARD (CANADA) KEN POTTER 3464=05"21"R6;0<:/=4:/26"GJ054:/26<@f049"G11/80"

%"D5<">BC"

!#D5<"B410:D"764KD<:S"345/60"06=/6005/6="<J08/4K/<:"

!!,!(,+*"

"

-I5506:"W2</:/26Q"

3464=05"21"2J054:/26<@"RT?"50<J26</HK0"125"<IJJ25:/6="/6;0<:/=4:/26<"5/=.:"4852<<":.0"82I6:5D"L/:."<J08/4K/<:<"<I8."4<"64;4K"458./:08:<"06=/6005<"469"K4H254:25D"06=/6005/6="<05;/80<"469"<I8."4<"L0KK"RT?"50<J26</HK0"K4:05"26"/6":.0"J5280<<"125"H2:.":.0"<410:D"464KD</<":.0"464KD</<"21":.0"148:<":.4:"450"=4:.0509"469":.0"<410:D"82??I6/84:/26":.4:"/<"49;/<25D"K0::05<"25"L5/:/6="21"5082??0694:/26<"469".4;/6=":.0?"J4<<":.52I=.":.0"H2459".050"125"4JJ52;4KM"G6":2J"21":.4:":4O/6="211":.4:".4:"469"JI::/6="26"?D":./59"RT?"4K<2":.0"?464=05"21":.0"806:54K"469"458:/8"50=/26"<2"4KK"/6;0<:/=4:/26<"/6":.0"806:54K"J45:"469"21":.0"82I6:5D"469"4H2I:"'+90=50<<"625:."14KK"I6905"?D"50<J26</H/K/:D"9/508:KDM"

@"

W50K/?/645D"7<<0<<?06:Q"

`0":5D":2"=0:":.050"1/5<:"HI:"21"82I5<0"L0"60;05"H04:":.0"1/5<:"50<J26905<"<2":DJ/84KKD":.050"L/KK"H0"J2K/80S"1/50"1/=.:05<S"25"-24<:"FI459S"25"B0458."469"b0<8I0"26":.0"<8060"1/5<:"H01250"I<M"

@"

b0K4:/26<./JQ"

`0"=06054KKD".4;0"4";05D"=229"L25O/6="50K4:/26<./J"L/:.":.0"1/5<:"50<J269/6="4=068/0<"/6"J45:/8IK45":.2<0"L0"L25O"L/:."H01250"b2D4K"-4649/46"?2I6:09"J2K/80"8="90J45:?06:"21"64:/264K"90106<0"L0".4;0"3GET<"L/:."<2"J50::D"?I8."H0:L006"I<"469":.0?"0;05D260"O62L<"L.2<0"92/6="L.4:":2"L.2?M"

"

>.0">BC"J52;/90<"/6125?4:/26":2":.2<0"1/5<:"50<J26905<"50=459/6="L.4:"L0"0\J08:M"P25"0\4?JK0S"2H;/2I<KD":.0D"450"50<J26</HK0"125"<4;/6="K/10"/1":.0D"846S"=06054KKD"L0"K0:":.0?"

Page 104: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!+%""

O62L":.4:"L0T9"K/O0":.0?":2"K22O"4:":.0"</:0":5D"62:":2"9/<:I5H"/:":22"?I8."/1"J2<</HK0"469"L0TKK"21:06"/6:05;/0L":.0?"41:05":.0"148:":2"<4D"L.4:"9/9"D2I"<00"L.06"D2I"=2:":.050["

@"

V;/90680"W5/25/:DQ"

G680"2I5"J02JK0"90JK2D"469"=0:":2":.0"</:0":.0D"2H;/2I<KD"/9":.0?<0K;0<":2":.0"211/805<"25":.0"1/5<:"50<J26905<"25":.0"211/805<"26":.0"<./J"/6":./<"84<0M">.0DTKK":4O0"4"K22O"4:":.0"</:0"/:<0K1"/6":./<"84<0"/:"L4<"46"06=/60"14/KI50M">.0":04?"L/KK"92"4"H5/01"J4<<":.52I=."K22O"469":.0"06=/60"522?"K22O"4:":.0"06;/526?06:":.0DT50"/6M":.0D"4K<2"K22O";05D"84501IKKD"4:"<410:D"L.06":.0D"H2459"4"<./J"260"21":.0"1/5<:"826</9054:/26<"/<":./<"</:0"<410"125"?0":2"H0":.050M"gI/:0"21:06"L/:."1250/=6";0<<0K<":.050"450":./6=<":4O06"125"=546:09"26"H2459":.4:"L0"<25:4"<.4O0"2I5".049<"4:"<2"L0T50";05D"82=6/:/;0"21"288IJ4:/264K"<410:D"469".04K:."/<<I0<M"G680"26"H2459"90J069/6="26":.0"</^0"21":.0";0<<0K"/1"/:".4<"4"jNb"L0"/??09/4:0"92L6K249":.4:"4<"UI/8OKD"4<"J2<</HK0M"P2KK2L/6=":.4:"L0TKK"=0:"46"/904"26"L.2<0"/6;2K;09"469"908/90"L.2"L0"L46:":2"/6:05;/0L"469"<0:"IJ"46"/6:05;/0L"JK46M">./<"/<"4KK"21"82I5<0"=2/6="26"L./K0":.0"<./J"/<":5D/6=":2"H0"<4K;4=09"25"J02JK0"450":5D/6=":2"92":.0/5"2L6":./6=<M"`0":5D":2"4882??294:0":.0?"L./K0":./<"/<"=2/6="26"4<"?I8."4<"J2<</HK0M"7<"6080<<45D":.0":04?"/6":.0"1/0K9"L/KK"H5/01":.0"<:411"4:".049"211/80"25"/1"6080<<45D"50UI0<:"4<</<:4680"152?"<J08/4K/<:<M"B2"/1":.0D"6009"4"64;4K"458./:08:"L0"L/KK"9/<J4:8."260M"4:":.0"<4?0":/?0":.0":04?"26"H2459"/<"<82J/6=":.0"/6;0<:/=4:/26"<4D/6=".2L"145"450"L0"=2/6="L/:.":./<"L.4:"450":.0"/6/:/4K"/<<I0<"L0"450"<00/6="82??I6/84:/6=":.2<0"/<<I0<"H48O":2".049"211/80"L.050":.0"464KD<:<".050"L/KK"<:45:"92/6="H48O=52I69"50<0458."<:4:/<:/84K"50<0458."J50;/2I<"288I550680":.4:"<25:"21":./6=M"G6"</:0"/6;0<:/=4:/26<".050"846"K4<:"46DL.050"152?"260"94D":2":L2"?26:.<"90J069/6="26":.0"<0;05/:D"469"<82J0"21":.0"288I550680M"E<I4KKD":.0"4;054=0"RT9"<4D"/<":L2@12I5"94D<"41:05":.4:"K06=:."21":/?0"D2IT;0"/6:05;/0L09"4KK"21"D2I5"L/:60<<0<"D2IT;0"82KK08:09"D2I5".459"0;/90680"D2I5"J.D</84K"0;/90680"125"0\4?JK0"/1"D2I"6009"J45:<"464KD^09"<:00K"<4?JK0<":.4:"<25:"21":./6="D2IT;0"82KK08:09":.0?"469"450"H5/6=/6=":.0?"H48O":2".049"211/80M"

"

F06054KKD"/:"L2IK9"H0S"k0<"<08I50":.0"J.D</84K"0;/90680"469":.06"=2"/6:2":.0"/6:05;/0L"J.4<0"2680"D2IT;0"<006"452I69":.0"<8060"L.4:".4<"=260"26M"k2I"846T:"ZI<:"K04J"/6:2"/6:05;/0L<"5/=.:"4L4DM"

@"

799/:/264K"j/</:<Q"

A2:";05D"21:06"L0T50"I<I4KKD"J50::D":.252I=."4H2I:"/:M"`0TKK"2884</264KKD".4;0":2"=2"H48O"469":4KO":2"L/:60<<0<"HI:"62:"4:":.0"</:0"/:<0K1M"25"?4DH0"2884</264KKD"469"RTKK"<4D"?4DH0"2680":L/80"4"D045"L0TKK".4;0":2"=2"H48O"469"K22O"4:"<2?0"J.D</84K"0;/90680"4=4/6"HI:"=06054KKD"L0T50"J50::D":.252I=."26"2I5"1/5<:"J4<<":.52I=.M"

@"

Page 105: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!+&""

N0J:.Q"

`0".4;0"4"<05/0<"21"?46I4K<"84KK09"?46I4K<"21"/6;0<:/=4:/26"469":.0D"J52;/90":.0"?2/";2KI?0"#"/<"<J08/1/8":2"1/0K9"/6;0<:/=4:/26"469"/:"J52;/90<"=I/94680"125":.0"82KK08:/26"21"0;/90680".2L":2":504:":.4:"0;/90680"J5280<<"/:"L.4:":2"K22O"125":.4:"<25:"21":./6=M"7<"L0KK"L0".4;0"8.08OK/<:<"125";45/2I<"/6:05;/0L<"469"125";45/2I<":DJ0<"21"488/906:<"K/O0"/1"/:T<"4":46O05"488/906:"L0TKK".4;0"4"8.08OK/<:"125":46O05"488/906:S"1/<./6=";0<<0K"9/110506:"8.08OK/<:"125":.4:M"CI:":.2<0"450"I<09"HD"<2?0"/6;0<:/=4:25<"62:"4KKM"F06054KKD":.0"?2/"J52;/90<":.0"=I/94680"6080<<45D"125":.0"8269I8:"21":.0/5"/6;0<:/=4:/26M"R"L/KK"<4D":.4:"=06054KKD"2680"46"/6;0<:/=4:25".4<"I<09"260".0"L26T:"2J06"/:"/6":.0"1/0K9".0"L/KK"=06054KKD"O62L"L.4:":2"K22O"125M"

@"

V;/90680"-.08OK/<:Q"

R":./6O"<2?0:/?0<"46"/6;0<:/=4:25"L/KK"H0"82?0";05D"J0946:/8"L/:.":.0"8.08OK/<:"469"48:I4KKD"?/<<"<2?0:./6=".0"?4D".4;0"J/8O09"IJ"26M"CI:"<2?0L.050":.050T<"4"H4K4680"H0:L006"/6:I/:/26"469":.0"8.08OK/<:"469"0\J05/0680"21"82I5<0"4<"L0KKM"R"<.2IK9"J2/6:"2I:":22":.4:"2I5"/6;0<:/=4:25<"/6":.0"1/0K9"4506T:"48:/6="/6"/<2K4:/26M">.0DTKK"H0"84KK"H48O"469"H5/01/6="I<".050"4:".049"211/80"469"2I5"J02JK0".050"/68KI9/6="?D<0K1"469":.0"2:.05"?464=05<"L/KK"H0"8.4KK06=/6="L.4:":.0DT;0"9260"469"<I==0<:/6="L.050":.0D"?4D"L/<.":2"K22OM"

@"

W52H4HK0"-4I<0Q"

`0"I<0"0\J50<</26"84KK09"X84I<0<"469"826:5/HI:/6="148:25<Y"L0".4;0"4"125?4K"J52809I50"125"464KD^/6="46"288I550680":.0"0;06:<"469":.0"148:<"21"46"288I550680"L./8."K049<"I<":2":.4:"84I<0"469"826:5/HI:/6="148:25"25":.0"J52H4HK0"84I<0"4<"D2I"90<85/H09"/:M"`0"84KK"/:":.0"RBR3"d/6:0=54:09"<410:D"/6;0<:/=4:/26"?0:.292K2=De"469"0<<06:/4KKD"L.4:"L0"92"/<":4O0"4"<410:D"</=6/1/846:"0;06:"469"L0"95/KK"92L6"152?"/:S"L0TKK"ZI<:"4<O"4"UI0<:/26"XL.D"9/9":.4:".4JJ06[Y"469":.06"L0"46<L05"/:"XH084I<0":./<]Y"469":.06"L0TKK"826:/6I0"4"L.D@H084I<0"464KD</<"I6:/K"L0"82?0":2":.4:"I6905KD/6="148:25":.4:"D2I"846T:"95/KK"92L6"46D?250M"B2"/6"2:.05"L259<":2"82?0":2":.4:"J52H4HK0"84I<0"L0T50"4JJKD/6="4"<8/06:/1/8"?0:.292K2=D":2":.0"488/906:"469":.0"148:<M"

@"

764KD</<"d_06=:.eQ"

`0"<:45:"/:"/??09/4:0KD"469"L0"0682I54=0"2I5"/6;0<:/=4:25<"/6":.0"1/0K9S"L.06":.0DT50"92/6="/6:05;/0L<"25"82KK08:/6="0;/90680"/6":.0"1/0K9":2"<:45:"K22O/6="4:"/6":.0/5"?/69"4"L.D@H084I<0M">DJ/84KKD"L0TKK"<:45:"46"0;06:<"9/4=54?"469"<:45:":.0"464KD</<"21":.2<0"0;06:<":.0"L.D@H084I<0"464KD</<"21":.2<0"0;06:<"L/:./6":L2"25":.500"94D<"21":.0"288I550680M"G680":.0"0;06:<"25":.0"/6125?4:/26"82?0<"/6":.0DTKK"<:45:"/:"/6":.0".2:0K"4:"6/=.:"41:05"H0/6="/6":.0"1/0K9"

Page 106: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!+'""

9I5/6=":.0"94DM"769":.0"464KD<:"H48O"4:".049"211/80"L/KK"<:45:"/:"4<"L0KK"/6"<IJJ25:"21":.0"1/0K9":04?M"B2"4K?2<:"/??09/4:0KD"469":.0"504<26"L0"92":.4:"/<"H084I<0"/1"D2I".4;0":2"=2"H48O"K4:05"/1"D2I".4;0":2":548O"J02JK0"92L6"/:"846"H0";05D"9/11/8IK:"469":.0/5"?0?25/0<"1490"25":.0D"?/=.:"62:".4;0":.0"/6125?4:/26"D2I"6009M"

@"

-I55068DQ"

`0".4;0"46"466I4K":54/6/6="JK46"JI:":2=0:.05"125":.0"?45/60"H5468.".050M"GH;/2I<KD"/:T<"K/?/:09"HD":.0"HI9=0:"826</9054:/26<"L0".4;0":2L459<":54/6/6=M"CI:"L.4:"L0":5D":2"92"/<"/906:/1D"0?05=/6="4504<"21":08.62K2=DS"<410:D":54/6/6=S"469"50150<.05":54/6/6="469"4<</=6":.4:":54/6/6=":2":.0"/6;0<:/=4:25":.4:"L/KK"?2<:"H0601/:"152?"/:M"

"

>.4:"<00?<":2"H0"4"826<:46:":.0?0".050M"B069"?0"2I:"26"4"<./J"<069"?0"2I:"26"4":5/J"152?"7":2"C"<2"R"846"2H<05;0M"769"L0T;0"H006"92/6=":.4:M"P25"0\4?JK0"125":.0"K4<:"</\"?26:.<"L0T;0"H006"<069/6="<:411"26"_/UI/9"A4:I54K"F4<"-455/05<"d_AF-e"152?"-46494":2">5/6/949"ZI<:":2"2H<05;0M"

@"

N0J:.Q"L.4:"/<":.0"I<I4K":I56452I69["

GI5"2L6"<:469459"469":./<"/<"/6:0564K".050"/<":2":5D":2".4;0"46"/6;0<:/=4:/26"152?"<:45:":2"1/6/<."469"82?JK0:09"469":.0"50J25:"2I:":.0"9225":2":.0"JIHK/8"/6"260"D045"25"K0<<M"

@"

b09I69468DQ"

>DJ/84KKD"/:"846"=2"H48O"H0:L006"?464=0?06:S"?D<0K1S"25":.0":04?"H084I<0"L0"L25O"/6":04?<"%S"&S"'":/?0<"469":4O0"</\"L00O<M""

@"

B21:L450Q"

7H<2KI:0KDM"`0"92L6K249"jNb<M"37N7B"qA>BC",37RCM"7<"145"4<"508259<"O00J/6="469"<410:D"464KD</<"125"0\"2I5"RBR3"J5280<<":.050T<"46"464KD</<":22K"L/:./6"2I5"82?JI:05"<D<:0?".050":.4:"4KK2L<"I<":2"92"0;06:"9/4=54?<S":2"92":.0"95/KK"92L6S":.0"L.D@H084I<0S":2"92":.0"5/<O"464KD</<"4KK"21":.4:"J5280<<"/<"9260"0K08:526/84KKDM"_4<:KD"L0".4;0"4"<D<:0?"/:T<"0<<06:/4KKD"4"508259<"?464=0?06:"<D<:0?<"84KK09">R3B":546<J25:4:/26"/6;0<:/=4:/26"?464=0?06:"<D<:0?":.4:"4KK2L<"I<":2"?464=0"1/K0<"0K08:526/84KKDM"`0T;0"?2;09"/6":.4:"9/508:/26":.4:"4KK"508259<"125"46"/6;0<:/=4:/26"450"0K08:526/8"469"0;05D260".4<"4880<<"152?":.0?"/6<:46:KDM"B2"<2?0260"846"H0"/6":.0"1/0K9"469"JIKK"4"928I?06:"152?":.0".049"211/80"<05;05"25"R"846"JIKK"4"928I?06:"152?":.0"L0<:"824<:"211/80"62"J52HK0?"4:"4KKM"

Page 107: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!+(""

STATENS HAVERIKOMMISSION (SWEDEN) SANNY SHAMOUN 3>G"V\J05:"d346@>08.62K2=D@G5=46/^4:/26e"

%"D5<"fI?46"148:25<"<J08/4K/<:"

!!,!*,+*"

"

W50K/?/645D"7<<0<<?06:Q"

GI5"4=068DS"J02JK0"L25O/6="L/:."2I5"H2459M">.0"BL09/<."488/906:"/6;0<:/=4:/26"H2459"L2IK9"92":.4:M">.0"824<:"=I459"?4D"455/;0"1/5<:"469":.0D"L/KK"62:/1D"I<"HI:"L0"926T:"90J069"26":.0?"125"46D"82KK08:/26"21"0;/90680"L0"92"2I5"2L6"/6;0<:/=4:/26M"

@"

b0K4:/26<./JQ"

R"926T:".4;0":.4:"?I8."0\J05/0680"L25O/6="L/:.":.0"?45/:/?0"49?/6/<:54:/26"HI:"?D"82KK04=I0<".4;0"469"152?"L.4:"RT;0".0459":.0D".4;0"4"=229"50K4:/26<./JM"

@"

G6"<8060Q"

`.06"<2?0:./6=".4JJ06<"L0"=0:"/6125?4:/26"152?":.0"J2K/80":.0D"84KK"I<"469":0KK"I<"46"488/906:".4<".4JJ0609M">.0"H2459"908/90<"L.0:.05"25"62:":2"/6;0<:/=4:0M"GI5":04?"=20<":2":.0"</=.:"469"L0"H0=/6"928I?06:/6=":.0"4504"

@"

V;/90680"J5/25/:DQ"

N0J069<"L.4:".4<".4JJ0609M"`0":5D":2"92"H2:.M"R1"L0"450"=2/6="4<"4":04?"L/:."#"25"$"<2?0:/?0<"%"25"&"J02JK0"L0":5D":2"9/;/90"<2"L0"846"92"0;05D:./6=M"

@"

N0J:.Q"

`0"12KK2L"4"8.08OK/<:"L0".4;0":2"<00"L.4:"L0"K22O"4:"469"82KK08:M"`0"4K<2"4KK".4;0"9/110506:"H48O=52I69<"469"L0"4KK".4;0"9/110506:"UI0<:/26<"L0"L46:":2"4<O"<2"H0:L006"I<"4KK"0;05D:./6="/<"82;0509M"

@"

Page 108: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!+)""

799/:/264K"j/</:<Q"

N0J069<"26".2L"H/=":.0"488/906:"/<M"R"<4D"I<I4KKD"!"25"#":/?0<M"R:"90J069<"26".2L"?46D"J02JK0"D2I".4;0":2":4KO":2"469"/1"D2I"846":4KO":2"4KK"21":.0?"26":.0"1/5<:":5/JM"R:T<"?250":.46"2680"R"846"<4DM""

@"

W52H4HK0"-4I<0Q"

`0"926T:".4;0"4"901/6/6="=I/90K/60M"`0":5D":2"=2"4<"145"4<"L0"846"469"L0"92":./<"HD"4<O/6=":.0"UI0<:/26"L.D"9/9":./<".4JJ06"469"L.06"L0"926T:"=0:"46"46<L05":2":./<"UI0<:/26"L0"H0K/0;0"L0".4;0"=260"145"062I=."/6:2":.0"84I<0"21":./<"488/906:M"

@"

764KD</<Q"

A2":.050"450"<2"?46D"?0:.29<":.4:"D2I"846"I<0"L0".4;06T:"901/609"4"?0:.29":2"I<0M"R":5D":2"I<0"4"?0:.29"84KK09"3#"?0:.29["G5"0;06:"84I<0"?0:.29"R":./6O"/:T<"H4<09"26"4"?0:.29"84I<0".I?46"J05125?4680"?0:.29M"`.06"D2I"92":.0<0"/6;0<:/=4:/26<"D2I":5D":2".4;0"4":04?"L/:."<J08/4K/<:<"L/:."9/110506:"H48O=52I69<M"k2I":5D":2"92"46"0;06:"84I<0"464KD</<M">.050"450"<2"?46D"?0:.29<"2I:":.050"469"R"J50105"H084I<0":.0"?0:.29"504KKD"128I<0<"26":.0"9/508:/26"H0:L006"?46":08.62K2=D"469"25=46/^4:/26M"""

@"

>/?0Q"

_4L"<4D<"/:"<.2IK9":4O0"260"D045M"`0".4;0":2"1/6/<.":.0"50J25:"HD"260"D045"41:05":.0"488/906:"HI:"H084I<0"L0T50"62:":.4:"H/="/:"<2?0:/?0<":4O0<"?250":.46"260"D045M"

@"

b0J25:"L5/:/6=Q"

B2?0:/?0<"260"J05<26"/<"L5/:/6="4"50J25:"469"<2?0:/?0<":4O0<"J45:<"152?"2:.05"L5/:05<M"R:"90J069<"26".2L"?46D"J02JK0"450"L5/:/6=":.0"50J25:"/:"82?JK/84:0<"/:"L/:."?250"J02JK0"0;06":.2I=."/:T<"?250"50<2I580<M""

@"

b09I69468DQ"

`0".4;0":L2"<:0J<":2"L5/:/6=":.0"50J25:M"`0"<069"/:":2"4KK"/6;0<:/=4:25":04?"?0?H05<"469":.0D"<4D"L.4:":.0D":./6O"4H2I:"/:M"769":.06":.0"/6;0<:/=4:/26"K049".4<":2"J5221":.4:"50J25:M"769"L.06"L0":./6O"/:T<"=229"062I=."L0"<069"/:":2":.0"2:.05"4=068/0<":.4:".4;0"<2?0"/6:050<:"/6":.0"50J25:":.4:"L0"L25O09"L/:.M">.0D"846"<4D"L.4:":.0D":./6O"4H2I:":.0"50J25:"

Page 109: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!+*""

H01250"/:T<"1/64KM"7K<2":.0"J02JK0"L0".4;0"/6:05;/0L09"846"5049":.0"148:<"L0".4;0"=4:.0509"152?":.0/5"/6:05;/0L<M"

@"

B21:L450Q"

A2M""

Page 110: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!!+""

MARSHALL ISLANDS MARITIME AUTHORITY ALAN BLUME N0JI:D"-2??/<</2605"21"345/:/?0"7114/5<""

!!,#$,+*"

#"D5<"

EB-F"

"

-I5506:"W2</:/26Q""

RT;0"H006"2;05<00/6="8I5506:"/6;0<:/=4:/26<"4<"L0KK"4<"92/6="<2?0"?D<0K1M"

@"

W50K/?/645D"7<<0<<?06:Q"

32<:":/?0<"L0"92"/:"/6".2I<0M"7"K2:"21":.0":/?0<"L0"L/KK"=0:":.0"50J25:"21"4"84<I4K:D"152?"4";0<<0K"2J054:25"469"H4</84KKD"L0"L/KK"92"4":5/4=0"469"908/90"/1"L0"6009":2"<069"<2?0H29DM""

@"

FI/90K/60<"125"_4I68.Q"

W5/?45/KDS"L0T50"/6":.0"J5280<<"21"50;/</6=":.0"J5280<<"125":.0"/?JK0?06:4:/26"125":.0"R3G"8290M"R:T<";05D"?I8."26"4"84<0"HD"84<0"H4<0<":.4:"L0"908/90":2"K4I68."25"62:M">./<"/<"H4<09"26"8I5506:"50<2I580<"469"L25OK249S"<2?0:/?0<"L0"?4D"/6;0<:/=4:0"50?2:0KD"469"4<O":.0?":2"82KK08:":./6=<"469"<069"/:":2"I<M"

@"

G6"<8060Q"

>DJ/84KKD"/:T9"H0"4":04?"21"260"?4DH0":L2"469"L0"954L"4<"?I8."4<"L0"846"26"2:.05"25=46/^4:/26<"L25O/6="/6":.0"4504<M"`0":5D":2"?4\/?/^0"26"L.4:"/<"4K5049D"H0/6="9260"HD"2:.05<M"R1"L0"6009"46"0\J05:"826:548:25"/6"4"J45:/8IK45"4504"L0TKK"L25O"L/:.":.0"2J054:25"<4D/6="D2I"92":.0"826:548:/6="469"L0TKK"=0:":.0"/6125?4:/26M"7":DJ/84K"52I:/60"/<"L.4:"92"L0"O62LS"4<<0<</6="L.4:"/<"O62L6S"L.4:":DJ0"21"84<I4K:D"L4<"/:S"L.050"469"L.06"L4<"/:S"469":.06"<:45:/6=":2"JI:":.0"UI0<:/26<":2=0:.05"152?":.050":4O/6="/:"J/080"HD"J/080M"7=4/6"/:"90J069<"26":.0"64:I50"21":.0"488/906:M"

@"

Page 111: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!!!""

V;/90680"W5/25/:DQ"

F/;0":.0"</:I4:/26"L./8."260"450"D2I"?2<:"K/O0KD":2"K2<0"1/5<:"469":.06":2":5D"469"J50<05;0"1/5<:M"R<"/:"<./1:"8.46=0"469":.0DT50":5D/6=":2"=0:"J02JK0".2?0"469":.0"148:":.4:"?0?25/0<"<:45:":2"1490M"7:":.0"<4?0":/?0"9/508:/6=":.0"8./01"06=/6005":.0<0"450":.0"928I?06:<":.4:"6009":2"H0"JI:":2=0:.05M"

@"

799/:/264K";/</:<Q"

F06054KKD"62M"

@"

N0J:.Q"

`0"926T:".4;0"46D:./6="928I?06:09"/:":069<":2"14KK"26":.0"/6;0<:/=4:25"469"L25O/6="L/:.":.0"211/80"H48O"469"125:.M"

@"

W52H4HK0"-4I<0Q"

R6":05?<"21"4"L5/::06"<:469459"901/6/:/26S"62M"R6":05?<"21"4"L25O/6="901/6/:/26":.4:T<"</::/6="92L6"469"L.4:"/<":.0"J2:06:/4K"<82J0"21"84I<4K"148:25<"125":./<"J45:/8IK45"84<I4K:DM"769":.06"<:45:/6=":2"L25O"H48O"L./8."260"/<"J52\/?4:0"469"4::0?J:/6=":2"/906:/1D":.4:"260"522:"84I<0":.4:"260"0;06:":.4:"/1"/:"L050"62:":2".4;0".4JJ0609"L2IK9":.0"488/906:".4;0"82?0":2".4JJ06M"`0"5082=6/^0":.050T<"4"L.2K0"8.4/6"21"84<I4K"148:25<":.4:"K049"IJ":2":.0"488/906:M"

@"

764KD:/84K"30:.29Q"

A2"L0"926T:".4;0"4":0\:H22O"J52809I50":.4:"L0"I<0M"`.4:"L0"92"62L"/<";05D"?I8."H4<09"26":.0"0\J05/0680"21":.0"/69/;/9I4K<"92/6=":.0"/6;0<:/=4:/26"469"2:.05<"L.2".4;0"50;/0L09"/:":2"1KI<."/:"2I:M"

@"

b0J25:"`5/:/6=Q""

R:"504KKD"90J069<h"4"K2:"21":./<"846"H0"9260"50K4:/;0KD"UI/8OKD"2680":.0"84<I4K:D".4<"288I5509M"R6":05?<"21"82?JK0:/26"469"=2/6="152?"<:45:":2"1/6/<.09"J529I8:":.4:"846"H0"41108:09"HD"L25OK249"H084I<0"D2I"?/=.:"H0"L25O/6=":L2"25":.500"/6;0<:/=4:/26<"4:"4":/?0M"R"<4D"L0"<.22:"125"'"?26:.<"HI:"RT9"<4D":.4:T<"?250"21"4":45=0:M"G6"<8060@"!@%"94D<S"464KD</<,L5/:/6=""

@"

Page 112: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!!#""

-I55068DQ"

R6"499/:/26":2"H0/6="/6<J08:25<"L0T50";05D".469<"26"L/:."L.4:T<"=2/6="26"/6":.0"1K00:M"`0"4K<2".469K0"4"K2:"21"_AF";0<<0K<"<2"L0"JI:"26"4"82I5<0"125"2I5"J05<2660K":.4:"904K"L/:."_AF";0<<0K<M"769":.050"450"10L"21"I<"L.2"L/KK":4O0"J45:"/6":54/6/6="4<"/:"45/<0<"469"<.450":.0"L04K:.M"

"

_AF@K/UI/9"64:I54K"=4<"

Gc>@26":.0"Z2H":54/6/6="

@"

B21:L450Q"

A2S"L0".4;0"I<09"4"928I?06:/6="84:4K2="ZI<:":2"25=46/^0"0;05D:./6="HI:"2:.05L/<0"62M"

Page 113: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!!$""

MARINE ACCIDENT INVESTIGATION BRANCH (UNITED KINGDOM) MIKE TRAVIS !#,$,+*"

"

-I5506:"W2</:/26Q"

W5/68/JK0"R6<J08:25"21"H548O0:"90;0K2J?06:"

&"J5/68/JK0"/6<J08:25<S"L/:."J02JK0"L25O/6="I6905":.0?"

f049"21":08.6/84K"90J45:?06:"

>22O"2;05":./<"D045"152?":08.6/84K"?464=05"

&"D045<"4:":.0"37RC"

7:"<04"125"!#"D045<"4<"4"?45/60"06=/6005"

B/^0"q"%#"/69/;/9I4K<"

GJ054:/264K"</90S"("211/805<S"!'"/6;0<:/=4:25<S"!":08.S"!"?464=05S"%"

@"

W50K/?/645D"7<<0<<?06:Q"

G6KD"K284:09"/6"B2I:."f4?J:26S"<2"0/:.05"J2K/80"25"-24<:"FI459"469":.0"488/906:"/<"50J25:09":2":.0?M"F/;06"/6<:5I8:/26<":2"<:4D"8K045"21"46D:./6=":.4:"?/=.:"H0"826<:5I09"4<"0;/90680M"E6K0<<":.0D"450"4L450"21"90=549/6="0;/90680S":.0"1/5<:"50<J26905<"450"4<O09":2"<:005"8K045"21"0;/90680M"W50105":2"=4:.05"2I5"2L6"0;/90680M"

@"

b0K4:/26<./JQ"

b0K4:/26<./J"=20<":.52I=."J.4<0<h"L0".4;0"J5/?48D"125"4KK"0;/90680M"V;05D260"I6905<:469<":.0/5"2L6"52K0S"HI:"<2?0:/?0<":.0D"92"=5/69"=045<M"

@"

A2:/1/84:/26"

A2:/1/84:/26"152?"-24<:"FI459"25":.52I=."K/;0"0;06:"122:4=0M"B2?0:/?0<"4"?250"211/8/4K"62:/1/84:/26M"79?/6/<:54:/;0"211/805"JI:<"26"94:4H4<0"469"/??09/4:0KD":4O0<"/:":2"J5/68/JK0"

Page 114: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!!%""

/6<J08:25<":2"=4:.05":04?"125"9I:D,<:469HDS"0:8M"NI:D"J5/68/JK0"?4O0<"1/5<:"4<<0<<?06:"E<I4KKD":L2"/6<J08:25<S"HI:"<2?0:/?0<"4"L.2K0":04?"/1"/:"/<"4"?4Z25"488/906:M"

@"

G6"<8060Q"

-26:/6I0<":2"?4O0"45546=0?06:<"152?"H4<0":2"?4O0"<I50"<8060"/<"J52:08:09"469"H0=/6":2"45546=0"/6:05;/0L<M"R6:05;/0L"L/:60<<0<S":.06"/6<J08:"<8060"9I5/6=":.0"/6:05;/0L"J5280<<M"G:.05"J02JK0"H082?0"4JJ4506:":.52I=.":./<"J5280<<"469":.06":.0D"/6:05;/0L":.0<0"J02JK0"4<":.0D"H082?0"4JJ4506:M""

@"

799/:/264K"j/</:<Q"

E<I4KKD"#"94D"90JK2D?06:"82?/6="H48O"26":.0":./59"94DM"N26T:"I<I4KKD"=2"H48OS"?4O0"0;05D"4::0?J:":2"?4O0"<I50":.0D"=4:.05"0;05D:./6=":.0D"6009"/6"260":5/JM"R1"L0"92"50:I56":2":.0";0<<0KS":.06":.0"<./J<"<:411"/<"62"K26=05":.050S"0:8M"R1":.0"<./J"6009<":2"H0"<4K;4=09S":.06":.0D"92".4;0":2"=2"H48O"25"/1":.0D"504K/^0":.0D"6009"0\:54"0;/90680":.4:":.0D"9/9"62:"=4:.05M"

@"

N0J:.Q"

7K?2<:"06:/50KD"H4<09"26"0\J05/0680"21":.0"/6;0<:/=4:25<S":4KO":2".049"/6;0<:/=4:25<"H48O"4:"H4<0":2"?4O0"<I50":.0D".4;0"0;05D:./6=S"<2"82KK4H254:/26M"b01050680<"4<"4"J50@90JK2D?06:":54/6/6="4/9":2"06<I50"/6;0<:/=4:25<"450"82?125:4HK0"L/:.":.0"J5280<<"L.06"455/;0"26"</:0"

@"

V;/90680"-.08OK/<:Q"

E<09"HD":.0"522O/0<S"HI:":.0D"?490"/:"H0::05"HD":5D/6=":2"<.2L"4KK"60L"/6<J08:25<"4";45/0:D"21"488/906:"/6;0<:/=4:/26<"<2":.0D"=0:"9/110506:"J05<J08:/;0<M"E<0"21"8.08OK/<:<"846"H0":22"826<:54/6/6=S"/1"/:"/<6T:"82?JK0:0"062I=."25"/1"/:"/<":22"K26=S"0:8M"

@"

W52H4HK0"-4I<0Q"

A2"901/6/:/26S"I<0":.0"H4</<"H0K2L"

@"

-4I<4K"764KD</<Q"

C4<09"26"-4649/46"L/:."46"488/906:"464KD</<"9/4=54?"L/:.":/?0K/60M"`0"E<0"/:"26@</:0"469"L.06"50:I5609":2"H4<0M"`0"JI:"/:"26"L./:0H2459"L/:.":.0"L.2K0":04?M"E</6=":.0"488/906:"

Page 115: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!!&""

464KD</<"9/4=54?"L0"=2"2;05"4KK"</=6/1/846:"0;06:<"26":.0":/?0K/60M"G:.05"?290K<S"-4649/46"?290K".49"4"K2:"21"464KD</<":.4:"=20<"=504:KD"152?":.0"0;/90680":.52I=.":.0"5082??0694:/26<M"E<0"84I<0"469"0;06:<"9/4=54?":2"<I/:";284HIK45D"4K<2":2"901/60":.0"522:"84I<0<M"

@"

W45:D"BD<:0?Q"

P4/5"/904"21"L.4:":.0"5082??0694:/26<"82IK9"H0S"<2":.0D"=4:.05":.0"J02JK0":.4:"L2IK9"/?JK0?06:":.0<0"5082??0694:/26<"469".4;0"4"?00:/6="L.050":.0D"4<</<:":.0?"/6"?4O/6="4"104</HK0"5082??0694:/26M"b082??0694:/26<"450"X<?45:"469"L25O4HK0YM"W02JK0"L.2"450"8455D/6="2I:":.0<0"5082??0694:/26<"450"/6;2K;09"/6":.0"J5280<<"<2"K0<<"50KI8:46:":2"492J:"L.06":.0D"450"/<<I09M"`25O/6="L/:.":.0"J02JK0"L.2":.0D"450"<85I:/6/^/6=".0KJ<"L/:.":.0"J5280<<"469"822J054:/26M"

@"

_06=:.Q"

_06=:."21"464KD</<"

WIHK/<."5082??0694:/26<"L/:./6"'"?26:."=24K"/6"504K/:D"/:"920<"=0:"<:50:8.09M"E<I4KKD"?/625"488/906:<"846"O00J"IJ"L/:.":.4:M"

"

`./:0H2459"?00:/6="L/:./6"#"L00O<".4;0"4"J50<06:4:/26"5049D"125"0;05D:./6=":.4:"/6;2K;0<":.0"488/906:"469"J02JK0"/6":.0"H5468."82??06:"26"L.4:"?4D".0KJ":.0"/6;0<:/=4:/26"4K26="

-./01"/6<J08:25"908/90<"26"H4</<"21"5082??0694:/26<":.4:"82IK9"82?0"2I:"152?":.0"488/906:S"82IK9"<:4D"4<"4"J50K/?/645D"0\4?/64:/26"/1"5082??0694:/26<"92"62:"<00?":2"H0"</=6/1/846:"L/:."62"JIHK/<.09"50J25:"

C4<09"<2K0KD"26"K0<<26<":.4:"6009":2"H0"K045609"152?":.0"488/906:"

@"

b0J25:"`5/:/6=Q"

rB0::/6="4"94:0"125"5082??0694:/26<"?00:/6=S"<0:"045KDS"HI:"H4K46809"L/:."L.06":.0D"846"1/6/<.":.0"464KD</<"

r$"L00O<"125"464KD</<S"HI:"82IK9"H0"K0<<"

rNI:D"<D<:0?"12580<":.0"25=46/^4:/26"21":.0":04?":2"H0"5049D"125"L./:0H2459"464KD</<"

r7KK284:0"50<2I580<"L/:./6"4"?26:."21":.0"488/906:"

Page 116: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!!'""

rG680"0;05D":L2"L00O<".4;0"?00:/6=<"

r-2IJK0"488/906:<":.4:"/68504<09"K06=:."9I0":2"/KK60<<"

rW50<<I50"152?":.0"H2<<0<"

@"

`25OK249Q"

>04?<"I<I4KKD".4;0"?IK:/JK0"84<0<"=2/6="

7KK284:0"9/110506:"Z2H<"<I8."4<"K049"/6<J08:25"

&"8268I5506:"/6;0<:/=4:/26<S"62L"<J5049"/<"?250"0;06"

B2?0":04?<"H082?0"?250"K24909S"HI:"/:"ZI<:".4JJ06<"

`.06":./<"0\05:<"0\:50?0"J50<<I50S"H4</<"21"4;4/K4H/K/:D":4O0<"J5/25/:D"

7KK"21":.0"/6;0<:/=4:25<"450"8./01"06=/6005<"25"64;4K"458./:08:<"

V48."260"=0:<"62?/64:09":2"46"4504"4<"4"<J08/4K/<:S"/:"/<":.0/5"50<J26</H/K/:D":2"O00J"IJ"L/:.":08.62K2=/84K"49;4680<"/6":.0/5"4504<"469"<.450"L/:."0;05D260"

R1":./<"/<":22"?I8."L25O"125"260"J05<26S":.06"

R1"600909S":.0D"846"92"/[email protected]<0":54/6/6="25":54/6/6="L/:."462:.05"4=068D"

@"

N4:4H4<0Q"

N0806:"J5280<<"?464=0?06:":22K":.4:":.0D"450"K22O/6="125"

-I5506:"94:4H4<0"/<"2OS"HI:"450"H0/6="JI<.09":2L459<"4"VI52J046"N4:4H4<0"

3464=0?06:"469"82<:"/<"H0D269"L.4:":.0D"846"?I<:05"

C4</8"211/80":22K<"L25O"<I11/8/06:KD"L0KK"

-2IK9"?4O0"I<0"21"2:.05":22K<S"HI:"/:"/<6T:"600909"

@"

b0J25:"`5/:/6="

R6/:/4K"/6<J08:25"L5/:0<"/:S":.06"L/:.":.0":04?":.0D"JI:"IJ"4"9541:":2"8./01"/6<J08:25"

>.06":.0"8./01"/6<J08:25"JI:<"/:"2I:":2":.0"JIHK/8"

Page 117: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!!(""

b0J25:"L5/:/6="<O/KK<":54/6/6=":2".0KJ":.0"/6;0<:/=4:25<"

`46:":2"O00J":.0?"J508/<0"469"<.25:"

@"

788/906:">5069<Q"

7"K2:"21"1/<./6=";0<<0K"488/906:<M"G1:06";05D"15I<:54:/6="H084I<0"62"?4::05".2L"?46D"5082??0694:/26<"450"?490S"488/906:<"<:/KK":069":2".4JJ06M"

"

>/?H05";0<<0K"488/906:<"9I5/6="J05/29<"

P4:/=I0"H4<09"=52I69/6=<"

3250"<?4KK";0<<0K"488/906:<"

@"

N0J:."R<<I0<Q"

V<<06:/4K":2"82;05"4KK"148:<"152?":.0"488/906:"

A2:"0<<06:/4K"125"90:4/K"21"464KD</<"/6":.0"50J25:"/1":.0"148:<"464KD^09"450"62:"/?J25:46:":2"5082??0694:/26<"

V\:0564K":0<:/6="450"/68KI909S"HI:"4<"46"4660\"<2"62:":2"?4O0":.0"50J25:"K26=05"

a00J":.0"50J25:"J508/<0"

b0806:KD"492J:09"?0:.29"21":4O/6="62:0<"9I5/6="/6:05;/0L<"

B2":.4:"0;05D260"I6905<:469<"0;05D260T<"/6:05;/0L<"

Page 118: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!!)""

AUSTRALIAN TRANSPORTATION SAFETY BUREAU MIKE WALKER 7>BC"7;/4:/26"

!#,!+"

B5M"R6;0<:/=4:25""

!%"D5<""

"

W50K/?/645D"7<<0<<?06:"

R:";45/0<"UI/:0"4"H/:".050"/:"82IK9"H0"J2K/80S"0?05=068D"<05;/80<S"25":.0"2J054:25M"R1"/:"/<"?250"21"46"4/5K/60"/68/906:":.06"D2I"0\J08:"/:":2"H0":.0"4/5K/60"25":.0"7>-"K0:"I<"O62LM"R1"/:T<"4"?250"21"46"488/906:"D2I"L2IK9"0\J08:"/:":2"H0":.0"J2K/80"26"GNB>7b["

"

345/60Q"A2"-24<:"FI459"29<:45"82;05<"?45/60"4<"L0KKM"k04.":.0D"L2IK9"=0:"62:/1/09"HD"GNB>7bM"7K<2":.050"/<":.0"7I<:54K/46"?45/:/?0"B410:D"7I:.25/:D"d73B7e"469":.0DT50"K/O0"4"64:/264K"25=46/^4:/26M"`/:."?45/60"4:":.0"?2?06:":./6=<"K/O0"1/<./6=";0<<0K<":.4:"<25:"21"8541:":.0DT50"4KK"50=/<:0509"469"904K:"L/:."26"4"<:4:0"HD"<:4:0"H4</<M"769"048."<:4:0"920<":.0/5"2L6"/6;0<:/=4:/26M"B2"73B7"904K"L/:."ZI<:"/6:0564:/264K"<./JJ/6=M"

"

7;/4:/26Q"/6":05?<"21":.0"=06054K"4;/4:/26"488/906:"/:"L/KK"H0"<08I509"K/O0"4"85/?0"<8060"469"L4/:"125"I<M"R1"L0"846T:"=0:":.050"125"<2?0":/?0"L0"L/KK"4<O":.0?":2":4O0"J/8:I50<"469"928I?06:":.0"<8060M"E6905"2I5"K0=/<K4:/26":.0D"450"4KK2L09":2"50?2;0"H29/0<"469"J05<264K"01108:<".2L0;05S"/1":.0D"92":.4:"L0"4<O":.0?":2":4O0"J/8:I50<"H01250":.0D"?2;0"/:0?<"<2":.4:"L0"846"<00M">.4:"/<"4H2I:":.0/5"/6;2K;0?06:M"

"

G6"<8060"d0;/90680"-.08OK/<:e"

>.4:T<"H006"4"H/:"21"826:06:/2I<"J2/6:M"B2?0"/6;0<:/=4:25<"90;0K2J":.0/5"2L6"HI:"125":.0"?2<:"J45:"62S"L0"926T:"I<0"8.08OK/<:<M"`0"ZI<:"1/69":.4:":.050"/<"<I8."4";45/0:D"21"</:I4:/26<":.4:"62"8.08OK/<:"846"82;05":.0?"4KK"469"L0":069":2"50KD"26":.0"0\J05/0680"469":04?L25O"L/:./6":.0":04?":2"82;05"0;05D:./6=M">.050"450"8.08OK/<:<"469":.0D"450"805:4/6KD"I<09"H0::05"4<"=I/90K/60<M"

"

Page 119: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!!*""

V;/90680"W5/25/:D"

R"=I0<<":.4:"920<"90J069S":.4:"L2IK9"504KKD"90J069"26":.0"</:I4:/26M"R"=I0<<"L0"L2IK9"504KKD"K/O0"4<"?I8."/6:0KK/=0680"26":.0"</:I4:/26"H01250"L0"=2":2":.0"488/906:"</:0M"R6"=06054K"/:"L2IK9"H0"=2":2":.0"</:0"469"?4O0"928I?06:4:/26M"769":.06"L508O4=0"469"L/:60<<0<"125"=06054K"4;/4:/26M"R:<"12I69"L/:.":.0"L/:60<<":.4:":.0/5"?0?25D"/<"H0::05"/6":L2".2I5<":.46"/6":L2"94D<":.46"/6":L2"L00O<":.06"/6":L2"?26:.<M"f2L0;05"</680"L0"J52H4HKD"L26T:"=0:":2":.0?"L/:./6":2".2I5<":.0D"926T:"069"IJ":.0"1/5<:"J5/25/:D"2680"26@<8060M"`0"L46:":2"=0:"J05/<.4HK0"0;/90680":.4:"L2IK9"H0"41108:09"HD"54/6"469":.2<0":./6=<M"

"

_06=:."26@<8060"

P25"4"=06054K"4;/4:/26"488/906:"J52H4HKD"<4D"#@%"94D<"26":.0"</:0M"7/5":546<J25:"L./8."L0T;0".49"50806:KD"6260"21"L./8."L050"14:4K":4O0"4"K/::K0"K26=05":.0D"L2IK9"=2"&@'"94D<M""

"

799/:/264K";/</:<"

R:T<"H006"4"<250"J2/6:"/6":.0"K4<:"D045"/6;0<:/=4:25<".4;0".49":2"=2"H48O"H084I<0"<2?0:./6="L4<"?/<<09"<2":.0D".49":2"=2"H48O":2":.0"L508O4=0"469"1/69"H/:<M">.0"/904"/<"K0:T<"92"0;05D:./6="L0"846"0;06"/1"/:":4O0<"462:.05"94D"K26=05"469"926T:"5I<."469"4;2/9".4;/6=":2"=2"H48OM"B2?0:/?0<"=2/6="H48O"/<"I64;2/94HK0"H084I<0"L0"1/69"4"14/KI50"L/:."<2?0"<D<:0?":.4:":.050"L4<"62"L4D"L0"82IK9"H0"4L450"21":.4:"L/:.2I:"1I5:.05"50<0458.M"CI:"/6"=06054K"4K?2<:"4KK"0;06:<"L0"<.2IK96T:"6009":2"=2"H48OM"B2"L0T;0"H006"=/;/6="J02JK0"?250":54/6/6="4H2I:"H0/6="26":.0"</:0"469"/6:05;/0L/6="26":.0"</:0M"B2"499/:/264K";/</:<"L2IK9"H0";05D"K2LM"

"

W52H4HK0"84I<0"

k0<"4H<2KI:0KDM"`0"926T:"84KK"/:"J52H4HK0"84I<0"L0".4;0"H/="J52HK0?<"L/:."I</6=":.0"L259"84I<0"469"L/:.":.0"82680J:"21"J52H4HK0"84I<0"4<"92"?46D"4=068/0<"/6":.0"/69I<:5DM"769"L0T;0"L25O09"UI/:0"0\:06</;0KD"26":.0"4504"RTKK"<069"D2I"<2?0"J4J05L25O"26":.0":2J/8M">.0":05?"L0"I<0"/<"X826:5/HI:/6="<410:D"148:25<Y"469"901/6/6="<410:D"148:25"469":.06"L.4:"/<"4"826:5/HI:/6="<410:D"148:25M"769":.0"901/6/:/26"/<"UI/:0"9/110506:"152?"2:.05<"469"UI/:0"90:4/K09"469"I<09":.52I=.2I:":.0"4=068DM"`.06"L0"?05=09":.0"9/110506:":546<J25:"9/;/</26<"4KK":.0"9/;/</26<"I<0":.0"901/6/:/26"62LM"R"846":0KK"A0L"s04K469"920<6T:".4;0"4"901/6/:/26"HI:":.0DT50"K22O/6="4:"2I5<M"-46494".4<"<2?0"O/69"21"901/6/:/26"HI:":.0D".4;06T:"0K4H254:09"/:M"37RC":.0D"926T:".4;0"46D"<:469459"901/6/:/26"I<09":.52I=.2I:":.0"4=068DM"R"H0K/0;0"62L":.0"R3G"901/6/:/26"/<";05D"</?/K45":2":.0"7>BC"901/6/:/26"/:"?/=.:"I<0"4"9/110506:":05?"HI:"/:T<"UI/:0"</?/K45"/6"<:5I8:I50":2"2I5"901/6/:/26M"

"

Page 120: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!#+""

-4I<4K"764KD</<"

`0"90;0K2J09"<2?0:./6="84KK09"464KD</<"J2K/8/0<"469"464KD</<"=I/90K/60<"469":22K<M"769"805:4/6KD"<2?0"?290<"450"492J:/6=":.0<0"/904<"?250":.46"2:.05<M"R6"?45/60S"/6;0<:/=4:25<"J52H4HKD".4;06T:"H006"I</6=":.0"?4:05/4K<"4<"?I8."4<"2I5"4;/4:/26M">.050"450":./6=<"L0"L46:"4KK"/6;0<:/=4:25<":2"92M"`0"L46:":.0?":2"92"<2?0"125?"21"<0UI0680"21"0;06:<"464KD</<"25":/?0K/60"21":.0"0;06:M"`0"L46:":.0?":2"92"4":.252I=."J5280<<"21"/906:/1D/6="4KK":.0"148:25<"=/;06"=06054K":22K<"469"8.08OK/<:<":2"826</905M"`0"L46:":.0?":2"=2":.52I=."4":.252I=."J5280<<"21":0<:/6="4KK"21":.2<0":./6=<"L0".4;0"<2?0:./6="84KK09"46"0;/90680":4HK0"L0/=./6="4KK":.0"H/:<"21"0;/90680"4=4/6<:"048."2:.05"125"/:<"/6;2K;0?06:M""

"

b0J25:"L5/:/6="

_0;0K"%@"K2L0<:"K0;0K"21"/6;0<:/=4:/26"@'"?26:.<")"J4=0<"

_0;0K"$@"!#"?26:.<"

_0;0K"#@"!)"?26:.<"

_0;0K"!@"L0T;0"60;05".49"260"21":.0<0"

"

`0"926T:"=0:"4"K2:"21"2I5"50J25:<"2I:"L/:./6":.0<0":/?0"154?0<M"b09I69468DQ"D0<"/:T<"4"?4Z25"J52HK0?"125"I<M"`0".4;0"L.4:"/<"84KK09"4"J005"50;/0L05"90J069/6="26"L.4:":DJ0"21"/6;0<:/=4:/26":.0D"50;/0L"469":.06"J4<<"26":2":04?"K04905"469"<2"26M"V48."<:4=0"846":4O0"<2?0":/?0"/6":.0"J4<:"L0T;0".49"J52HK0?<"L/:."L5/:/6="<:DK0<"469"L259"I<4=0":./6=<":.4:"926T:".4;0"46D"504K"41108:"26"50494H/K/:DM"`0T50":5D/6=":2"0682I54=0":.0"J./K2<2J.D"/<"D2I"<.2IK9"1/6/<."4KK"464KD</<"H01250"92/6="D2I5"50J25:"L5/:/6=M"B2?0"<:45:"L5/:/6="H01250"4KK":.0"464KD</<"/<"9260"?4O/6="/:":4O0"4"K26=":/?0M"

"

"

"

-I55068D"

R6":05?<"21"<D<:0?<S"125"<2?0"/6;0<:/=4:25<":.0D"H0K/0;0":.0D"<.2IK9"H0"<J069/6="?250":/?0":54/6/6=":.46"92/6="/6;0<:/=4:/26M"R6":05?"21"/6;0<:/=4:/26":54/6/6="L0".4;0"<2?0:./6="84KK09"4"9/JK2?4"L/:."!#"?29IK0<"L./8."R"=I0<<"L2IK9"H0"84KK09"26":.0"Z2H":54/6/6=M">.0D"/68KI90"82??I6/84:/26S"/6:05;/0L/6=S"?464=/6="46"/6;0<:/=4:/26S"I</6="4"82?JI:05S"0:8]"L0"4K<2".4;0"4"8K4<<"4"12I5"94D"82I5<0"26"84I<0"464KD</<"L./8."L0"92h"469".4;0"4"#"94D"82I5<0"26"/6:05;/0L/6=h"260"L00O"82I5<0"/6".I?46"148:25<"125"4KK"/6;0<:/=4:25<M"

Page 121: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!#!""

"

B21:L450"

`0"90;0K2J09"2I5"2L6"<D<:0?"84KK"BRR3B"dB410:D"R6;0<:/=4:/26"R6125?4:/26"3290<"469"BD<:0?<e"/:T<"</?/K45":2":.0"-4649/46">R3BM"7KK"2I5"928I?06:4:/26"/<"0K08:526/8"469"=20<"26":./<"<D<:0?"L./8."/<"H4</84KKD"12K905<"D2I"846"25=46/^0"/:":.4:"L4DM"V;05D:./6=".4;/6=":2"92"L/:.":.0"/6;0<:/=4:/26"/<"<:2509":.050"469"/6"504K":/?0M"`0"4K<2"I<0"B.450W2/6:"L/:."K2=<"469"928I?06:"?464=0?06:S"826:48:<S"J52Z08:"4<J08:<S"469"82??I6/84:/26<M"

Page 122: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!##""

JAPAN TRANSPORT SAFETY BOARD MABITO HAMADA 345/60"788/906:"R6;0<:/=4:25"

!#,!+"

"

W50K/?/645D"7<<0<<?06:Q"

R:"90J069<"26"</:I4:/26"469"O/69"21"84<I4K:DM">DJ/84KKDS"L0"=0:"/6/:/4K"/6125?4:/26"21"84<I4K:D"152?"c4J46"-24<:"FI459M"`0"?490"?0?25469I?"21"I6905<:469/6="L/:."c-F"50=459/6="J52;/9/6="/6/:/4K"/6125?4:/26M"71:05"L0"=54<J"/6/:/4K"/6125?4:/26"50=459/6="288I550680"21"84<I4K:DS"2I5"/6;0<:/=4:/26"L/KK"H0"8455/09"2I:"4<"<226"4<"J2<</HK0"HD"/6;0<:/=4:25"H0K26=/6=":2"c>BC"25"H0K26=/6=":2"9/<:5/8:"211/80M"

@"

b0K4:/26<./JQ"

C4</84KKD"/6/:/4K"/6125?4:/26"21"288I550680"/<"062I=."125"2I5"/6;0<:/=4:/26<S"469"2I5"/6;0<:/=4:/26"/<"8455/09"2I:"<?22:.KD"I6905":.0"3GE"J50;/2I<KD"?06:/2609S"<0J454:0KD"152?":.0"c-FM"R6";/0L"21"J52;/9/6="/6/:/4K"/6125?4:/26S"50K4:/26<./J"/<"=229M"

@"

G6"<8060Q"

R6<J08:/6=":.0"K284:/26h"?04<I5/6=S":4O/6="J.2:2<S":4O/6="<4?JK0S"<O0:8./6=S"0:8]"F4:.05/6="0;/90680M"V<J08/4KKDS"508259<"21"7RBS"jNb"469"54945"d2HZ08:/;0"0;/90680e"450"/?J25:46:M"R6:05;/0L/6="/6"J05<26S"UI0<:/26/6="HD"928I?06:"469"K/<:06/6="HD"J.260"84KKM""

@"

799/:/264K"j/</:<Q"

R:"90J069<"26":.0"</:I4:/26"21"84<I4K:DM"N450":2"<4DS"I6:/K"R"100K":.4:"=4:.05/6="21"0;/90680"/<"<I11/8/06:M"

@"

N0J:.Q"

_/O0"46DH29DS"R"84662:"8261/5?":.4:"R"?4O0"<I50"4KK":.0"148:<"450"=4:.0509M"N450":2"<4DS"/1"R"82IK9"?4O0"1/64K"/6;0<:/=4:/26"50J25:"<?22:.KD"469"64:I54KKD"L/:."JI50"?/69S"R"100K"4H2;0"<I50KDM"

Page 123: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!#$""

@"

V;/90680"-.08OK/<:"

E6905"J50J454:/26"21":.0"/6;0<:/=4:/26"?46I4KM"R6"?D"2J/6/26S"928I?06:09"?0:.292K2=D"/<"62:"6080<<45D"125"82KK08:/6="488/906:"148:<M"`0"82IK9"I6905<:469".2L":2":504:"488/906:"148:<S"0<J08/4KKD"0;/90680S"HD"R3G"8290M"32<:"/?J25:46:"/<<I0"/<".2L":2"J52:08:"L/:60<<0<"469"826805609"J45:/0<S"R":./6OM"

@"

W52H4HK0"-4I<0Q"

R:"90J069<"26"</:I4:/26"469"/:"/<"9/11/8IK:":2"0\JK4/6M"CI:S"R"82IK9"<4D":.4:"/:"L/KK"H0"901/609"HD"0;06"0D0"469"?/69"L/:."62"O62LK09=0M"

@"

>/?0"26"<8060Q"

R:"90J069<"26"</:I4:/26"

@"

N0J:.Q"

P/64KKDS"L0"90:05?/60":.4:"HD"9/<8I<</26"21":.0"C2459M"

@"

-I55068DQ"

`.06"R"1480"9/11/8IK:"J52HK0?<S"R"4<O"0\J05:<"125"464KD</<M"B2S"/:"/<"/?J25:46:":2"?4O0"826608:/26<"L/:."0\J05:<M"

"

R"L2IK9"K/O0":2"5080/;0":54/6/6="21"V6=K/<."125"06:05/6="R6:0564:/264K"?00:/6=M"3D"V6=K/<."/<"?45/605T<"<:DK0S"62:"1/:"125"259/645D"</:I4:/26"K/O0"4<"4H2;0"?00:/6=M"

Page 124: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!#%""

"

ADDITIONAL COMMENTS WILLIAM WOODY B5M"788/906:"R6;0<:/=4:25"

!!,$+,+*"

"

G3B".4<"62"211/80"L/90"901/6/:/26"125"J52H4HK0"84I<0M"

"

>.0"H2459"I<09"H455/05"464KD</<"9I5/6=":.0/5"0\4?/64:/26"21":.0"-2<82"CI<26M"

"

JIM SCHEFFER 7<<28/4:0"N/508:25"21"R6;0<:/=4:/26"gI4K/:D"

!#,!!,+*"

"

P52?":/?0":2":/?0"L0".4;0"4880<<09"FRBRB":2"=4:.05":5069"94:4"0:8M">.0"EB-F"/<":.0"5082=6/^09"I6/:"125":.0"49?/6/<:54:/26"dEBe":2"R3GM"R6":.0"J4<:":.0"A>BC".4<"<IH?/::09"/:<"/6;0<:/=4:/26<":2"R3G":.52I=.":.0"-FM"G6":.0"2:.05".469"A>BC@G3B".4<"4"5082=6/^09"464KD<:":2"R3G@FRBRB"d?D<0K1eS"50;/0L/6="R3G"?0?H05"<:4:0"488/906:"50J25:M"

"

Page 125: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!#&""

APPENDIX C "

>BC"V;/90680"-.08OK/<:"

"

_V7N"RAjVB>RF7>Gb"A7E>R-7_"-fV-a_RB>"

"

>2"<I880<<1IKKD"8455D"2I:"./<"9I:/0<":.0"_049"R6;0<:/=4:25"A4I:/84K"<.2IK9"826</905":.0"12KK2L/6="8.08OK/<:"/:0?<M""

"

"

FVAVb7_"

"

N0:05?/60"0;/90680"4;4/K4HK0h"

b0;/0L"469"4<<0<<";0<<0K"508259<h"

N28I?06:"0;/90680h"

b082=6/^0"469"464KD<0"?4:05/4K"14/KI50h"

W52:08:"J05/<.4HK0"0;/90680h"

B0/^0"0;/90680"I<01IK":2"/6;0<:/=4:/26h"

B0/^0"25"J.2:282JD"8.45:<S"928I?06:<S"K2=<h""

90:05?/60"50UI/50?06:":2"I<0"</?IK4:25<h"

0;4KI4:0".IKK,06=/60,?48./605D,;45/2I<"<D<:0?<h"

464KD<0S"/6:05J50:"0K08:526/8"94:4"469"<IH?/:":2">04?"_04905h"

<25:"2I:"/6125?4:/26"469"0;/90680"=4:.0509"469"/6125?">04?"_04905h"

"

NG-E3VA>7bk"VjRNVA-V"

Page 126: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!#'""

"

jVBBV_"

"

F06054K"45546=0?06:"JK46h"

84J48/:D"JK46h"

:2664=0"805:/1/84:0h"

8K4<</1/84:/26"<28/0:D"805:/1/84:0<h"5082??0694:/26<","50<05;4:/26<"," 8269/:/26<"21"8K4<<h"

850L"K/<:h"

<:469/6="25905<h"

RB3"928I?06:4:/26h"

62:0"21"J52:0<:h"

?4/6:064680"50J25:<h"

?4/6:064680"508259"J5/25":2"<4/K/6=h"

/6<J08:/26S"50J4/5"469"?4/6:064680"<8.09IK0h"K2=<h"

?4:0t<"5080/J:<"469"49Z2/6/6="8K4I<0<h"

H/KKt<"21"K49/6=h"

8.45:05@J45:Dh"

06=/60"K2=<h"

06=/60"94:4"50825905h"

82I5<0"50825905"J5/6:2I:h"

V-NRB"92L6K249h"

L25O/6="8.45:"L/:."25/=/64K"?45O/6=<h"

549/2"K2="469"549/2"?0<<4=0<h"

28046"52I:0/6="?0<<4=0<h"

9541:"<I5;0Dt<"469"4882?J46D/6="84K8IK4:/26<h"

908O"K2="4H<:548:<"/68KI9/6="K249/6="469"9/<8.45=/6="2J054:/26<h"

Page 127: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!#(""

.2K9"8K046/6="2J054:/26<h"

J50;/2I<";2D4=0"/1".04;D"L04:.05"0682I6:0509h"

;06:/K4:/26"508259<h"

:0?J054:I50"508259<h"

H/K=0"<2I69/6="508259<h"

<:4:0?06:"21"148:<"4:"K249"469"9/<8.45=0"J25:h"

62:/80"21"5049/60<<"4:"K249"469"9/<8.45=0"J25:h"

845=2"?46/10<:h"

845=2"<08I5/6="?46I4KdR3G"8290"21"B410"W548:/80<"125"845=2"<:2L4=0"469"<08I5/6=eh"

<:2L4=0"JK46h"

82550<J2690680h"8.45:0505<h"<./JJ05<h"4=06:<h"<:0;09250<h"<IJ05845=2h"

J.2:2=54J.<"469";/902h"HD"850Lh"HD"<./JJ05h"HD"845=2"2L605<h"

82?JI:05"<:4H/K/:D"J5/6:2I:h"

"

RAcEbk">G"W7BBVAFVbB"7AN,Gb"-bV`"

"

64?0"

49950<<"

94:0"21"H/5:."

60\:"21"O/6"

64:/264K/:D"

54:/6="

488/906:":DJ0S":/?0S"JK480"

804<0"L25O"

/684J48/:4:09":2:4KKD"25"J45:/4KKD"

_284:/26"21"<./J"4:":/?0"21"288I550680"dJ25:"25"<04e"

Page 128: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!#)""

64:I50"21"L25O"

<IJ05;/</26"21"L25O"

4882?J46/09"/6"L25O"JK480"

I6I<I4K":4<O"

4I:.25/^4:/26":2"J05125?"L25O"

L/:60<<":2"488/906:"

50K4:09":2"4"<./JJ/6="488/906:"

?09/84K":504:?06:"26H2459"469"4<.250"

<.250":504:?06:"

60=K/=0680S"?/<8269I8:S"9014IK:"

4K82.2KS"95I=<S":0<:"

06;/526?06:S"K/=.:/6=S"62/<0S"L04:.05S"4880<<"

"

"

BV7b-f"7AN"bVB-EV"

"

@" NP"H045/6="21"9/<:50<<"?0<<4=0":4O06h"

@" F3NBB"9/<:50<<"?0<<4=0<h"

?0<<4=0"50K4Dh"

:/?0"21"?4D94D"84KKh"

:DJ0"21"?4D94D"84KKh"

:/?0"21":4<O/6="50<8I0"I6/:<h"

826:/6I2I<"K/<:06/6="L4:8."26"9/<:50<<"150UI068/0<h"

@" B7b"?46I4K"826<IK:09h"

@" <I51480,4/5"82??I6/84:/26<h"

@" b--"50<J26<0h"

Page 129: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!#*""

4<</=609"I6/:<"d4/5"469"<I51480eh"

@" 82I5<0"469"<J009h"

3090;48h"

"

"

37A"GjVbCG7bN"

"

@" N0K4D"H0:L006"488/906:"469":/?0"21"62:/1/84:/26h"

_/10HI2D"L/:."K/=.:S"1K450"25"<?2O0"</=64K"50K04<09h"

@" 7;2/9/6="48:/26":4O06h"

@" W2</:/26"21"K/10HI2D"4<"<0458."94:I?"62:09h"

@" B./J"?4620I;50h"

@" _22O2I:<"J2<:09":2"O00J"J05<26"/6"</=.:h"

@" $"K26="HK4<:<"<2I6909h"

@" b0<8I0"H24:"850L"4<<0?HK09h"

@" 34<:05"/6125?09h"

@" V6=/60@522?"/6125?09h"

@" j0<<0K"469"?46"2;05H2459"J2</:/26"JK2::09"d346I4KKD"469,25"0K08:526/84KKDeh"

@" b49/2"H524984<:"J05125?09h"

W2</:/26<"IJ@94:09"/6"549/2"522?h"

"

"

7C7ANGA3VA>"

"

AI?H05"21"K/10H24:<h""

Page 130: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!$+""

>DJ0"21"K/10H24:<"d<:056"25"</90"K4I68./6=S"82;0509"25"2J06"826<:5I8:/26S"84J48/:DS"?2:25/^09"[eh"

B:5I8:I54K"8269/:/26"21"K/10H24:<h"

VUI/J?06:"469"J52;/</26h"

>DJ0"469"8269/:/26"21"K/10H24:<"50K04<0"<D<:0?h"

C2459/6="469"K4I68./6="45546=0?06:h"

AI?H05"469">DJ0"21"K/10541:<"h"

-269/:/26"21"K/10541:"d94:0"21"/6<J08:/26S"0UI/J?06:"469"J52;/</26eh"

B:2L4=0"K284:/26"21"K/10541:<h""

AI?H05"469"4880<</H/K/:D"21"BI5;/;4K"<I/:<h"

`4<"850L"J52J05KD"4::/509"dK/10Z48O0:S"L45?"8K2:./6=S"<I5;/;4K"<I/:<"0:8eh"

`4<"850L"J52J05KD":54/609h"

`4<":.0";0<<0K"K/<:09h"

B04"469"L04:.05"8269/:/26h"

_/=.:/6="8269/:/26h"

`4<":.050":/?0":2"J50J450"<I5;/;4K"0UI/J?06:"J5/25":2"4H46926?06:"</=64Kh"

`4<":.0"908/</26":2"4H46926"4JJ52J5/4:09"469":/?0KDh"

7H46926?06:"</=64K<"=/;06"d:DJ0S"01108:/;060<<eh"""

`4<":.050"46D"J52HK0?"9I5/6=":.0"K4I68./6=h"

V11/8/068D"21"J4<<06=05"826:52Kh"

`4<":.050"4"9/<:50<<"84KK"<06:S"HD"L.4:"?046<"469"L4<":.050"4"50JKD":2"/:h"

N/9":.0"VWRbC"469"4<<28/4:09"0UI/J?06:"48:/;4:0h"

>/?0"<J06:"/6"8541:"H01250"50<8I0h"

"

-GAPRAVN"BW7-V"VA>bk"

"

Page 131: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!$!""

WI5J2<0"469":DJ0"21"<J480h"

b04<26"125"06:5Dh"

W50@06:5Dh"

BJ480"<0=50=4:09"HD"HK46O/6="J/J0<S";4K;0<S"0K08"0UI/Jh"

j06:09h"

7:?2<J.050":0<:"5049/6="d2\D=06"#!pS".D952845H26"K0<<"!pS":2\/8"=4<0<"JJ?"?04<I50eh"

BJ480"8K04609h"

BJ480":0<:09h"

N0:05?/609"<410h"

W05?/:"50UI/509S"50UI0<:09S"2H:4/609S"82JD"d-.0?/<:"805:/1/84:0eh"

45546=0?06:"125"826:/6I09"8.08O<h"

826:/6I2I<";06:/6="21"<J480h"

490UI4:0"4880<<h"

490UI4:0"K/=.:/6="d0\JK2</26"J5221eh"

50<8I0"i"50<I<8/:4:/26"0UI/J?06:"4;4/K4HK0h"

50<J26</HK0"J05<26"90</=64:09h"

826<:46:"4::0694680h"

CI99D"<D<:0?"I<09h"

82??I6/84:/26<"0UI/J?06:":0<:09"HD"4KK"826805609h"

0?05=068D"469"0;48I4:/26"J52809I50<"0<:4HK/<.09h"

0UI/J?06:"I<09"2J054:/264Kh"

J05<2660K"J52J05KD"8K2:.09h"

/6<:5I8:/26<"=/;06":2"J05<26<"826805609h"

I6905<:469/6="21"82??I6/84:/26"J52809I50<h"

4=500"26"4"50J25:"/6:05;4Kh"

?I<:";484:0"/6"0;06:"21"/??09/4:0";06:/K4:/26"14/KI50h"

Page 132: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!$#""

14?/K/45"L/:."H504:./6="4JJ454:I<"I<09h"

H504:./6="4JJ454:I<":0<:09h"

H504:./6="4JJ454:I<"=4I=0":0<:09"2J054:/264Kh"

H504:./6="4JJ454:I<"84J48/:D"4/5"<IJJKD"<I11/8/06:h"

H504:./6="4JJ454:I<"1480"?4<O"62:"K04O/6="d8.08O"125"H0459eh"

0?05=068D"</=64K<":0<:09h"

50<8I0".4560<<"26h"

I<0"21"K/10K/60<"L.050"J548:/84HK0h"

J05?/:"</=609"HD"?4<:05S"50<J26</HK0"J05<26S"J05<26"06:05/6=h"

94:0"469":/?0"21"</=64:I50h"

Z2H"82?JK0:09S"94:0"469":/?0h"

<J480"<08I509"4=4/6<:"06:5D"94:0"469":/?0h"

G11/805"21":.0"L4:8."9IKD"/6125?09"94:0"469":/?0h"

82?JK0:/26"21"J05?/:"</=609"HD"50<J26</HK0"J05<26h"

8.08O"125"J05?/:"J05/29";4K/9/:Dh"

"

jGk7FV"W_7AARAF"

"

7WWb7RB7_"

"

R6125?4:/26"50K4:09":2":.0"H4</8"?4620I;5/6="8.4548:05/<:/8<"469"64;/=4:/26"0UI/J?06:"21":.0";0<<0KS"64;/=4:/26"8.45:<S"<4/K/6="9/508:/26<S"HI2D4=0S"64;/=4:/26"?45O<"469"2:.05"4/9<"<I8."4<"b7-GABS"K/=.:<S"K049/6="K/=.:<h"8I5506:"469":/90":4HK0<S"<./J=<"455/;4K"9541:"469"L04:.05h"

"

WbVW7b7>RGA"

;2D4=0"JK466/6=h"

Page 133: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!$$""

N54L"/6:06909":548O"26"8.45:h"

/6<J08:"125"/<2K4:09"946=05<h"

0<:4HK/<."<410:D"?45=/6<h"

954L"=I/90K/60<"125"?4620I;50<"4<"50UI/509h"

90:05?/60"1/64K"4H25:"J2/6:"826</905/6=":I56/6="8/58K0"469"<:2JJ/6="9/<:4680h"

90:05?/60"J2<</HK0"<410:D"48:/26<":2"0?05=068D"</:I4:/26<h"

64;/=4:25<"62:0<"

K049/6="K/=.:<"469"?45O<h"

HI2D4=0"<D<:0?h"

<O0:8."J52J2<09":548Oh"

/<2K4:09"946=05<h"

L.00K"2;05"J2</:/26<h"

O0D"J2</:/26<"125"J454KK0K"/690\/6=h"

8I5506:<"469":/94K"<:504?<h"

H5/01/6="125"H5/9=0":04?h"

:0<:"=045"H01250"90J45:I50h"

8.08O"/6<:5I?06:<h"

O00J"K2=<h"

?4<:05"J/K2:"0\8.46=0"21"/6125?4:/26h"

H5/9=0"0UI/J?06:"14?/K/45/<4:/26h"

H5/9=0"469"908O"K/=.:/6="

0?05=068D"/6"84<0"21"J2L05"14/KI50"

<D68.526/^4:/26"21"H5/9=0"0UI/J?06:"

"

EANVb"`7k","37aRAF"`7k"

"

Page 134: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!$%""

GG`"/<"<2K0"K22O2I:"dB>-`"$M#M!M!e"

BI11/8/06:"50<:"

76:/8/J4:09"L25OK249"4KK2L<"GG`":2"?4/6:4/6"J52J05"K22O2I:"i"50?4/6"/6"1IKK"826:52KhdI6905:4O0"9I:/0<"/6:05105/6="L/:."<410"64;/=4:/26e"

C48O@IJ"4<</<:4680".4<"H006"8K045KD"90</=64:09"

GG`"O62L<"L.2S".2L"469"L.06":2"84KK"125"H48O@IJ"

"

"

3GAR>GbRAF"

"

?26/:25"4:"4KK":/?0<h"

82?JKD"L/:."JK46h"

8.08O";2D4=0"L/:."64;/=4:/264K"4/9<h"

;/<I4K"1/\0<"469"54945"1/\0<h"

I<0"21"26"H2459"64;/=4:/26"0UI/J?06:h"

I<0"21"J454KK0K"/690\/6=h"

@"<410:D"?45=/6<"I</6="92::09"K/60<h"

@"L.00K"2;05"J2</:/26<h"

@"546=0"469"H045/6="/6125?4:/26h"

826:52K"21"<J009"469".049/6=h"

:548O"/6:06909h"

82I5<0"<:00509h"

J4:."12KK2L09h"

05525<"25"2?/<</26"HD"J/K2:h"

48:/26":22"K4:0h"

73jVbS"73P7uh"

Page 135: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!$&""

"

WbVW7b7>RGA"PGb"7bbRj7_"

"

K284K"50=IK4:/26<"

jfP"8.4660K<"<0K08:09"

64;/=4:/26"0UI/J?06:"J50J4509"469"8.08O09"

<:005/6="=045"8.08O09"

06=/60"<L/:8."152?"HI6O05":2"9/0<0K"d/1"4JJK/84HK0e"

J25:"82??I6/84:/26<S"<J08/4K"H05:./6=S"K249/6="50UI/50?06:<"

"

7A-fGb"`7>-f"

"

JK2:"<./J=<"J2</:/26"

I<0"21"4JJ52J5/4:0"8.45:"

150UI068D"21"J2</:/26"8.08O<"

J52J05"K22O@2I:"

/6<J08:/26"52I69"26H2459"

L04:.05"50J25:<"

L04:.05"508259<"

<:4:0"21"<04"

8I5506:"469":/90"

8.08O"125"954==/6="468.25"

5049/60<<"21"06=/60"

0\./H/:"J52J05"K/=.:<"469"<.4J0<"

J52J05"<2I69"</=64K<"

82?JKD"L/:."K284K"50=IK4:/26<"

Page 136: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!$'""

"

bVB>bR->VN"jRBRCR_R>k"

"

64;/=4:/26"0UI/J?06:"1IKKD"2J054:/264K"

jfP"8.4660K<"

12="</=64KK/6="4JJ454:I<"

64;/=4:/26"K/=.:<"

L4:05:/=.:"9225<"

K22O@2I:<"

?4<:05"/6125?09"

B410"<J009"

8.46=0"/6";2D4=0"JK46"

826</905"468.25/6="

"

fV7jk"`V7>fVb"B7R_RAF"

"

?4<:05"469"850L"/6125?09"21".04;D"L04:.05"125084<:"

?2;4HK0"2HZ08:<"<08I509"

<./J=<"4882??294:/26"<08I509"

L04:.05"908O"2J06/6=<"<08I509"

<J009"469"82I5<0"49ZI<:09"488259/6=KD"

850L"L45609"62:"=2"2I:</90"

<410:D"K/60<""

?26/:25/6="L04:.05"50J25:<"

"

BfRWWRAF"7--RNVA>"

Page 137: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!$(""

"

NRB>bVBB"7_Vb>"PbVgEVA-RVB"

"

I<0"21"<0K08:/;0"84KK/6="dNB-e"

jfP"8.4660K"!'"

jfP"(+"

3P"#!)#"Of^"

3P"#!)(M&"Of^"

fP"%#+(M&"Of^"

fP"'$!#"Of^"

)%!%M&"Of^"

!#&(("Of^"

!')+%M&"Of^"

:.0"9/<:50<<"4K05:"<.2IK9"/69/84:0"L./8."150UI068D":.0"12KK2L"IJ"?0<<4=0"L/KK"H0":546<?/::09"469":.0"?290"21":546<?/<</26"d:0K0J.26D"25":0K0\e"

:.0"9/<:50<<"4K05:"HD"<4:0KK/:0"<.2IK9"H0":546<?/::09"L/:."4H<2KI:0"J5/25/:D":2"4"b0<8I0"-2259/64:/26"-06:50"db--e"

"

-G__RBGA"

"

@" L04:.05h"

@" <04"<:4:0h"

@" :/90,8I5506:h"

@" 82I5<0"2L6"<./Jh"

@" <J009"2L6"<./Jh"

@" L.2"90:08:09"2:.05";0<<0Kh"

@" 1/5<:"54945"826:48:h"

Page 138: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!$)""

@" <J009h"

@" 82I5<0h"

@" :/?0h"

@" 9/<:4680h"

@" JK2::/6=h"

546=0"/6"I<0"

b4945"?290"/6"I<0S">5I0"32:/26,"B./J=<".049"IJ"

6I?H05"21"54945"<0:<"/6"I<0"

@" 7bW7"?0?25D"H46Oh"

@" 1/5<:";/<I4K"826:48:h"

@" L.4:"?046<h"

@" 9/<:4680h"

@" H045/6=h"

@" K/=.:<"469"<.4J0<h"

@" 4JJ4506:"82I5<0h"

@" :/?0h"

@" 1/5<:"549/2"826:48:h"

@" J05<2660K"26"H5/9=0h"

@" 82?J2</:/26"21"L4:8.h"

@" ?4620I;50<":2"?/6/?/<0"82KK/</26"01108:h"

@" <J009"8.46=0h"

@" 82I5<0"8.46=0h"

@" K/=.:"469"<2I69"</=64K<h"

@" L.06"?490h"

@" L.06".0459h"

@" 8.46=0"21"82I5<0"469,25"<J009"21"2:.05";0<<0Kh"

Page 139: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!$*""

@" 82K"50=<"2H0D09h"

@" 0?05=068D"4K45?"<2I6909h"

@" 46=K0"21"/?J48:h"

@" :/?0"21"/?J48:h"

@" 488I548D"21"8K28O<h"

@" J2</:/26"21"/?J48:h"

@" 82I5<0"4:"/?J48:h"

@" <J009"4:"/?J48:h"

@" 8.45:09"82I5<0h"

@" 82I5<0"50825905"=54J./8h"

@" H0KK"H22O"469,25"94:4"K2==05"dV-NRBeh"

@" /6=50<<"21"<04"L4:05h"

@" L4:05:/=.:"469"1/50"9225<"8K2<09h"

@" 908O"K/=.:/6="<L/:8.09h"

78:/26"410:5"/?J48:"

@" 82??I6/84:/26"H01250"469"41:05"/?J48:h"

@" 850L"469"J4<<06=05"?I<:05h"

@" ;0<<0K"J2</:/26h"

@" 549/2"522?"IJ@94:09h"

@" <2I69/6="21"H/K=0"469":46O<h"

@" 9/<:4680"152?"H5/9=0":2"H2Lh"

@" ;0<<0K<"/6";/8/6/:Dh"

@" 82??I6/84:/26"L/:.";0<<0K<"/6";/8/6/:Dh"

@" /690J06906:"L/:60<<0<h"

@" J.2:2=54J.<"469";/902h"

@" <854J"J4J05,82I5<0"84K8IK4:/26<h"

Page 140: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!%+""

"

"

B>bRaRAF"

"

@" 94:0"469":/?0"21"/68/906:h"

@" J2</:/26"469"6I?H05"/906:/1/84:/26"21"J/05h"

@" ;/</H/K/:Dh"

@" <04"<:4:0h"

@" <:4:0"21":/90"469"8I5506:<h"

@" 2K9"25"60L"2HZ08:h"

@" 94?4=0":2";0<<0Kh"

@" 94?4=0":2"2HZ08:h"

@" J2</:/26"21"2HZ08:"/906:/1/09"26"8.45:,62:<./Jh"

@" J.2:2=54J.<"469";/902h"

@" 1/\09,1K24:/6="2HZ08:h"

@" <410:D"21"J25:d/6125?4:/26"469"J/K2:"<05;/80eh"

@" 06;/526?06:4K"/6:050<:<h"

"

"

FbGEANRAF"

"

@" K4<:"O62L6"488I54:0"J2</:/26h"

@" ?0:.29"21"J2</:/26"1/\/6=h"

@" :/?0"21"/?J48:h"

@" 488I548D"21"8K28O<h"

@" J2</:/26"21"/?J48:h"

Page 141: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!%!""

@" 82I5<0"4:"/?J48:h"

@" =D52,?4=60:/8"82?J4<<"05525"

<J009"4:"/?J48:h"

@" 8.45:09"82I5<0h"

@" 82I5<0"50825905"=54J./8h"

@" H0KK"H22O"469,25"94:4"K2==05"dV-NRBeh"

@" /6=50<<"21"<04"L4:05h"

@" L4:05:/=.:"469"1/50"9225<"8K2<09h"

@" 9541:h"

@" :/90"J509/8:/26h"

@" 8I5506:"/6125?4:/26h"

@" 54945"J05125?4680h"

546=0"/6"I<0h"

@" ;/</H/K/:Dh"

@" 06=/60"<:2JJ09h"

@" 0?05=068D"4K45?"<2I6909h"

@" L4:05:/=.:"469"1/50"9225<h"

@" 06=/60@522?"/6125?09h"

@" K/=.:<"469"<.4J0<"0\./H/:09h"

@" J50@<:5469/6="9541:h"

BUI4:h"

@" H/K=0<"469":46O<"<2I6909h"

@" .2==/6="25"<4==/6=h"

@" 2;05</90"<2I69/6=<h"

@" ;0<<0K"J2</:/26h"

@" <J009"509I8:/26h"

Page 142: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!%#""

@" :5411/8h"

@" .049/6="41:05"=52I69/6=h"

@" 94?4=0"50J25:<h"

@" ?08.46/84K"H504O92L6<h"

@" 549/2"82??I6/84:/26<"L/:."J4<</6=":5411/8h"

@" 468.25<h"

@" ;0<<0K"50<J26<0h"

@" :I="4<</<:4680h"

@" 94:0"469":/?0"21"4<</<:4680h"

@" 8.46=0"21"<:005/6="?290h"

@" 6I?H05"21"JI?J<"26h"

@" 64:I50"21"H2::2?h"

"

"

P_GGNRAF"

"

@" 0?05=068D"4K45?h"

@" 0?05=068D"<:4:/26<h"

@" L4:05:/=.:"9225<"8K2<09h"

@" ?4<:05"/6125?09h"

@" 06=/60@522?"/6125?09h"

N4?4=0"<:4H/K/:Dh"

@" ;0<<0K"J2</:/26"IJ@94:09":2"549/2"522?h"

@" K/10"H24:,K/10"541:"?I<:05"4:"<:4:/26<h"

"

"

Page 143: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!%$""

PGEANVbRAF"

"

@" L04:.05"8269/:/26<h"

@" <./1:/6="21"845=2h"

@" H504O/6="21"HIKOh"

@" <:5I8:I54K"90108:<h"

@" 0\JK2</26h"

@" K/UI0148:/26"21"HIKO"845=2h"

@" <:4H/K/:Dh"

@" L4:05"H4KK4<:h"

@" ;0<<0K"./<:25Dh"

@" K4<:"L/:60<<0<h"

@" K4<:"<:0;09250<h"

@" K4<:"826:48:"L/:."2L605<h"

@" 94:0"469":/?0"21"K4<:"J25:"90J45:09h"

@" ;2D4=0"JK46h"

@" J25:"21"84KKh"

@" V>7h"

@" :DJ0"21"845=2h"

@" K4<:"549/2"82??I6/84:/26<h"

@" 2L605<h"

@" J4<</6=";0<<0K<h"

@" J50;/2I<"?08.46/84K"H504O92L6<S"0K08:5/84K"14/KI50<"469"0UI/J?06:"?4K1I68:/26h"

"

3V-f7AR-7_"P7R_EbVB"

<:469459"J52809I50<"/906:/1D/6="850L"9I:/0<"469"50<J26</H/K/:/0<h"

Page 144: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!%%""

125"64552L"8.4660K"64;/=4:/26"

125"J52JIK</26"0?05=068/0<"

125"<:005/6="=045"14/KI50"

125"0K08:5/84K"<D<:0?"14/KI50"

:2"J05125?"0?05=068D"468.25"90JK2D?06:"

:2"8269I8:"1I0K"<L/:8./6="

K4<:"95/KK"

J52809I50"125"850L/6="/6"625?4KKD"I64::06909"?48./605D"<J480<"

;2/80"82??I6/84:/26"H0:L006"H5/9=0"469"06=/60"826:52K"<:4:/26,0?05=068D"<:005/6="<:4:/26,468.25"<:4:/26h"

J50;06:4:/;0"?4/6:064680"508259<"

IJ":2"94:0"?46I148:I505=<":08.6/84K"JIHK/84:/26<"

<I11/8/06:":22K<S"<J450"J45:<""

J50;/2I<"469,25"J50<06:"?08.46/84K"14/KI50<"50J25:09":2"j>B","J/K2:","8.46=0"21"850L"

468.25<"5049D"125"/??09/4:0"50K04<0"

J2L05"/??09/4:0KD"4;4/K4HK0"125"468.25"50K04<0"

"

37RA"VAFRAV"P7R_EbV" "

"

@" ?4<:05"/6125?09h"

@" 5I9905"469"H2L":.5I<:05<"I<09":2"H0<:"49;46:4=0h"

@" J50J454:/26":2"468.25/6=h"

@" 62:"I6905"82??469"<.4J0<"469"K/=.:<h"

@" HI6O05"UI4K/:D"26"H2459,<.250"K4H":0<:<h"

@" <J08/1/8"=54;/:Dh"

@" ;/<82</:Dh"

Page 145: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!%&""

@" 82?J4:/H/K/:Dh"

@" L4:05"826:06:h"

@" ;0<<0K"J2</:/26h"

@" 946=05"21"=52I69/6=h"

@" L/69"469"8I5506:"9/508:/26h"

@" 468.254=0h"

@" 6009"125":2L4=0h""

@" J25:"21"501I=0h"

"

"

B>VVbRAF"P7R_EbV"

"

@" 06=/60"522?"/6125?09h"

@" 0?05=068D"<:005/6="06=4=09h"

@" 62:"I6905"82??469"K/=.:<"469"<.4J0<h"

@" 4JJ52J5/4:0"<2I69"</=64K<h"

@" L4D":4O06"211"<./Jh"

J5/25":0<:/6="21"?46I4K"<:005/6="=045"

?IK:/JK0"J2L05"I6/:<"

I<0"21":L2"25"?250"J2L05"I6/:<"</?IK:4602I<KD"

K4<:"?4/6"<:005/6="8.08O"

K4<:"4I\/K/45D"<:005/6="8.08O"

K4<:"50?2:0"<:005/6="8.08O"

K4<:"8.08O"J2L05"<IJJKD"

K4<:"852<<"8.08O"21"5I9905"/69/84:25";<"48:I4K"5I9905"?2;0?06:"

K4<:"4K45?"8.08O"

Page 146: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!%'""

/<2K4:/6="45546=0?06:<"

?4\/?I?"1IKK"5I9905"46=K0"

:/?/6="21"5I9905"?2;0?06:"

82??I6/84:/26<"H0:L006"H5/9=0"469"<:005/6="1K4:"

8.46=0"2;05"J52809I50<"

0?05=068D"<:005/6="95/KK<"

508259<"21"<:005/6="8.08O<"469"95/KK<"

"

RAB>bE3VA>"P7R_EbV"

"

FkbG,-G3W7BB"P7R_EbV"

"

@" 4K:0564:/;0"?046<h"

@" ?4=60:/8"82?J4<<"I<09"125"<:005/6=h"

@" ?4<:05"/6125?09h"

@" =D52"?4/6:064680h"

@" 06=/60"522?"/6125?09h"

@" 14/KI50"21"2:.05"64;/=4:/264K"0UI/J?06:h"

"

CbRNFV"-GA>bG_,>V_VFb7Wf"P7R_EbV"

"

@" <L/:8.":2"06=/60"522?"826:52Kh"

@" 06=/60"522?"/6125?09h"

@" ?4<:05"/6125?09h"

@" H5/9=0,06=/60@522?"82??I6/84:/26<h"

"

Page 147: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!%(""

FWB"P7R_EbV"

"

@90:05?/60":DJ0"21"5080/;05h"

@</?IK:4602I<"d:548O"12I5"25"?250"<4:0KK/:0<"4:"4":/?0eh"

@?IK:/"8.4660K"5080/;05h"

@<0UI06:/4K"d?04<I50"260"<4:0KK/:0"4:"4":/?0eh"

@</6=K0"8.4660K"5080/;05h"

@6I?H05"25"<4:0KK/:0<"?04<I509h"

@6I?H05"21"8.4660K<":.0"5080/;05".4<h"

@76:0664"826608:/26"

@76:0664"K284:/26""

@82?J450"94:4"L/:."2:.05"0UI/J?06:h"

@5080/;05"8K28O"?4K1I68:/26h"

@J2L05"<2I580"50=IK45h"

@:0?J054:I50"41108:<"50K/4H/K/:Dh"

@90:05?/60"94:4"<0:":2"H0"9/<JK4D09"26"?26/:25h"

@8.08O"J2</:/26"469";0K28/:D"L.06";0<<0K"<:4:/2645Dh"

@=229"5080/;05"L/KK"8.46=0"5049/6=";05D"K/::K0h"

@J225"5080/;0"L/KK"1KI8:I4:0"4KK"2;05":.0"JK480h"

@9I4K"150UI068D"5080/;05h"

@;05D"<2J./<:/84:09h"

@0K/?/64:0"/6262<J.05/8"05525h"

@<0UI06:/4Kh"

@K0<<"488I54:0h"

@</6=K0"8.4660Kh"

@<:45;09@J2L05h"

Page 148: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!%)""

@J25:4HK0h"

@5I6"<?4KK"H4::05/0<h"

@J2</:/26"5049/6="26KD"2680"25":L/80"J05"?/6h"

@?250"488I54:0":.46"K2546h"

@;0K28/:D"5049/6="UI0<:/264HK0h"

@826:/6I2I<"5I66/6=h"

@62:"K/?/:09"HD"J2L05"<2I580h"

@846"?04<I50";0K28/:D"/1"62"4880K054:/26"25"82I5<0"8.46=0h"

@92"62:"?26/:25"J2</:/26"826:/6I2I<KDh"

@84662:":5I<:";0K28/:D"/1"8.04J"5080/;05"8K28Oh"

@14<:"?IK:/JK0\/6=h"

@?2;0<"152?"<4:0KK/:0":2"60\:"14<:05h"

@5I6<"826:/6I2I<KDh"

@:L2"8.4660Kh"

@92IHK0<":.0"<D<:0?t<"</=64K@:2@62/<0"54:/2h"

@846"K28O"26:2"<4:0KK/:0"I6905"?250"49;05<0"8269/:/26<h"

@:548O"<4:0KK/:0<"8K2<05":2".25/^26h"

@;0K28/:D"?04<I50?06:<"?250"488I54:0h"

@L/KK"I<0"?250"J2L05":.46"</6=K0"8.4660Kh"

@</?IK:4602I<h"

@L/KK"?26/:25"12I5"25"?250"<4:0KK/:0<"4:"4":/?0h"

@/6<:46:4602I<"J2</:/26"469";0K28/:Dh"

@I<09"/6"./=."9D64?/8"469,25"488I548D"4JJK/84:/26<h"

@I<09"125"<I5;0D/6="25"<8/06:/1/8"JI5J2<0<h"

@:548O"4KK"<4:0KK/:0<"/6";/0L"/6<:049"21"84K8IK4:/6="J2</:/26h"

@12I5":/?0<"</=64K@:2@62/<0"21"</6=K0"8.4660Kh"

Page 149: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!%*""

@K45=0"J2L05"826<IJ:/26h"

"

@:.500"8.4660K"5080/;05"82?J52?/<0h"

@260"8.4660K":2"?04<I50":/?0h"

@:L2"8.4660K<":2"0<:4HK/<."549/2"K28O"26"60\:"<4:0KK/:0h"

@B,7"<0K08:/;0"4;4/K4H/K/:D"90:05?/609"HD"N0J:"21"N010680h" " "

"

b7N7b"P7R_EbV"

"

J2</:/26"26"H5/9=0"

14?/K/45"?290K"

b4945"150UI068/0<"

826<IK:09"2J054:25=<"?46I4K"

125?05"0\J05/0680"L/:."54945"

BVA"805:/1/84:/26"

7bW7"805:/1/84:/26"

8.08O"21"4KK"826:52K<"

?4K1I68:/26"21"826:52K<"

:I6/6="9/11/8IK:D"

H5/KK/4680,=4/6"49ZI<:?06:"

J50<06:4:/26"50<:5/8:/26"

.4H/:":2"I<0"L./8."J50<06:4:/26"

=D52"?4K1I68:/26"

9/<:25:/26"4:"UI4D"</90"

0K08:526/8"H045/6="K/60":22":./8O""

?IK:/JK0"08.2"

Page 150: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!&+""

</90"K2H0"01108:"

9/<:25:/26"/6"260"UI49546:"

9/11/8IK:D"/6"/906:/1D/6="K469?45O<"L/:."54<:05"

82?J450">\,b\"469"?26/:25<"

.049/6="?45O05"05525"L/:."K469?45O","4:.L45:<./J<"

J454KK4\"

?IK:/JK0";08:25"26"</6=K0":45=0:"

45J4"?4J"J50<06:4:/26"488I548D"

54945"05525"125=2::06":2"4JJKD"

?0:4K"84K8IK4:/26"1I5:.05":2"82I5<0"4K:054:/26"

54945"05525"50J25:09"HD"850L"

I6905<:469"850L"54945"05525"/6:05J50:4:/26"

82?J45/<26"J/K2:"469"850L"54945"05525"

82?J45/<26"21"05525"H01250"469"41:05"/?J48:"

05525"26"9/110506:"<84K0<"

405/4K"</^0"

405/4K"K284:/26"

64:I50"21"54945"H504O92L6"

0K08:5/84K"J2L05"14/KI50"

54945":08.6/8/46"26"H2459,<.250"

H5/9=0"54945"K2=""

"

V-NRB"

"

:54/6/6="<8.22K"

4JJ52;09"/6<:5I8:25"

Page 151: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!&!""

0\J05/0680"/6"=06054K"

14?/K/45/:D"L/:."?290K"4H2459"

W-"JK4:125?"25"8K2<09"<D<:0?"

W-"JK4:125?"d2:.05"I<0e"

W-"J05125?4680"

<21:L450"K/80680"

K/80680"0\J/5D"94:0"

<05;05<"/6"I<05h"

64;"0UI/J"K/6Oh"FWBS"NFWBS"b4945S"7bW7S"=D52h"

J50;/2I<"?4K1I68:/26"

508259<h"K2=h"

CK48OH2\S";2D4=0"./<:25Dh"

;0<<0K"469";2D4=0"06:5/0<"

?29/1/84:/26"21"06:5/0<"

8.2/80"21"?4J<"

B&(S"B04?4J<S"CBCS"7b-"

?46I4K"825508:/26"<4;09""

50/6<05:"?46I4K"825508:/26<"41:05"<.I:"92L6"

64;/=4:25<"2L6"<0K08:09"2J:/26<"

<85006"K/=.:"/6:06</:D"

62:0":/?0"154?0"H0:L006"488/906:"469"92L6K249"d?4\"(#".2I5<e"

"

A7jR"B7R_Gb"@">b7AB7B"

"

8261/=","0\/:"@"<.I:"92L6">546<4<"

"

Page 152: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!&#""

B7R_"B7PV"@"gM37b"

"

PRbV"

"

NRB-GjVbk"

"

@" L.2"9/<82;0509"1/50h"

@" JK480"469":/?0"9/<82;05D"21"1/50h"

@" 148:25"L./8."950L"4::06:/26":2"1/50h"

@" 504<26"125"J50<0680"6045"1/50h"

@" J50<0680"21"2:.05"J02JK0"H01250"1/50h"

@" 2H<05;0"<2?0260"K04;0"</:0"H01250":.0"1/50h"

@" K4<:"J05<26"O62L6":2".4;0";/</:09":.0"82?J45:?06:h"

@" L/69h"

@" J52=50<<"21"1/50h"

@" 0;/90680"21"46"4880K0546:h"

@" 82K2I5"21"1K4?0h"

@" 82K2I5"21"<?2O0h"

@" <?0KK"21"1/50h"

@" G5/=/6"469".04:"<2I580M"

@" .045"0\JK2</26h"

@" 125809"06:5D"/6:2"82?J45:?06:"d7bBGAeh"

"

7_7b3"

"

@" L.2"<2I6909"4K45?h"

Page 153: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!&$""

@" 84KK"2I:"v1/50"26"H2459vh"

@" 2J054:0"60450<:"1/50"4K45?h"

@" /6125?"850L"?0?H05h"

@" :0K0J.260"64;/=4:/26"H5/9=0h"

@" 0?05=068D"4K45?"<2I6909h"

@" ?4<:05"/6125?09h"

@" 06=/60@522?"/6125?09h"

@" W7"4662I680?06:"50"1/50h"

@" /906:/1D"D2I5<0K1"/6"0?05=068D"?0<<4=0h"

@" <:2J"K249/6=,9/<8.45=/6="2J054:/26h"

@" :/?0"1/50"J45:D"455/;09h"

PRbV"PRFf>RAF"GWVb7>RGA"

"

@" 850L"?0?H05"/6"8.45=0"21"1/50"J45:Dh"

@" 0UI/J?06:"26".469h"

@" 0UI/J?06:"I<09h"

@" K/10"H24:,K/10"541:"?I<:05"<:4:/26<h"

C24:"K2L0509":2"0?H45O4:/26"908O"K0;0Kh"

@" 850L"5048."0?05=068D"<:4:/26<h"

@" J4<<06=05<"50UI0<:09":2"12KK2L"0<84J0"52I:0":2"K/10H24:<h"

@" 8.08O"62"260"?/<</6=h"

@" 8.08O"?46"2;05H2459h"

@" O00J"8./K9506"26"J05?4606:"2H<05;4:/26h"

@" <0K08:"J52J05"K/10"Z48O0:<"125"8./K9506"469"J4<<06=05<h"

@" 8.08O"8K2:./6="21"J4<<06=05<h"

@" 4KK"826805609"62:/1/09"21"<04:"21"1/50h"

Page 154: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!&%""

@" ;06:/K4:/26h"

@" 4I:2?4:/8"1/50"9225<h"

@" 1/50"9225<"Z4??09"/6"2J06"J2</:/26h"

@" L4:05:/=.:"9225<"8K2<09h"

@" 908O"K/=.:/6="<L/:8.09"26h"

@" ;0<<0K"J2</:/26":2"549/2"522?h"

@" 48:/;4:0"UI/8O"8K2</6="<D<:0?h"

@" 50?2:0";4K;0"<.I:@211"<D<:0?h"

@" 8K2<0"4KK"?46I4K"2J054:09";4K;0<"26"_WF@<D<:0?h"

@" <L/:8."211"4KK"0K08:5/84K"<IJJKDh"

@" <:2J"4KK";06:/K4:/26":2"82?J50<<25"469"?2:25"522?h"

@" 1/50"H2\"6045HDh"

@" 1/50"0\:/6=I/<.05"26".469h"

@" 1/50".2<0h"

@" I6/;05<4K"826608:/26h"

@" 0?05=068D"1/50"JI?Jh"

@" /<2K4:0"<J480"469"<2I580"21"1I0K"

@" 826:48:"1/50"<.250"H5/=490h"

@" JK46"/6/:/4K"4::48Oh"

@" 48:/;4:0".4K26,-G#"1/50"1/=.:/6="<D<:0?h"

@" <:45:"L4:05"/55/=4:/26"/6"1526:"21"H5/9=0h"

@" 822K"92L6"845=2"4504"L/:."L4:05"<J54Dh"

@" 4<</<:"1/50"<.250"H5/=490h"

@" <:469@HD"06=/60h"

@" I<0"21"1/50":I=<h"

@" :/?0":2"0\:/6=I/<.h"

Page 155: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!&&""

@" <:45:"L4:05"/55/=4:/26"<D<:0?"/6"1526:"21"H5/9=0h"

@" <:45:"L4:05"/55/=4:/26"2;05":46O"4504h"

"

BR>V"BEbjVk"

"

@" 90:05?/60"L.2".49"4880<<":2"</:0"H01250"/6<J08:/26h"

@" 90:05?/60"/1";2KI6:45D"25"/6;2KI6:45Dh"

@" 90:05?/60"H4</8"0K0?06:<h"

@" 1K4??4HK0"?4::05"d1I0Keh"

@" /=6/:/26"<2I580"d.04:eh"

@" <2I580"21"2\D=06h"

@" /6<:4KK4:/26h"

@" 0K08:5/8"d<J45Oeh"

@" 8.0?/84K"d5048:/26eh"

@" ?08.46/84K"d15/8:/26eh"

@" 8.08O"125"2I:"21"JK480"2HZ08:<h"

@" :5480"25/=/6"21"1/50h"

@" <0458."125"?2<:"94?4=09"4504h"

@" <0458."125"K2L0<:"</:0"21"82?HI<:/26h"

@" 8.08O"125"90J:."21"845H26/<4:/26"/6"1K4??4HK0"?4:05/4Kh"

@" 8.08O"/1"<2I580".4<"25/=/6"2I:</90";0<<0Kh"

@" 8.08O"2J06/6=<"L./8.".4;0"8504:09"4/5"9541:h"

@" L4<":.0"82?J45:?06:"2J0609"25"8K2<09h"

@" JI:"8/=450::0"/6"4JJ52J5/4:0"826:4/605h"

@" I<0"21"64O09"K/=.:h"

@" I<0"21"8469K0h"

Page 156: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!&'""

@" .46="2HZ08:"6045"0K08:5/8"HIKHh"

@" 0K08:5/8"/526,2:.05"=49=0:"/6"84H/6h"

@" <0458."125":/?05<h"

@" 0K08:5/8"?2:25h"

@" L/5/6="<.2IK9"62:"HI56"I6K0<<"?2:25"82?JK0:0KD"HI5609h"

@" /6:0564K"HI56/6="/1"H526^0"H045/6=<"<0/^09"26"<.41:h"

@" /1"?2:25"5I66/6="9I5/6="1/50S"J45:<"?0K:09"/6</90h"

@" 0K08:5/84K"J460Kh"

@" 8.08O"J52J05"1I<0<h"

@" <.25:"8/58I/:"L/KK"HI56"06:/50KD":.0"1I<0h"

@" 2;05K249"25"J45:/4K"<.25:"8/58I/:"L/KK"?0K:"26KD":.0"1I<0"H469h"

@" /6<IK4:/6="?4:05/4K"26"L/5/6="L/KK".4;0"90:05/254:09"26"H2:."</90<"21":.0"<.25:"8/58I/:h"

@" 90:08:"1/50,<?0KK,<?2O0h"

@" =4<"K04O"L/KK"84I<0"0\JK2</26h"

@" 8.08O"J/J/6=h"

@" 8.08O"/64JJ52J5/4:0"K284:/26"21"=4<"4JJ454:I<h"

"

"

RANR-7>RGA"GP"7bBGA""

"

@" :54/K05<"H0:L006"<2I580<"21"1/50h"

@" 8469K0<h"

@" ?4:8.0<"L54JJ09"452I69"1K4??4HK0"?4:05/4Kh"

@" 8.0?/84K<"2I:"21"JK480h"

@" 508/J/06:"L./8."826:4/609":.0"4880K0546:h"

Page 157: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!&(""

@" /1"K/UI/9"4880K0546:"I<09S":5480<"L/KK"H0"12I69"/6"Z2/6:<"469"825605<"L.050"K/UI/9"1K2L09h"

@" L4D"/:"L4<"908K4509h"

@" <J46"21":/?0"H01250"908K45/6="1/50h"

@" /6<I54680"J2K/8Dh"

@" 1/6468/6=";0<<0Kh"

@" 50JK48/6=";0<<0Kh"

@" ;0<<0K"125"<4K0h"

@" 9/<<06</26"H0:L006"850L,2J054:25<,2L605<h"

@" 850L"?0?H05"50806:KD"1/509h"

@" 4H<0680"21"J05<2664K"01108:<"/6"84H/6h"

1IKKD"8K2:.09"850L"?0?H05<".4K1"L4D":.52I=.":.0"6/=.:h"

@" ?/<</6="25"626@2J054:/264K"1/50"1/=.:/6="0UI/J?06:h"

"

fRB>Gbk"

"

@" 8455D"2I:"1/50"J4:52Kh"

@" 0\JK2</?0:05<"5049/6=<h"

@" =4<"1500/6="21":46O<h"

@" 8261/5?"1/50"9225<"8K2<09h"

@" 8261/5?"9225<"4880<<":2"82?J46/26L4D<"8K2<09h"

@" K4<:".2I<0O00J/6=h"

@" :46O"8K046/6=h"

@" <?2O0"/6"H09h"

@" 0\:/6=I/<."8/=450::0h"

@" 8.46=0"/6<:4KK4:/26<"50806:KDh"

Page 158: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!&)""

@" ?29/1D";0<<0K"50806:KDh"

@" "

VjRNVA-V"

"

@" :4O0"J.2:2=54J.<"H01250"50?2;/6="2HZ08:<"125"1I:I50"464KD</<h"

@" :4O0"62:0"21"4KK"J05<26<"?46/JIK4:/6="0;/90680h"

"

"

-7bFG"GWVb7>RGAB"

"

_G7NRAF"

"

@" J50;/2I<"845=2h"

@" L04:.05"0682I6:0509h"

@" H4KK4<:"9/<:5/HI:/26h"

@" 8269/:/26"21".2K9<"J5/25":2"K249h"

@" H/K=0"<I8:/26"8.08O09h"

@" H4KK4<:":46O<"J50<<09"IJh"

@" -45=2"B08I5/6="346I4K"d!"cIKD"!**'eh"

@" 1/\09"<08I5/6="45546=0?06:<h"

@" 90<85/J:/26"dJ49@0D0S"0D0H2K:<S"0:8eh"

@" K284:/26h"

@" J25:4HK0"<08I5/6="=045h"

@" K284:/26h"

@" <:2L4=0h"

@" /6;06:25D"469"H504O/6="<:506=:.h"

Page 159: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!&*""

@" 0\4?JK0<"21"825508:"I<0"21"=045h"

@" K4<./6=":4O/6="/6":2"826</9054:/26";0<<0K"?2;0?06:h"

@" 64?0"K249"J25:h"

@" 94:0"469":/?0"21"455/;4Kh"

@" K249/6="H05:.h"

@" 845=2":DJ0<h"

8.0?/84K"469"J.D</84K"J52J05:/0<h"

<410"826:4/6?06:"21"845=2"?04<I50<h"

J52:08:/;0"8K2:./6=h"

K249/6=":0?J054:I50h""

82I6:05"?04<I50<"4=4/6<:"488/906:4K"J05<264K"826:48:h"

@" /6<:5I8:/26<"H01250"K249/6=h"

845=2"<:2L4=0"JK46h"

@" ?0:.29"21"K249/6=h"

@" <./Jt<"0UI/J?06:h"

@" <.250"0UI/J?06:h"

@" =54H<h"

@" 0K0;4:25h"

@" K249/6="<0UI0680h"

@" 94:0"469":/?0"21"K249/6=h"

@" UI46:/:D"K24909h"

@" <:2JJ4=0<h"

845=2"L4:05"<J54D"<D<:0?h"

@" ?45O/6="21"845=2"dL.050S"L.06"469"L.2eh"

@" H504O"4L4D"152?"?225/6=<h"

@" <:2J"K249/6=h"

Page 160: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!'+""

@" 0?05=068D"J52809I50<h"

@" <:469@HD"4:"1/50"<:4:/26"60\:":2"845=2"852<<2;05h"

@" <:469@HD":2"<:45:"L4:05"/55/=4:/26"/1"<J/KK4=0h"

@" UI/8O"8K2</6=h"

@" :5D":2"<08I50"?225/6=<h"

@" <:469@HD":2"952J"468.25h"

@" <:469@HD"06=/60<h"

@" 84KK"<:469@HD":I=<h"

J52809I50"125"845=2":546<105h"

@" 2;05K249/6=h"

@" UI/8O"8K2</6=h"

@" 0?05=068D"J52809I50<h"

@" <:2J"4KK"06=/60<"469"L25Oh"

@" 8.08O".2<0"82IJK/6=<h"

@" 8.08O"JI?J/6="54:0h"

@" 150UI068D"21"IKK4=0"8.08O<h"

"

_7BfRAFS"B>G`7FV"i">bR33RAF"

"

@" <J08/1/8"/6<:5I8:/26<h"

@" /6;2K;0?06:"21"<.250"K4H2I5h"

@" ?0:.29"21"845=2".469K/6=h"

@" 9I664=0h"

@" <IJJK/05h"

@" 64:I50h"

@" L.2"8455/09"2I:":5/??/6="2J054:/26<h"

Page 161: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!'!""

@" L.2"8455/09"2I:"K4<./6="2J054:/26<h"

@" 6I?H05"469"9/?06</26"21"K4<./6="L/50<h"

@" 9541:"<I5;0Dh"

@" H/KKt<"21"K49/6="826</<:06:"L/:."?4:0t<"5080/J:h"

@" 90:4/K<"21"8K2</6="21".4:8.0<h"

@" L.2h"

@" 46D"J52HK0?<"0682I6:0509h"

@" J.2:2=54J."469";/902"21"94?4=09"0UI/J?06:h"

@" K4<./6="I6905":/?0"J50<<I50":2"=0:"I6905L4Dh"

@" /6490UI4:0"6I?H05"21"L/50"K4<./6=<h"

@" ?/<?4:8.09"469"/?J52;/<09"K4<./6="=045h"

@" I<0"21"94?4=09"K4<./6="=045h"

@" HK28O/6="469"H548/6=h"

@" :22"?46D";45/0:/0<"21"<08I5/6="=045"26"H2459"

"

-GA>7RAVbB"

"

@" 1K4:"<.20"JK4:0"d908O"i".4:8."82;05<eh"

@" 8260<"/6"<.20"JK4:0<h"

@" H2::2?":/05"26"50<:/6="26"8260<h"

@" :L/<:"K28O<"H0:L006"048.":/05h"

@" 9/4=264KKD"1/::09"K4<./6=<h"

@" N@5/6=<"26"908Oh"

@" J06=I/6".22O"26"826:4/605<h"

@" N@5/6=<,:I56HI8OK0<,L/50"K4<./6=<,J06UI/6".22O<h"

"

Page 162: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!'#""

EANVb`7k"

"

@" 94:0"21"<4/K/6=h"

@" <J009h"

@" 82I5<0"/6:06909"469"12KK2L09h"

@" J05/29<"21".04;D"L04:.05h"

@" ?0:.29"21"4<<0<<?06:"21"L/69"<J009h"

@" L04:.05"94?4=0":2"908O"1/::/6=<"469"0UI/J?06:h"

@" I<0"21"28046"52I:0/6="25"2:.05"</?/K45"<05;/80h"

@" 549/2"826:48:"L/:."2:.05";0<<0K<"d64?0<eh"

@" H4KK4<:"9/<:5/HI:/26"469"8.46=0<"9I5/6="<4/K/6=h"

@" .4:8.0<"2J0609"4:"<04h"

@" 5049/6=<h"

@" 845=2":0?J054:I50<h"

@" H/K=0"<2I69/6=<h"

@" <04L4:05":0?J054:I50<h"

@" 4/5":0?J054:I50<h"

@" ?046<":2"82I6:0548:"<D68.5262I<"52KK/6=h"

@" 4<<0<<?06:"21"<:25?h"

@" :4O0"045KD"48:/26"/6"90:05/254:/6="L04:.05h"

"

NRB-f7bFRAF""

"

@"" 64?0"J25:h"

@" 94:0"469":/?0"21"455/;4Kh"

@" 9541:"<I5;0D"26"455/;4Kh"

Page 163: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!'$""

@" 64?0"469":/?0"21"H05:.h"

@" /6<:5I8:/26<"=/;06"8268056/6="9/<8.45=/6=h"

@" 9/<8.45=0"0UI/J?06:"d<./J"25"<.250eh"

@" 94:0":/?0"21"9/<8.45=/6=h"

@" UI46:/:D"9/<8.45=09h"

@" ?0:.29"21"845=2".469K/6=h"

@" J52HK0?<"9I5/6="9/<8.45=/6=h"

@" H4KK4<:"JI?J/6="<0UI0680h"

@" /6;2K;0?06:"21"<.250"K4H2I5h"

@" .2<0"HI5<:S"1548:I509"J/J0h"

@" 48:/;4:0"UI/8O"8K2</6="<D<:0?h"

@" 0?05=068D"J52809I50<h"

@" <:45:"1/50"4K45?h"

@" <:45:"L4:05"/55/=4:/26"/1"<J/KK4=0"26"908Oh"

@" 1/50@1/=.:/6="0UI/J?06:"469"H504:./6="4JJ454:I<"<:469@HDh"

@" J50J450"0?05=068D"J45:Dh"

@" <:45:"1/50"JI?Jh"

@" <:2J"4KK"06=/60<"469"L25O"/1"6080<<45Dh"

@" H504O"L4D"?225/6=<h"

@" <:2J"K249/6=h"

@" 0?05=068D"J52809I50<h"

@" <:469@HD"4:"1/50"<:4:/26"60\:":2"845=2"852<<2;05h"

@" <:469@HD":2"<:45:"L4:05"/55/=4:/26"/1"<J/KK4=0h"

@" UI/8O"8K2</6=h"

@" :5D":2"<08I50"?225/6=<h"

@" <:469@HD":2"952J"468.25h"

Page 164: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!'%""

@" <:469@HD"06=/60<h"

@" 84KK"<:469@HD":I=<h"

@" :/94K"9/508:/26"469"<:506=:.h"

"

_GBBS"BfGb>7FV"Gb"N737FV"GP"-7bFG"

"

@" 1/5<:"50J25:"21"K2<<S"94?4=0"25"<.25:4=0h"

@" L.2"50J25:09h"

@" Z2/6:"/6<J08:/26h"

@" J45:/0<"/6;2K;09"/6"/6<J08:/26h"

@" L.050"845=2"<:2509h"

@" <0=50=4:/26"21"94?4=0"845=2"152?"=229"845=2h"

@" 4H4692609"845=2"dZ0::/<2609eh"

@" L04:.05"8269/:/26<h"

@" J48O/6="901/8/068Dh"

@" 0UI/J?06:"14/KI50h"

@" :DJ0"21"9I664=0h"

@" ?0:.29"21"<:2L/6="469"K4<./6=h"

@" 94?4=0"50J25:h"

@" ?4<:05h"

@" 8./01"06=/6005h"

@" 2:.05"J05<2660K"/6;2K;09h"

@" J.2:2=54J."469";/902"21"94?4=0"845=2h"

@" K2="H22O<"469"508259<h"

@" 52I:0/6="8.46=0"/?J2<09h"

@" J25:"21"501I=0h"

Page 165: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!'&""

@" <.250"4<</<:4680"J52;/909h"

@" J25:"4I:.25/:D"J52;/</26<h"

"

"

PRBfRAF"jVBBV_"GWVb7>RGAB"

"

14:/=I0h""

AI?H05"21"850L"490UI4:0"125"2J054:/26h"

.2I5<"21"826:/6I2I<"L25Oh"

.2I5<"21"50<:h"

100K"<:54/6h"

906/4K"21"J.D</84K"90:05/254:/26"K049/6=":2"9/<4H/K/:Dh"

H48O"/6ZI5D"25"J4/6"41108:"H29/KD"?2;0?06:h"

<./J"?2:/26h"

H29D"5048:/26":2"<./J"?2:/26h"

L04:.05h"

<./JJ/6="<04<"

?4/6:4/6"J2<:I50h"

41:05908O"J225"90</=6h"

J225"25"1IKKD"0UI/JJ09h"

0UI/J?06:"I<09"4<"90</=609h"

?4/6:064680"21"0UI/J?06:h"

60L"0UI/J?06:"H0::05"HI:"J50<06:"/6<:4KK4:/26"L25O<h"

L/KK/6=60<<":2":5D"60L"1/<./6="?0:.29<h"

/6:06909"2J054:/26h"

826;05:09"152?"260"1/<."<J08/0<":2"84:8."462:.05h"

Page 166: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!''""

K/10"<4;/6="0UI/J?06:"4880<</HK0h"

"

W7BBVAFVb"jVBBV_"GWVb7>RGAB"

"

W7BBVAFVb"CbRVPRAF"`fVA"CG7bNRAFh"

"

14?/K/45"L/:."0?05=068D"0\/:<h"

4L450"21"?I<:05"<:4:/26h"

5049"/6<:5I8:/26"JK48459h"

<:I9D"9/4=54?<S";/902"0:8Mh"

12I69"K/10"Z48O0:h"

:5/09"26"K/10"Z48O0:h"

50<J/54:25D"?4<O<"/1"4;4/K4HK0h"

J45:/8/J4:0"4:"H24:"95/KKh"

"

RAB>bE->RGAB"

"

L4<"/:"L0KK"I6905<:229h"

82??I6/84:/26<h"

@" 8269I8:"21"J4<<06=05<h"

@" 0;48I4:/26h"

@" H24:"95/KK<h"

@" J50;06:/;0"?04<I50<h"

@" 5049"/6<:5I8:/26<"26"H48O"21"84H/6"9225h"

@" <410:D"50=IK4:/26<h"

@" 50J25:/6="1/50<h"

Page 167: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!'(""

@" <2I69/6="4K45?h"

@" J4<<06=05<"I6905<:469"<./J"L./<:K0"</=64Kh"

@" .2L":2"950<<"469"L.4:":2":4O0":2"4<<0?HKD"<:4:/26h"

@" 90?26<:54:0".2L"JI:"26""K/10"Z48O0:h"

@" .2L":2"06:05,8269I8:,K04;0"K/10H24:h"

@" H5/01"26"J52./H/:09"4504<S"908O<S"<J480<h"

"

bVWGb>RAF"GP"BfRWWRAF"7--RNVA>"

"

4H625?4K"</:I4:/26h"

/6125?"850L"?0?H05h"

4K45?"/??09/4:0KD"=/;06"41:05L459<h"

292I5S"<?2O0S"1/50"

488/906:"/68/906:h"

W7"4662I680?06:"50"1/50h"

?2;0"4L4D"152?"946=05"^260h"

@" J52:08:/;0"?04<I50<"125"8./K9506h"

4::069/6="J4<<06=05<"41:05"4K45?h"

9/508:09":2L459<"?I<:05"<:4:/26h"

J4<<06=05<"=2/6=":2":.0/5"84H/6<h"

J4<<06=05<"=2/6="/6</90";0<<0Kh"

50UI0<:"50K4:/;0"4:"?I<:05"<:4:/26h"

J4<<06=05<"12KK2L"/6<:5I8:/26<h"

J548:/80"94/KD"=2/6=":2"?I<:05"<:4:/26h"

/6125?"26"J50<06:"</:I4:/26":2"54/<0"8261/90680h"

0<825:".0KJK0<<"J4<<06=05<h"

Page 168: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!')""

92"62:"I<0"K/1:,0K0;4:25<h"

2H0D"/6<:5I8:/26<h"

8K045"06:54680,4880<<":2"K/10H24:,K/10"541:h"

62"JI<./6="L./K0"0?H45O/6=h"

O00J"<I5;/;4K"0UI/J?06:"26h"

J52;/</26<"9/<:5/HI:09"HD":.0"211/805h"

9/<8/JK/60"/6"K/10H24:h"

K22O2I:"125"J02JK0"2;05H2459h"

4::069"J02JK0"50UI/5/6="?09/84K"4/9h"

62"95/6O/6="21"<04"L4:05h"

<069"9/<:50<<"549/2"84KKh"

1/50"528O0:<h"

"

"

>7AaVb"dGR_e"GWVb7>RGAB"

"

WbV@7bbRj7_h"

"

/612"152?":05?/64Kh"

82?J4:/H/K/:D"?46/12K9,.2<0"<IJJ25:S"14/5K049<,K/60<S"H/::<h"

/612"152?"845=2h"

0\8.46=0"21"/612"<./J","<.250":05?/64Kh"

;0<<0K"90108:<h"

1I68:/264K":0<:S";0<<0K"?48./605D"0:8"469"2J054:/264K"0UI/Jh"

2J06/6=<"<08I509h"

85I90"2/K"L4<./6="#%".2I5"62:/80h"

Page 169: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!'*""

4=500?06:"26"L25O/6="K46=I4=0h"

?/6/?I?"850L/6="dB410"3466/6="N28I?06:eh"

805:/1/84:/26"850Lh"

:I=<"4<</<:4680h"

50UI/509"90J:.h"

<.0K:05"<04"<L0KKh"

?4620I;5/6="4504h"

0?05=068D"468.254=0h"

4K:0564:0"468.254=0h"

L04:.05"8269/:/26<h"

L04:.05"125084<:h"

125084<:"?26/:25/6=h"

/80"8269/:/26<h"

J52J05"850L"8K2:./6=h"

:5411/8"906</:Dh"

J.D</84K"8261/=I54:/26"21":05?/64Kh"

J25:"4I:.25/:D"62:/1/84:/26h"

549/2"826:48:h"

106905h"

?225/6="JK46h"

?225/6="J52809I50<h"

:2L/6="L/50<h"

UI/8O"50K04<0"?225/6=<h"

4I:2?4:/8"L/68.0<h"

2;05.46=/6="J52:5I</26<h"

IJ5/=.:S":5/?S"K/<:h"

Page 170: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!(+""

49;/<0"K284K":5411/8h"

K22O2I:h"

06=/60"522?"62:/80h"

8K046@IJ"2J054:/26"J52809I50<h"

.2<0<":0<:09"469"805:/1/84:0<h"

"

>b7ABPVb"GWVb7>RGA"

"

:546<105"J52809I50<h"

L4:8."<IJ05;/</26h"

82??I6/84:/26"?046<h"

0?05=068D"<.I:"92L6h"

5049D":2"?2;0"J52809I50<h"

<410"4880<<"4L4D"152?".2<0<h"

K/10"<4;/6="0UI/J?06:"6045"4880<<h"

K/10"HI2D"6045"=46=L4Dh"

826608:,9/<826608:"J52809I50<h"

1K46=0<"1IKKD"H2K:09h"

1K46=0<"?45O09h"

:22K<"4;4/K4HK0"125"54J/9"9/<826608:/26h"

.2<0"K/1:/6="0UI/J?06:"2J054:/264Kh"

<8IJJ05<"469"95/J":54D<h"

I6I<09"845=2"826608:/26<h"

HK46O"1K46=0<"1IKKD"H2K:09h"

;4K;0<"<04K09"<.I:h"

;4K;0<"?26/:25/6=h"

Page 171: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!(!""

50?2:0"2J054:09";4K;0<h"

92IHK0"8.08O";4K;0<h"

845=2":46O"K/9<h"

2J06/6=<"450"8K2<09"=4<":/=.:h"

2J0609"9I5/6="IKK4=0"469"<4?JK/6="26KDh"

:46O";06:/6=h"

;06:/6="J52809I50<"4880J:09"HD"<./J"469"<.250h"

2J06":2"4:?2<J.050";/4"2J06"IKK4=0h"

1K4?0"<85006"26"2J06"IKK4=0"J25:<"9I5/6=";06:/6=h"

1/\09";06:/6="I</6="R605:"F4<"BD<:0?"dRFBe"

<.250";06:/6="I</6=";4J2I5"8/58IK4:/6="<D<:0?h"

.469":258.0<"4JJ52;09"469"2J054:/264Kh"

J25:4HK0"jfP"4JJ52;09"469"2J054:/264Kh"

J4=05<"469"J25:4HK0"J.260<"<.I:"211h"

54945"469"?4/6"549/2"<.I:"211h"

J25:4HK0"0K08:5/84K"0UI/J?06:"J52./H/:09h"

<IJJKD"84HK0<"9/<826608:09h"

<./J"<.250"82??I6/84:/26"84HK0<"4L4D"152?"L25O"^260h"

4882??294:/26"9225<"L/692L<"<.I:h"

9225<"O0J:"I6K28O09h"

</=6<"J2<:09"6045"9225<h"

4/5"8269/:/26/6="/6:4O0<"<.I:h"

L/692L":DJ0"4/5"8269"<.I:"211h"

<?2O/6="J52./H/:09"26"H2459"469"Z0::Dh"

62"<?2O/6="JK48459<h"

64O09"K/=.:<S"1/50S"<J45O"125?4:/26h"

Page 172: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!(#""

.2:"L25O"J05?/::09"<IHZ08:":2"J25:"4I:.25/:Dh"

62"64O09"K/=.:<"JK48459<h"

0?05=068D"0<84J0h"

K/10H24:"5049Dh"

<I11/8/06:"?466/6=h"

82?J0:06:"J05<2660K"<./J"469"<.250h"

?225/6="L/50"L/:."1/H50":4/Kh"

84:.29/8"J52:08:/26"J52809I50h"

JI?J522?";06:/K4:/26h"

?08.46/84K";06:/K4:/26h"

826:/6I2I<";06:/K4:/26h"

8K2<0"IKK4=0"J25:<"I6K0<<"90</=6"9/8:4:0<"2:.05L/<0h"

=4<"9/<JK4809";06:09";/4";06:"<:48Oh"

;06:"<:48O";0K28/:D";4K;0h"

=4<0<"8K045"21"845=2"908Oh""

:46O"L4<./6="J52809I50<h"

/KKI?/64:/26"<410"4880<<S":546<105"L25O"469"0?05=068D"4504h"

4KK"52I69"509"64;/=4:/26"K/=.:h"

H"1K4=h"

"

R605:"F4<"BD<:0?"dRFBe"469"-5I90"G/K"`4<."d-G`e"

"

1/\09"2\D=06"464KD<05"469"50825905h"

84K/H54:/26"21"0UI/J?06:h"

1/K:05"8K046/6="25"50JK480?06:h"

J25:4HK0"2\D=06"469".D952845H26"?0:05<h"

Page 173: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!($""

:0<:"I</6="2J054:/26<"?46I4K"25"?46I148:I505h"

/605:"=4<"90K/;0509h"

2\D=06"H0K2L"&p"/6"/605:"=4<"90K/;05Dh"

2\D=06"H0K2L")p"/6";2KI?0"/6":46O<h"

J2</:/;0"J50<<I50"21":46O"4:?2<J.050"d!++??"=4I=0eh"

85I90"2/K"L4<."JK46h"

85I90"2/K"L4<."4H25:"8269/:/26h"

?48./60<"I<09"J52J05KD"<0:h"

<I11/8/06:"95/;0"I6/:<h"

?I<:"62:"?2;0"4"95/;0"I6/:"?250":.46":L/80h"

;4K;0<"469"K/60<"J52J05KD"<0:h"

62"J/J0K/60"/6"?48./605D"<J480h"

.04:05"<I11/8/06:KD"/<2K4:09h"

.D9546:";4K;0<"HK46O09"211h"

;4K;0<"1/\09":2":46O"L4<./6="<.I:"211h"

:46O"L4<./6="<D<:0?"J50<<I50":0<:09"125"625?4K"2J054:/26h"

<:5/JJ/6="?26/:25/6="0UI/J?06:h"

50?2:0"5049"2I:"148/K/:/0<h"

"

PRbV"l"BWR__"l"V3VbFVA-k"

"

<./J"<.250"1/50"?4/6h"

<I11/8/06:"J50<<I50h"

/6:0564:/264K"826608:/26h"

1/50"62:/80h"

0?05=068D"1/50"826:52K"JK46<h"

Page 174: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!(%""

JK46<"<:254=0"/6"L04:.05":/=.:"068K2<I50"2I:</90"908O.2I<0h""

826:/6=068D"JK46h"

@" .2<0"HI5<:S"1548:I509"J/J0h"

@" 48:/;4:0"UI/8O"8K2</6="<D<:0?h"

@" 0?05=068D"J52809I50<h"

@" <:45:"1/50"4K45?h"

@" <:45:"L4:05"/55/=4:/26"/1"<J/KK4=0"26"908Oh"

@" 1/50@1/=.:/6="0UI/J?06:"469"H504:./6="4JJ454:I<"<:469@HDh"

@" J50J450"0?05=068D"J45:Dh"

@" <:45:"1/50"JI?Jh"

@" <:2J"4KK"06=/60<"469"L25O"/1"6080<<45Dh"

J52809I50":2"50J25:"J2KKI:/26h"

K/<:"21"4I:.25/:/0<":2"62:/1Dh"

48:/26":4O06":2"826:52K"9/<8.45=0h"

J52809I50<S"826:48:"125"82259/64:/6="<./JH2459"48:/26h"

0;48I4:/26"J52809I50<"/1"50UI/509h"

1/5<:"/6125?09"21"</:I4:/26h"

@" :/?0"21"<J/KKh"

@" UI46:/:D"21"<J/KKh"

@" :46O<"41108:09h"

@" ?0:.29":2"K/?/:"<J/KKh"

@" <04K/6=h"

@" 2/K"H22?h"

@" :546<105"21"J529I8:<h"

@" <K2J":46O<h"

@" IKK4=0,L.06h"L.2?h"508259<h"

Page 175: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!(&""

@" 84J48/:D"JK46h"

@" :0?J054:I50"21"K/UI/9h"

@" </:I4:/26"?26/:2509"HD"L.2?h"

@" I<0"21"/605:"=4<"<D<:0?h"

@" 94?4=0"4<<0<<?06:h"

@" 48:/26":4O06h"

@" 826:/6=068D"JK46h"

@" K4<:"J548:/80"94:0"469"JK480,H24:S1/50S<J/KK"95/KK<h"

@" /6125?"?4<:05h"

@" /6125?"K284K":5411/8h"

@" 84KK"4<</<:4680h"

@" ?2;0?06:"21"<./Jh"

@" 9I:D"J05<2660Kh"

@" j>Bh"

@" 50?2:0";4K;0"<.I:@211"<D<:0?h"

@" 2/K"508259"H22Oh"

@" K04O4=0"/6";2/9"<J480h"

@" 48:/;4:0"UI/8O"8K2</6="<D<:0?h"

@" K/UI/9"JI?J09"/6"I694?4=09":46O"25"4<.250h"

@" 4882??294:/26"9225<"469";06:/K4:/26"8K2<09h"

@" <?2O/6="469"64O09"K/=.:<"J52./H/:09h"

@" 1/50@1/=.:/6="0UI/J?06:"469"H504:./6="4JJ454:I<h"

@" 0?05=068D"J45:/0<h"

@" L4:05"H4KK4<:"H0K2L"94?4=09"845=2":46O":2"H0"8/58IK4:09":2"4;2/9"1500^/6="469"1548:I50"21"<:00K".IKKh"

@" 50@K/UI0148:/26"JK46:"I<09":2"509I80":46O"J50<<I50h"

Page 176: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!('""

>7AaVb"d-fV3R-7_"i"WbGNE->e"GWVb7>RGAB"

"

90:05?/60"<./J":DJ0"RS"RRS"RRRh"

805:/1/84:0"21"1/:60<<h"

<I5;0D<h"

90:05?/60"J05<2660K"805:/1/84:/26S"K0;0K"RS"RRh"

0\0?J:/26"125"626@82?JK/4680h"

501050680"928I?06:":J")!#+h"

H01250"!cIKD)'h"CIKO"-.0?/84K<"-290dC-fe,41:05"R6:0564:/264K"CIKO"-.0?/84K"-290"dRC-eh"

@" R3G":DJ0"R"469"RR".4;0"92IHK0"H2::2?<"469"L/6=":46O<,:DJ0"RRR"</?/K45":2":46O05h"

@" ?45/60"J2KKI:/26"345J2K"($,()h"

34:05/4K"B410:D"N4:4"B.00:"d3BNBe"<.00:";4K/9"$"D045<h"

3BNB"/<<I0"94:0h"

3BNB"<IJJK/09"HD"?46I148:I505"25"<./JJ05h"

K4H0K"50?2;09"4K:0509h"

501050680"928I?06:4:/26"26"H2459h"

J52809I50"469"45546=0?06:"?46I4K<h"

/6125?4:/26"4880<</HK0":2"850Lh"

:54/6/6="21"850Lh"

84KK"-7AE>V-"d'!$e"**'"@"''''"125"/6125?4:/26h"

845=2"508259"H22OS"$"D045<"26"H2459h"

/6<J08:/26S"805:/1/84:/26"928I?06:<h"

0\J05/0680"L/:."8.0?/84K<"26"H2459h"

:22"14?/K/45/^09S"62:"J4D/6="4::06:/26h"

.49"J52HK0?<"L/:."8.0?/84K"H01250h"

824:/6="826:4?/64:/26h"

Page 177: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!((""

J50@;06:/K4:/26h"

:46O"8K046/6=S"/605:"=4<S";06:<h"

/605:"=4<"<D<:0?"2J054:/264K"469"/6"I<0h"

845=2"JK46"152?"850L"25"<.250h"

845=2"JK46"4JJ52;4Kh"

845=2"5048:/;/:Dh"

J529I8:"JI5/:Dh"

826608:/26<":54809"HD"<.250"850Lh"

?46/12K9<"?45O09S"K4H0KK09h"

845=2"64?0"8261I<09"L/:."462:.05"J529I8:h"

501050680"21"J529I8:h"

845=2"9/<<2K;0"L/:."49Z4806:"J529I8:<h"

;06:/K4:/26"<D<:0?"41108:".04K:.h"

845=2"K24909"/6"J5090:05?/609":46O"d8.46=0eh"

=4I=0"5049/6=h"

=4I=0"50<0:h"

50?2:0"5049/6="21"845=2"<:4:0h"

2;051/KK/6=h"

L456/6="</=6<h"

P54?0"JI?J<S"<IH?05=09"JI?J<h"

K/?/:"21"J25:4HK0"JI?J<h"

900J"L0KK"845=2"JI?J"/6"048.":46Oh"

JI?J"50J4/5S"?4/6:064680h"

JI?J";/H54:/26h"

50=IK4:0"JI?J"1K2L"L/:.":.52::K0";4K;0h"

JI?J"50<:5/8:/26<h"

Page 178: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!()""

JI?J"94?4=0"HD"J529I8:h"

JI?J"522?"26"H2459S"8.08O<h"

.04:"0\8.46=05<h"

.04:09":46O<h"

<K2J":46O"/6"I<0h"

<410:D"8K2:./6="I<09h"

50<J/54:25"<:254=0"H4=S"H/6S"148/K/:Dh"

50<J/54:25"2J054:/264K"K/?/:S"0\J/5D"94:0h"

8.46=0"21"845:5/9=0S"<04K<h"

0UI/J?06:"8.08O09"HD"82?J0:06:"J05<26h"

1/:60<<"125"I</6="<410:D"0UI/J?06:h"

;/8:/?"L/<.09":2"</:"92L6"25"K4D"92L6"H01250"82?4h"

?09/84K"1/5<:"4/9"=I/90h"

<./J"<.0KK"JK4:/6=S"/6<J08:/26S"J/::/6=S":./8O60<<h"

J2</:/26/6="21"4KK"82110594?<h"

J50<<I50"152?"<./JJ05<S"HID05<S"<0KK05<h""

@" /6<J08:/26"21":46O<"HD"<./JJ05t<"<I5;0D25,8K046i95Dh"

@" 9I5/6=";[email protected]":0?J054:I50,J50<<I50,8.08O"<:4:0"21"/6./H/:25"/6"845=2h"

9I54:/26"21";2D4=0h"

50=IK45"8.08O<h"

/6./H/:25<"49909h"

:0?J054:I50"826:52Kh"

I<0"21"/6:5/6</84KKD"<410"0UI/J?06:"/6"I<0"26"908Oh"

:0K0\"469"2:.05"?09/4"/612"<2I580h"

0<:/?4:0"UI46:/:D"21"<J/KKh"

826:/6=068D"JK46h"

Page 179: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!(*""

:46O"824:/6="94?4=09h"

:46O"824:/6="82?J4:/HK0"L/:."845=2h"

805:/1/84:0"125"4880<<h"

82K2I5"21"805:/1/84:0"

84K/H54:/26"?46I4K"21"0UI/J?06:h"

0UI/J?06:"?290KS"<05/4K"6I?H05h"

<06</:/;/:D"21"?0:50S"0UI/J?06:h"

6I?H05"21"928I?06:<h"

:0<:"?0:.29h"

J50;/2I<":.500"845=2<S"J/J/6=S":46Oh"

JI5=/6="82110594?<S":46O<S"J/J0<h"

0;4KI4:0"8269/:/26"H01250"06:05/6=":46Oh"

<4:/<1/09"L/:.":46O"8269/:/26h"

805:/1/84:0"21"/6./H/:/26h"

K/10:/?0"21"/6./H/:25h"

/6./H/:25"26"H2459h"

9/<8.45=0"J52809I50<h"

L4<."J52809I50<h"

<:4:/8"HI/K9"IJS"508/58IK4:/6="L4:05S"<:04?h"

:46O"8K046/6="J529I8:S"/612S"<08I5/:D"?04<I50<h"

954/64=0"21";06:"<D<:0?h"

K249"54:0h"

K249"8.46=0"?490"HD"UI4K/1/09"J05<26h"

1K4?0"<85006"/6"JK480h"

4K45?<S"HD@J4<<S"2;05@5/90S"/<2K4:0S"<.I:"92L6h"

62"J52:0/6"124?"125"1/50"1/=.:/6=h"

Page 180: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!)+""

JI?J"522?"L/:."-G#"25"2:.05"<?2:.05/6="8.0?/84Kh"

124?"82?J4:/HK0"L/:."J529I8:<"8455/09h"

<I11/8/06:"<IJJKD"21"124?h"

"

b2@b2"BfRW"GWVb7>RGAB"

"

</KK"H0K2L":.0"IJJ05?2<:"K249"K/60h"

.D954IK/8"<08I5/6="90;/80"J52;/909h"

4K:0564:0"<08I5/6="90;/80"HD"=54;/:Dh"

?08.46/84KKD"K28O/6=":.0".D954IK/8"<D<:0?h"

4I9/HK0"4K45?h"

0?05=068D"<2I580h"

4K45?"/6"06=/60@522?h"

:0K0;/</26"<I5;0/KK4680"<D<:0?h"

H2L"9225"4K45?"26"64;/=4:/26"H5/9=0h" " "

L4:05"K04O4=0"90:08:/26"<D<:0?h"

?26/:25"/6"06=/60@522?h"

?26/:25"J2</:/26"21"/6605"469"2I:05"H2L"9225h"

6I?H05"21"<08I5/6="90;/80<h"

"

GWVb7>RAF"7AN"37RA>VA7A-V"37AE7_"

"

4JJ52;09"82JDh"

K4<:"4?069?06:<h"

<I5;0D25"06925<0?06:h"

"

Page 181: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!)!""

GWVb7>RAF"WbG-VNEbVB"37AE7_"

"

/612"2J06/6="8K2</6="9225<h"

O0J:"26"H2459h"

J2<:09"/6<:5I8:/26<h"

"

jRBE7_"RABWV->RGA"

"

.IKKS"9225"</90"<IJJ25:S"./6=/6="45?<"469"L0K9/6=h"

8.08O"125"8548O<"

50J4/5<h"

:./8O60<<"?04<I50?06:<"HD"<I5;0D25h"

82552</26h"

90125?4:/26"d62";/</HK0"90125?4:/26"4880J:09eh"

./6=0<"469"H045/6=<S":5I<:"H045/6=<h"

8K0454680<"?04<I509"4:"466I4K"<I5;0D<h"

9/<?46:K/6="125"?04<I50?06:"JI5J2<0<"4:"<J08/4K"<I5;0D<h"

8K0454680"62:":2"0\8009h"

<08I5/6=S"<IJJ25:/6="469"K28O/6="90;/80<h"

8.08O"<I5;0D25"466I4K"<I5;0D"94:4h"

8K0454680<h"

:./8O60<<"?04<I50?06:<"4:"<J08/4K"<I5;0D<h"

82552</26"?4\/?I?"4KK2L09"d!&peh"

J48O/6="?4:05/4KS"5IHH05"=4<O0:<h"

4880J:4HK0"J05?4606:"82?J50<</26"21"<04K<"d!+peh"

8548O<S".45906/6="9I0":2"4=0/6="5060L09h"

Page 182: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!)#""

2;05"J4/6:/6=h"

=504<0"26"J48O/6="d4=4/6<:"/80"82K9"8K/?4:0eh"

954/64=0"45546=0?06:h"

:0<:"21"H/K=0"<D<:0?"H0:L006"/6605"2I:05"9225h"

626"50:I56";4K;0<"4:"<J08/4K"<I5;0D<h"

<.0KK"469"2I:05"9225<h"

2J054:/26"9I5/6="2J06/6="469"8K2</6="2J054:/26h"

<?22:."<D??0:5/84K"?2;0?06:h"

J52J05"L25O/6="21"./6=0<S":5I<:"H045/6=S"K28O/6="90;/80h"

<08I5/6="<IJJ25:/6="469"K28O/6="90;/80<h"

5I6"<?22:.KD"L/:.2I:"/6:051050680h"

?08.46/84K"K28Oh"

.D954IK/8"K28O":2"50?4/6"K28O09"/6"0;06:"21"1KI/9"K2<<h"

2J06,8K2<0"J2</:/26"J52J05KD"/69/84:09"4:"50?2:0"826:52Kh"

826:52K"J460K<"/64880<</HK0":2"I64I:.25/^09h"

/6<:5I8:/26"JK48459"4KK"<08I509"469"K28O09"H01250"90J45:I50h"

<04K/6="45546=0?06:h"

.2<0":0<:"4:"466I4K"<I5;0Dh"

/69/84:25"<D<:0?h"

:0<:"4:"466I4K"<I5;0Dh"

26"H5/9=0S";/</HK0"/69/84:/26"469"4I9/HK0"4K45?h"

.45H2I5","<04"1I68:/26<h"

K4?J":0<:"1I68:/26h"

?I<:"62:"H0"J2<</HK0":2":I56"211"/69/84:25"K/=.:h"

14/K"<410"J05125?4680h"

J2L05"<2I580"125"9225<"/690J06906:"21"/69/84:25"<D<:0?h""

Page 183: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!)$""

8269/:/26"21"<06<25<"469"J52:08:/26"152?"L4:05S"/80S"94?4=0h"

L4:05"K04O4=0"90:08:/26"<D<:0?h"

L25O":0<:"4:"466I4K"<I5;0Dh"

4I9/HK0"4K45?"26"H5/9=0"J460K"469"06=/60"522?"J460Kh"

:0K0;/</26"<I5;0/KK4680"<D<:0?h"

L25O/6=":0<:"4:"466I4K"<I5;0Dh"

?26/:25"/6"06=/60@522?"469"H5/9=0h"

0K08:5/84K"0UI/J?06:h"l"2J06/6="8K2</6="<08I5/6="9225"4:"466I4K"<I5;0Dh"

Page 184: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!)%""

P7B>"PVbbk"GWVb7>RGAB"

"

@" K469"/?J090"64;/=4:/26"

@" 826=0<:09"8.4660K"

@" J/K2:4=0":08.6/UI0"4JJK/09"

@" L04:.05"8269/:/26<"=229"

@" JK2::/6="26":549/:/264K"8.45:"

@" ?46I4K"JK2::/6="

@" 0K08:526/8"8.45:"

@" 54945"J52;/90"0<<06:/4K"8.08O/6="

@" 92IH:"4<":2"J2</:/26"25"64;/=4:/26"</:I4:/26"

@" </:I4:/26"4<<0<<09";/<I4KKD"25"0K08:526/84KKD"

@" 8.08O";0<<0Kt<":548O"

@" 8.08O"<I552I69/6="946=05<"

@" <J009"21"64;/=4:/26"d4H2;0"%+"O6e"

@" <410:D"?45=/6<"

@" ;/</H/K/:D":L/80"<:2JJ/6="9/<:4680"

@" :5411/8"<0J454:/26"<8.0?0<"

@" 4JJ5248."<D<:0?<"

@" J4<<"8K2<0":2"946=05<":2"64;/=4:/26"

@" O00J"4":/?0@:4HK0";<"<410"2J054:/26"

@" 945O60<<""

@" J225";/</H/K/:D"

@" 4/9<":2"64;/=4:/26"4JJ52J5/4:0":2"14<:"64;/=4:/26"

@" K/=.:<"1K4<.":22"<K2LKD"

@" 52I=."<04"

Page 185: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!)&""

@" 14?/K/45/:D"L/:."52I:0"

@" 50UI/50?06:":2"12KK2L"90</=64:09"52I:0"

@" 52I:0"=2;056?06:"4JJ52;09"

@" /?J52;0?06:<"21"64;/=4:/264K"?45O<"

@" <J009"=I/90K/60<"

@" <J009<"=2;056?06:"4JJ52;09"

@" :L2"/690J06906:"64;/=4:/26"<D<:0?<"26"H2459"

@" 0K08:526/8"8.45:"

@" 54945"

@" 54945"9/110506:/4:0"H0:L006"HI2D<"469";0<<0K<" "

@" 9/<:4680":2"=2":2"4K:05"82I5<0"

@" J5/?45D"<D<:0?"I<09"4:"6/=.:"9/110506:":2"94D:/?0"<D<:0?"

@" 0UI/J?06:"14/KI50":5/==05"509I8:/26"21"<J009"

@" 14<:"O00J"2I:"21"L4D"21"<K2L"?2;/6=":5411/8"d=229";/<e"

@" /6"J225";/</H/K/:DS"54945"4<<0<<?06:"H01250";/<I4K"826:48:"

@" L.4:"/<"01108:/;0"<J009"509I8:/26"/6"J225";/</H/K/:D"

@" L/:."<?4KK"9541:S"14<:"1055D"4;2/9"<./JJ/6="K460"

@" <?4KK"8541:"2I:</90"<./JJ/6="K460<"

@" 4<</<:09"HD":5411/8"826:52K"

"

CE_a"-7bbRVb"GWVb7>RGAB"

"

<:4H/K/:D"94:4h"

R3G"B410"W548:/80"125"B2K/9"CIKO"-45=20<"dC-"-290eh"

K249"I6K249"K/?/:<"<0:"HD"90</=605<h"

?4\"12580<"469"?2?06:<"26"<./J".IKKh"

Page 186: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!)'""

=06054K"K249/6="I6K249/6="/6<:5I8:/26<h"

?4\"K249"J05"I6/:"<I51480"26":46O":2Jh"

?4\"K249"J05".2K9h"

4JJ52;09"<:4H/K/:D"469"K249/6="H22OK0:h"

.4:8."2J06/6=<"21"<I11/8/06:"</^0h"

4JJ52;09"K249/6="/6<:5I?06:":2"84K8IK4:0"<:50<<h"

.4:8.0<"/6"=229"8269/:/26h"

50UI/50?06:":2"O00J":L006"908O".4:8.0<"8K2<09"25"2J0609h"

/6/:/4K"./=."1500@14KK"952Jh"

;0<<0K"<I/:4HK0"125"4880<<":2":05?/64Kh"

K/<:"K/=.:":0<:09"J5/25":2"K249/6=h"

4/5"954I=.:h"

90:4/K<"469"84J48/:D"21";0<<0K"845=2".469K/6="=045h"

J50;/2I<"K249/6="I6K249/6="J25:h"

J50809/6="d$e"845=20<h"

GCG"K4<:"2/K"845=2"9/<8.45=0h"

826:06:"21"<K2J":46O<S"/605:09S"<04K09h"

94:0S"JK480S"4I:.25/:D"21"K4<:"=4<"1500"805:/1/84:0h"

=4<"/6"J/J0K/60<"469"JI?J<h"

L4:05"90J:."4:"H05:.h"

L4:05"906</:Dh"

?4\/?I?"</^0"<./J"125"H05:.h"

845=2".469K/6="0UI/J?06:"62:"/6./H/:"<410:Dh"

?26/:25/6="0UI/J?06:h"

106905/6=h"

6I?H05"469":DJ0"21"?225/6="K/60<h"

Page 187: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!)(""

<410"4880<<":2"L.451h"

805:/1/84:/26"21"K249/6="0UI/J?06:h"

K4<:"84K/H54:/26"21"K249/6="0UI/J?06:h"

:54/6/6="21"<.250"850L"50=459/6="K249/6="0UI/J?06:h"

L25O"<8.09IK0"21"<.250"J05<2660K"d14:/=I0eh"

:05?/64K"62?/64K"K249/6="I6K249/6="54:0h"

826;0D25"L0/=.:"?0:05<"L/:./6"!p"21"54:09"UI46:/:Dh"

UI46:/:/0<"21"845=2":2"H0"5I6"211"469"0?J:D"826;0D25h"

9/<:"152?"L4:05K/60":2":2J".4:8."824?/6=h"

9/<:"152?"<./J=<"</90":2".4:8."824?/6=h"

4=500?06:"26"K249/6="I6K249/6="J52809I50<h"

/6:06909"K249"JK46h"

8.08O"K/<:"82?JK0:09"J5/25":2"K249/6=h"

/6:06909"845=2"JK46"</=609"HD"?4<:05"469":05?/64K"J05<2660Kh"

50<:5/8:/26<"?06:/2609h"

K249/6="<0UI0680h"

J52J05:/0<"21"845=2"50UI/5/6="<J08/4K"4::06:/26h"

49;4680"/612"26"845=2h"

49;4680"/612"26"845=2".469K/6=h"

:5/??/6="J52809I50<h"

48:I4K"K249/6=,I6K249/6="JK46h"

?4<:05"469":05?/64K"49;/<09"26"4?069?06:<":2"K249"JK46h"

J52809I50":2"12KK2L"/6:06909"K249/6="JK46h"

6I?H05<"21"K249/6="I6K249/6="85460<h"

49;/<0"488259/6=KD"L.06"8.46=/6="6I?H05"21"85460<h"

/1"K249/6="I6K249/6=".IKK"K/?/:<"4::4/6S"?4<:05":2"<:2J"2J<h"

Page 188: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!))""

4?069?06:<":2"K249"JK46"8.46=09"26KD"H2:."J45:/0<"4=500h"

826:/6I2I<"?26/:25/6="HD"J05<2660Kh"

850L"J521/8/06:"/6"<.250"K46=I4=0"I<09h"

82;05"/6/:/4KKD"4KK":46O":2Jh"

50J4/5<"90K4D/6="K249/6="I6K249/6="2J<h"

455/;4K"469"J52J2<09"90J45:I50"9541:h"

:/?0"50UI/509"125"90@H4KK4<:/6=h"

O00J";0<<0K"IJ5/=.:h"

46D"50<:5/8:/26<"26"90@H4KK4<:/6=h"

<I<J06909":/?0":2"06<I50"90@H4KK4<:/6=h"

49;/<0":05?/64K"/1"90;/4:/26":2"90@H4KK4<:/6="JK46h"

8.08O"K/<:"82?JK0:09"9I5/6="K249/6=h"

.45?26/^4:/26"H0:L006"90@H4KK4<:/6="469"K249/6=h"

9I5/6="K249/6="82?J450"K24909"L0/=.:"L/:."954I=.:h"

<I11/8/06:"850L"26"H2459":2"?26/:25"845=2"2J<h"

48:/26":2"H0":4O06"/6"84<0"21"54/6,L04:.05"8.46=0h"

826:48:"J05<26"4<.250h"

:05?/64:0"90@H4KK4<:/6="H01250":5/??/6=h"

49;/<0":05?/64K"L.06":5/??/6="82??0680<":2"5I6"211"826;0D25h"

Page 189: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!)*""

50J25:"94?4=0":2"<./Jh"

.2:"L25O"50J4/5<"9I5/6="845=2"2J<h"

82JD"21"JK46<"K29=09"L/:."4JJ52J5/4:0"4I:.25/:D,J25:"<:4:0h"

K249"JK46"O0J:"</\"?26:.<"/6":05?/64K"458./;0<h"

"

"

EA_G7N"

"

:4O0"/6":2"826</9054:/26".IKK"<:50<<"469".2K9"94?4=0h"

?4<:05"469":05?/64K".4;0"4880<</HK0"/612h"

I6K249/6="JK46"/6"L5/:/6=h"

J2<</H/K/:D"21":.0"9/<8.45=0"21"1K4??4HK0";4J2I5<h"

;06:/K4:/26"21"845=2"<J480<h"

I6K249/6="469"H4KK4<:/6="488259/6=":2"4=5009"JK46h"

<./J=<"954I=.:"5049"50=IK45KDh"

94?4=0"469,25".4^459"50J25:09":2"H2:."?4<:05"469":05?/64Kh"

;0<<0K"O0J:"IJ5/=.:h"

I6K249"I6/125?KD"H2:."</90<"21"048.".2K9h"

I6K249"260"</90"4:"260"069"469"2JJ2</:0"4:"2:.05"069,:L/<:h"

54:0<","<0UI0680"62:"?29/1/09"I6:/K"<./J,:05?/64K"62:/1/09h"

62:/1/84:/26"L.06"048.".2K9"/<":05?/64:09h"

Page 190: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

!*+""

APPENDIX D, E, F, G The following appendixes are included in the project folder as separate files.

Appendix D --- msc-mepc_3-circ.2.pdf

Appendix E --- Appendix E (MAIIF Witnesses and Interviews Chapter4).pdf

Appendix E (Physical Evidence Chapter5).pdf

Appendix F --- MORT.pdf

Appendix G --- Appendix G [E&CF MAIB (OLIVIA JEAN)].pdf

"

Page 191: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

I:\CIRC\MSC-MEPC\3\2.doc

INTERNATIONAL MARITIME ORGANIZATION 4 ALBERT EMBANKMENT LONDON SE1 7SR Telephone: 020 7735 7611 Fax: 020 7587 3210

IMO

E

Ref. T1/12.01 MSC-MEPC.3/Circ.2

13 June 2008

CASUALTY-RELATED MATTERS

Code of the International Standards and Recommended

Practices for a Safety Investigation into a Marine Casualty or Marine Incident

1 The Maritime Safety Committee, at its eighty-third session (3 to 12 October 2007), and the Marine Environment Protection Committee, at its fifty-seventh session (31 March to 4 April 2008), approved the Code of the International Standards and Recommended Practices for a Safety Investigation into a Marine Casualty or Marine Incident (Casualty Investigation Code). 2 The Maritime Safety Committee, at its eighty-fourth session (7 to 16 May 2008), adopted the Casualty Investigation Code by resolution MSC.255(84) and a new regulation 6 in chapter XI-1 of the SOLAS Convention by resolution MSC.257(84) to make the Code mandatory. The Committee agreed that the Casualty Investigation Code should take effect on 1 January 2010, noting that the effective date should be the same as the date of entry into force of the new SOLAS regulation XI-1/6. 3 The annex to this circular is the Casualty Investigation Code, the text of which is identical to the text of the Code adopted by MSC 84. 4 Governments are invited to start implementing the Code of the International Standards and Recommended Practices for a Safety Investigation into a Marine Casualty or Marine Incident on a voluntary basis prior to the effective date of the Code.

***

Page 192: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews
Page 193: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

MSC-MEPC.3/Circ.2

I:\CIRC\MSC-MEPC\3\2.doc

ANNEX

CODE OF THE INTERNATIONAL STANDARDS AND RECOMMENDED PRACTICES FOR A SAFETY INVESTIGATION INTO A MARINE CASUALTY

OR MARINE INCIDENT (CASUALTY INVESTIGATION CODE)

Table of Contents Page

Foreword 2 Part I �– General provisions 4 Chapter 1 �– Purpose 4 Chapter 2 �– Definitions 5 Chapter 3 �– Application of chapters in Parts II and III 8 Part II �– Mandatory standards 9 Chapter 4 �– Marine safety investigation Authority 9 Chapter 5 �– Notification 9 Chapter 6 �– Requirement to investigate very serious marine casualties 10 Chapter 7 �– Flag State�’s agreement with another substantially interested State to conduct a marine safety investigation 10 Chapter 8 �– Powers of an investigation 11 Chapter 9 �– Parallel investigations 11 Chapter 10 �– Co-operation 11 Chapter 11 �– Investigation not to be subject to external direction 11 Chapter 12 �– Obtaining evidence from seafarers 12 Chapter 13 �– Draft marine safety investigation reports 12 Chapter 14 �– Marine safety investigation reports 13 Part III �– Recommended practices 13 Chapter 15 �– Administrative responsibilities 13 Chapter 16 �– Principles of investigation 14 Chapter 17 �– Investigation of marine casualties (other than very serious marine casualties) and marine incidents 15 Chapter 18 �– Factors that should be taken into account when seeking agreement under chapter 7 of Part II 16 Chapter 19 �– Acts of unlawful interference 16 Chapter 20 �– Notification to parties involved and commencement of an investigation 17 Chapter 21 �– Co-ordinating an investigation 17 Chapter 22 �– Collection of evidence 19 Chapter 23 �– Confidentiality of information 19 Chapter 24 �– Protection for witnesses and involved parties 20 Chapter 25 �– Draft and final report 21 Chapter 26 �– Re-opening an investigation 22

Page 194: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

MSC-MEPC.3/Circ.2 ANNEX Page 2

I:\CIRC\MSC-MEPC\3\2.doc

Foreword

1 This Code incorporates and builds on the best practices in marine casualty and marine incident investigation that were established by the Code for the Investigation of Marine Casualties and Incidents, adopted in November 1997 by the International Maritime Organization (the Organization), by resolution A.849(20). The Code for the Investigation of Marine Casualties and Incidents sought to promote co-operation and a common approach to marine casualty and marine incident investigations between States. Background 2 The Organization has encouraged co-operation and recognition of mutual interest through a number of resolutions. The first was resolution A.173(ES.IV) (Participation in Official Inquiries into Maritime Casualties) adopted in November 1968. Other resolutions followed including: resolution A.322(IX) (The Conduct of Investigations into Casualties) adopted in November 1975; resolution A.440(XI) (Exchange of Information for Investigations into Marine Casualties) and resolution A.442(XI) (Personnel and Material Resource Needs of Administrations for the Investigation of Casualties and the Contravention of Conventions), both adopted in November 1979; resolution A.637(16) (Co-operation in Maritime Casualty Investigations) adopted in 1989. 3 These individual resolutions were amalgamated and expanded by the Organization with the adoption of the Code for the Investigation of Marine Casualties and Incidents. Resolution A.884(21) (Amendments to the Code for the Investigation of Marine Casualties and Incidents resolution A.849(20)), adopted in November 1999, enhanced the Code by providing guidelines for the investigation of human factors. 4 The International Convention for the Safety of Life at Sea (SOLAS), 1948, included a provision requiring flag State Administrations to conduct investigations into any casualty suffered by a ship of its flag if an investigation may assist in identifying regulatory issues as a contributing factor. This provision was retained in the 1960 and 1974 SOLAS Conventions. It was also included in the International Convention on Load Lines, 1966. Further, flag States are required to inquire into certain marine casualties and marine incidents occurring on the high seas*. 5 The sovereignty of a coastal State extends beyond its land and inland waters to the extent of its territorial sea**. This jurisdiction gives the coastal State an inherent right to investigate marine casualties and marine incidents connected with its territory. Most national Administrations have legal provisions to cover the investigation of a shipping incident within its inland waters and territorial sea, regardless of the flag.

* Reference is made to the United Nations Convention on the Law of the Sea (UNCLOS), article 94(7) or

requirements of international and customary laws. ** Reference is made to the United Nations Convention on the Law of the Sea (UNCLOS), article 2 or

requirements of international and customary laws.

Page 195: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

MSC-MEPC.3/Circ.2 ANNEX

Page 3

I:\CIRC\MSC-MEPC\3\2.doc

Treatment of Seafarers 6 Most recently, the International Labour Organization�’s Maritime Labour Convention, 2006 (which has not yet come into force), provides a provision for the investigation of some serious marine casualties as well as setting out working conditions for seafarers. Recognizing the need for special protection for seafarers during an investigation, the Organization adopted, in December 2005, the �“Guidelines on Fair Treatment of Seafarers in the Event of a Maritime Accident�” through resolution A.987(24). The Guidelines were promulgated by the IMO and the ILO on 1 July 2006. Adoption of the Code 7 Since the adoption of the first SOLAS Convention, there have been extensive changes in the structure of the international maritime industry and changes in international law. These changes have potentially increased the number of States with an interest in the process and outcomes of marine safety investigations, in the event of a marine casualty or marine incident, increasing the potential for jurisdictional and other procedural differences between affected States. 8 This Code, while it specifies some mandatory requirements, recognizes the variations in international and national laws in relation to the investigation of marine casualties and marine incidents. The Code is designed to facilitate objective marine safety investigations for the benefit of flag States, coastal States, the Organization and the shipping industry in general.

Page 196: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

MSC-MEPC.3/Circ.2 ANNEX Page 4

I:\CIRC\MSC-MEPC\3\2.doc

PART I

GENERAL PROVISIONS

Chapter 1

PURPOSE 1.1 The objective of this Code is to provide a common approach for States to adopt in the conduct of marine safety investigations into marine casualties and marine incidents. Marine safety investigations do not seek to apportion blame or determine liability. Instead a marine safety investigation, as defined in this Code, is an investigation conducted with the objective of preventing marine casualties and marine incidents in the future. The Code envisages that this aim will be achieved through States: .1 applying consistent methodology and approach, to enable and encourage a broad

ranging investigation, where necessary, in the interests of uncovering the causal factors and other safety risks; and

.2 providing reports to the Organization to enable a wide dissemination of

information to assist the international marine industry to address safety issues. 1.2 A marine safety investigation should be separate from, and independent of, any other form of investigation. However, it is not the purpose of this Code to preclude any other form of investigation, including investigations for action in civil, criminal and administrative proceedings. Further, it is not the intent of the Code for a State or States conducting a marine safety investigation to refrain from fully reporting on the causal factors of a marine casualty or marine incident because blame or liability, may be inferred from the findings. 1.3 This Code recognizes that under the Organization�’s instruments, each flag State has a duty to conduct an investigation into any casualty occurring to any of its ships, when it judges that such an investigation may assist in determining what changes in the present regulations may be desirable, or if such a casualty has produced a major deleterious effect upon the environment. The Code also takes into account that a flag State shall* cause an inquiry to be held, by or before a suitably qualified person or persons into certain marine casualties or marine incidents of navigation on the high seas. However, the Code also recognizes that where a marine casualty or marine incident occurs within the territory, including the territorial sea, of a State, that State has a right** to investigate the cause of any such marine casualty or marine incident which might pose a risk to life or to the environment, involve the coastal State�’s search and rescue authorities, or otherwise affect the coastal State.

* Reference is made to the United Nations Convention on the Law of the Sea (UNCLOS), article 94 or

requirements of international and customary laws. ** Reference is made to the United Nations Convention on the Law of the Sea (UNCLOS), article 2 or

requirements of international and customary laws.

Page 197: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

MSC-MEPC.3/Circ.2 ANNEX

Page 5

I:\CIRC\MSC-MEPC\3\2.doc

Chapter 2

DEFINITIONS

When the following terms are used in the mandatory standards and recommended practices for marine safety investigations they have the following meaning. 2.1 An agent means any person, natural or legal, engaged on behalf of the owner, charterer or operator of a ship, or the owner of the cargo, in providing shipping services, including managing arrangements for the ship being the subject of a marine safety investigation. 2.2 A causal factor means actions, omissions, events or conditions, without which:

.1 the marine casualty or marine incident would not have occurred; or .2 adverse consequences associated with the marine casualty or marine incident

would probably not have occurred or have been as serious; .3 another action, omission, event or condition, associated with an outcome in .1

or .2, would probably not have occurred. 2.3 A coastal State means a State in whose territory, including its territorial sea, a marine casualty or marine incident occurs. 2.4 Exclusive economic zone means the exclusive economic zone as defined by article 55 of the United Nations Convention on the Law of the Sea. 2.5 Flag State means a State whose flag a ship is entitled to fly. 2.6 High seas means the high seas as defined in article 86 of the United Nations Convention on the Law of the Sea. 2.7 Interested party means an organization, or individual, who, as determined by the marine safety investigating State(s), has significant interests, rights or legitimate expectations with respect to the outcome of a marine safety investigation. 2.8 International Safety Management (ISM) Code means the International Management Code for the Safe Operation of Ships and for Pollution Prevention as adopted by the Organization by resolution A.741(18), as amended. 2.9 A marine casualty means an event, or a sequence of events, that has resulted in any of the following which has occurred directly in connection with the operations of a ship: .1 the death of, or serious injury to, a person; .2 the loss of a person from a ship; .3 the loss, presumed loss or abandonment of a ship;

.4 material damage to a ship;

Page 198: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

MSC-MEPC.3/Circ.2 ANNEX Page 6

I:\CIRC\MSC-MEPC\3\2.doc

.5 the stranding or disabling of a ship, or the involvement of a ship in a collision; .6 material damage to marine infrastructure external to a ship, that could seriously

endanger the safety of the ship, another ship or an individual; or .7 severe damage to the environment, or the potential for severe damage to the

environment, brought about by the damage of a ship or ships. However, a marine casualty does not include a deliberate act or omission, with the intention to cause harm to the safety of a ship, an individual or the environment. 2.10 A marine incident means an event, or sequence of events, other than a marine casualty, which has occurred directly in connection with the operations of a ship that endangered, or, if not corrected, would endanger the safety of the ship, its occupants or any other person or the environment. However, a marine incident does not include a deliberate act or omission, with the intention to cause harm to the safety of a ship, an individual or the environment. 2.11 A marine safety investigation means an investigation or inquiry (however referred to by a State), into a marine casualty or marine incident, conducted with the objective of preventing marine casualties and marine incidents in the future. The investigation includes the collection of, and analysis of, evidence, the identification of causal factors and the making of safety recommendations as necessary. 2.12 A marine safety investigation report means a report that contains: .1 a summary outlining the basic facts of the marine casualty or marine incident and

stating whether any deaths, injuries or pollution occurred as a result; .2 the identity of the flag State, owners, operators, the company as identified in the

safety management certificate, and the classification society (subject to any national laws concerning privacy);

.3 where relevant the details of the dimensions and engines of any ship involved,

together with a description of the crew, work routine and other matters, such as time served on the ship;

.4 a narrative detailing the circumstances of the marine casualty or marine incident; .5 analysis and comment on the causal factors including any mechanical, human and

organizational factors; .6 a discussion of the marine safety investigation�’s findings, including the

identification of safety issues, and the marine safety investigation�’s conclusions; and

.7 where appropriate, recommendations with a view to preventing future

marine casualties and marine incidents.

Page 199: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

MSC-MEPC.3/Circ.2 ANNEX

Page 7

I:\CIRC\MSC-MEPC\3\2.doc

2.13 Marine safety investigation Authority means an Authority in a State, responsible for conducting investigations in accordance with this Code. 2.14 Marine safety investigating State(s) means the flag State or, where relevant, the State or States that take the responsibility for the conduct of the marine safety investigation as mutually agreed in accordance with this Code. 2.15 A marine safety record means the following types of records collected for a marine safety investigation: .1 all statements taken for the purpose of a marine safety investigation; .2 all communications between persons pertaining to the operation of the ship; .3 all medical or private information regarding persons involved in the

marine casualty or marine incident; .4 all records of the analysis of information or evidential material acquired in the

course of a marine safety investigation; .5 information from the voyage data recorder. 2.16 A material damage in relation to a marine casualty means: .1 damage that: .1.1 significantly affects the structural integrity, performance or operational

characteristics of marine infrastructure or a ship; and .1.2 requires major repair or replacement of a major component or components;

or .2 destruction of the marine infrastructure or ship. 2.17 A seafarer means any person who is employed or engaged or works in any capacity on board a ship. 2.18 A serious injury means an injury which is sustained by a person, resulting in incapacitation where the person is unable to function normally for more than 72 hours, commencing within seven days from the date when the injury was suffered. 2.19 A severe damage to the environment means damage to the environment which, as evaluated by the State(s) affected, or the flag State, as appropriate, produces a major deleterious effect upon the environment.

Page 200: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

MSC-MEPC.3/Circ.2 ANNEX Page 8

I:\CIRC\MSC-MEPC\3\2.doc

2.20 Substantially interested State means a State: .1 which is the flag State of a ship involved in a marine casualty or marine incident;

or .2 which is the coastal State involved in a marine casualty or marine incident; or .3 whose environment was severely or significantly damaged by a marine casualty

(including the environment of its waters and territories recognized under international law); or

.4 where the consequences of a marine casualty or marine incident caused, or

threatened, serious harm to that State or to artificial islands, installations, or structures over which it is entitled to exercise jurisdiction; or

.5 where, as a result of a marine casualty, nationals of that State lost their lives or

received serious injuries; or .6 that has important information at its disposal that the marine safety investigating

State(s) consider useful to the investigation; or .7 that for some other reason establishes an interest that is considered significant by

the marine safety investigating State(s). 2.21 Territorial sea means territorial sea as defined by Section 2 of Part II of the United Nations Convention on the Law of the Sea. 2.22 A very serious marine casualty means a marine casualty involving the total loss of the ship or a death or severe damage to the environment.

Chapter 3

APPLICATION OF CHAPTERS IN PARTS II AND III

3.1 Part II of this Code contains mandatory standards for marine safety investigations. Some clauses apply only in relation to certain categories of marine casualties and are mandatory only for marine safety investigations into those marine casualties. 3.2 Clauses in Part III of this Code may refer to clauses in this part that apply only to certain marine casualties. The clauses in Part III may recommend that such clauses be applied in marine safety investigations into other marine casualties or marine incidents.

Page 201: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

MSC-MEPC.3/Circ.2 ANNEX

Page 9

I:\CIRC\MSC-MEPC\3\2.doc

PART II

MANDATORY STANDARDS

Chapter 4

MARINE SAFETY INVESTIGATION AUTHORITY 4.1 The Government of each State shall provide the Organization with detailed contact information of the marine safety investigation Authority(ies) carrying out marine safety investigations within their State.

Chapter 5

NOTIFICATION

5.1 When a marine casualty occurs on the high seas or in an exclusive economic zone, the flag State of a ship, or ships, involved, shall notify other substantially interested States as soon as is reasonably practicable. 5.2 When a marine casualty occurs within the territory, including the territorial sea, of a coastal State, the flag State, and the coastal State, shall notify each other and between them notify other substantially interested States as soon as is reasonably practicable. 5.3 Notification shall not be delayed due to the lack of complete information. 5.4 Format and content: The notification shall contain as much of the following information as is readily available: .1 the name of the ship and its flag State; .2 the IMO ship identification number; .3 the nature of the marine casualty; .4 the location of the marine casualty; .5 time and date of the marine casualty; .6 the number of any seriously injured or killed persons; .7 consequences of the marine casualty to individuals, property and the environment;

and .8 the identification of any other ship involved.

Page 202: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

MSC-MEPC.3/Circ.2 ANNEX Page 10

I:\CIRC\MSC-MEPC\3\2.doc

Chapter 6

REQUIREMENT TO INVESTIGATE VERY SERIOUS MARINE CASUALTIES 6.1 A marine safety investigation shall be conducted into every very serious marine casualty. 6.2 Subject to any agreement in accordance with chapter 7, the flag State of a ship involved in a very serious marine casualty is responsible for ensuring that a marine safety investigation is conducted and completed in accordance with this Code.

Chapter 7

FLAG STATE�’S AGREEMENT WITH ANOTHER SUBSTANTIALLY INTERESTED

STATE TO CONDUCT A MARINE SAFETY INVESTIGATION 7.1 Without limiting the rights of States to conduct their own separate marine safety investigation, where a marine casualty occurs within the territory, including territorial sea, of a State, the flag State(s) involved in the marine casualty and the coastal State shall consult to seek agreement on which State or States will be the marine safety investigating State(s) in accordance with a requirement, or a recommendation acted upon, to investigate under this Code. 7.2 Without limiting the rights of States to conduct their own separate marine safety investigation, if a marine casualty occurs on the high seas or in the exclusive economic zone of a State, and involves more than one flag State, then the States shall consult to seek agreement on which State or States will be the marine safety investigating State(s) in accordance with a requirement, or a recommendation acted upon, to investigate under this Code. 7.3 For a marine casualty referred to in paragraph 7.1 or 7.2, agreement may be reached by the relevant States with another substantially interested State for that State or States to be the marine safety investigating State(s). 7.4 Prior to reaching an agreement, or if an agreement is not reached, in accordance with paragraph 7.1, 7.2 or 7.3, then the existing obligations and rights of States under this Code, and under other international laws, to conduct a marine safety investigation, remain with the respective parties to conduct their own investigation. 7.5 By fully participating in a marine safety investigation conducted by another substantially interested State, the flag State shall be considered to fulfil its obligations under this Code, SOLAS regulation I/21 and article 94, section 7 of the United Nations Convention on the Law of the Sea.

Page 203: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

MSC-MEPC.3/Circ.2 ANNEX Page 11

I:\CIRC\MSC-MEPC\3\2.doc

Chapter 8

POWERS OF AN INVESTIGATION 8.1 All States shall ensure that their national laws provide investigator(s) carrying out a marine safety investigation with the ability to board a ship, interview the master and crew and any other person involved, and acquire evidential material for the purposes of a marine safety investigation.

Chapter 9

PARALLEL INVESTIGATIONS

9.1 Where the marine safety investigating State(s) is conducting a marine safety investigation under this Code, nothing prejudices the right of another substantially interested State to conduct its own separate marine safety investigation. 9.2 While recognizing that the marine safety investigating State(s) shall be able to fulfil obligations under this Code, the marine safety investigating State(s) and any other substantially interested State conducting a marine safety investigation shall seek to co-ordinate the timing of their investigations, to avoid conflicting demands upon witnesses and access to evidence, where possible.

Chapter 10

CO-OPERATION 10.1 All substantially interested States shall co-operate with the marine safety investigating State(s) to the extent practicable. The marine safety investigating State(s) shall provide for the participation of the substantially interested States to the extent practicable*.

Chapter 11

INVESTIGATION NOT TO BE SUBJECT TO EXTERNAL DIRECTION

11.1 Marine safety investigating State(s) shall ensure that investigator(s) carrying out a marine safety investigation are impartial and objective. The marine safety investigation shall be able to report on the results of a marine safety investigation without direction or interference from any persons or organizations who may be affected by its outcome.

* The reference to �“extent practicable�” may be taken to mean, as an example, that co-operation or participation is

limited because national laws make it impracticable to fully co-operate or participate.

Page 204: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

MSC-MEPC.3/Circ.2 ANNEX Page 12

I:\CIRC\MSC-MEPC\3\2.doc

Chapter 12

OBTAINING EVIDENCE FROM SEAFARERS 12.1 Where a marine safety investigation requires a seafarer to provide evidence to it, the evidence shall be taken at the earliest practical opportunity. The seafarer shall be allowed to return to his/her ship, or be repatriated at the earliest possible opportunity. The seafarers human rights shall, at all times, be upheld. 12.2 All seafarers from whom evidence is sought shall be informed of the nature and basis of the marine safety investigation. Further, a seafarer from whom evidence is sought shall be informed, and allowed access to legal advice, regarding: .1 any potential risk that they may incriminate themselves in any proceedings

subsequent to the marine safety investigation; .2 any right not to self-incriminate or to remain silent; .3 any protections afforded to the seafarer to prevent the evidence being used against

them if they provide the evidence to the marine safety investigation.

Chapter 13

DRAFT MARINE SAFETY INVESTIGATION REPORTS

13.1 Subject to paragraphs 13.2 and 13.3, where it is requested, the marine safety investigating State(s) shall send a copy of a draft report to a substantially interested State to allow the substantially interested State to make comment on the draft report. 13.2 Marine safety investigating State(s) are only bound to comply with paragraph 13.1 where the substantially interested State receiving the report guarantees not to circulate, nor cause to circulate, publish or give access to the draft report, or any part thereof, without the express consent of the marine safety investigating State(s) or unless such reports or documents have already been published by the marine safety investigating State(s). 13.3 The marine safety investigating State(s) are not bound to comply with paragraph 13.1 if: .1 the marine safety investigating State(s) request that the substantially interested

State receiving the report to affirm that evidence included in the draft report will not be admitted in civil or criminal proceedings against a person who gave the evidence; and

.2 the substantially interested State refuses to provide such an affirmation.

Page 205: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

MSC-MEPC.3/Circ.2 ANNEX Page 13

I:\CIRC\MSC-MEPC\3\2.doc

13.4 The marine safety investigating State(s) shall invite the substantially interested States to submit their comments on the draft report within 30 days or some other mutually agreed period. The marine safety investigating State(s) shall consider the comments before preparing the final report and where the acceptance or rejection of the comments will have direct impact on the interests of the State that submitted them, the marine safety investigating State(s) shall notify the substantially interested State of the manner in which the comments were addressed. If the marine safety investigating State(s) receives no comments after the 30 days or the mutually agreed period has expired, then it may proceed to finalize the report. 13.5 The marine safety investigating State(s) shall seek to fully verify the accuracy and completeness of the draft report by the most practical means.

Chapter 14

MARINE SAFETY INVESTIGATION REPORTS

14.1 The marine safety investigating State(s) shall submit the final version of a marine safety investigation report to the Organization for every marine safety investigation conducted into a very serious marine casualty. 14.2 Where a marine safety investigation is conducted into a marine casualty or marine incident, other than a very serious marine casualty, and a marine safety investigation report is produced which contains information which may prevent or lessen the seriousness of marine casualties or marine incidents in the future, the final version shall be submitted to the Organization. 14.3 The marine safety investigation report referred in paragraphs 14.1 and 14.2 shall utilize all the information obtained during a marine safety investigation, taking into account its scope, required to ensure that all the relevant safety issues are included and understood so that safety action can be taken as necessary. 14.4 The final marine safety investigation report shall be made available to the public and the shipping industry by the marine safety investigating State(s), or the marine safety investigating State(s) shall undertake to assist the public and the shipping industry with details, necessary to access the report, where it is published by another State or the Organization.

PART III

RECOMMENDED PRACTICES

Chapter 15

ADMINISTRATIVE RESPONSIBILITIES 15.1 States should ensure that marine safety investigating Authorities have available to them sufficient material and financial resources and suitably qualified personnel to enable them to facilitate the State�’s obligations to undertake marine safety investigations into marine casualties and marine incidents under this Code.

Page 206: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

MSC-MEPC.3/Circ.2 ANNEX Page 14

I:\CIRC\MSC-MEPC\3\2.doc

15.2 Any investigator forming part of a marine safety investigation should be appointed on the basis of the skills outlined in resolution A.996(25) for investigators. 15.3 However, paragraph 15.2 does not preclude the appropriate appointment of investigators with necessary specialist skills to form part of a marine safety investigation on a temporary basis, neither does it preclude the use of consultants to provide expert advice on any aspect of a marine safety investigation. 15.4 Any person who is an investigator, in a marine safety investigation, or assisting a marine safety investigation, should be bound to operate in accordance with this Code.

Chapter 16

PRINCIPLES OF INVESTIGATION 16.1 Independence: A marine safety investigation should be unbiased to ensure the free flow of information to it. 16.1.1 In order to achieve the outcome in paragraph 16.1, the investigator(s) carrying out a marine safety investigation should have functional independence from: .1 the parties involved in the marine casualty or marine incident; .2 anyone who may make a decision to take administrative or disciplinary action

against an individual or organization involved in a marine casualty or marine incident; and

.3 judicial proceedings. 16.1.2 The investigator(s) carrying out a marine safety investigation should be free of interference from the parties in .1, .2 and .3 of paragraph 16.1.1 with respect to: .1 the gathering of all available information relevant to the marine casualty or marine

incident, including voyage data recordings and vessel traffic services recordings; .2 analysis of evidence and the determination of causal factors;

.3 drawing conclusions relevant to the causal factors; .4 distributing a draft report for comment and preparation of the final report; and .5 if appropriate, the making of safety recommendations. 16.2 Safety focused: It is not the objective of a marine safety investigation to determine liability, or apportion blame. However, the investigator(s) carrying out a marine safety investigation should not refrain from fully reporting on the causal factors because fault or liability may be inferred from the findings.

Page 207: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

MSC-MEPC.3/Circ.2 ANNEX Page 15

I:\CIRC\MSC-MEPC\3\2.doc

16.3 Co-operation: Where it is practicable and consistent with the requirements and recommendations of this Code, in particular chapter 10 on Co-operation, the marine safety investigating State(s) should seek to facilitate maximum co-operation between substantially interested States and other persons or organizations conducting an investigation into a marine casualty or marine incident.

16.4 Priority: A marine safety investigation should, as far as possible, be afforded the same priority as any other investigation, including investigations by a State for criminal purposes being conducted into the marine casualty or marine incident. 16.4.1 In accordance with paragraph 16.4 investigator(s) carrying out a marine safety investigation should not be prevented from having access to evidence in circumstances where another person or organization is carrying out a separate investigation into a marine casualty or marine incident. 16.4.2 The evidence for which ready access should be provided should include: .1 survey and other records held by the flag State, the owners, and classification

societies; .2 all recorded data, including voyage data recorders; and .3 evidence that may be provided by government surveyors, coastguard officers,

vessel traffic service operators, pilots or other marine personnel. 16.5 Scope of a marine safety investigation: Proper identification of causal factors requires timely and methodical investigation, going far beyond the immediate evidence and looking for underlying conditions, which may be remote from the site of the marine casualty or marine incident, and which may cause other future marine casualties and marine incidents. Marine safety investigations should therefore be seen as a means of identifying not only immediate causal factors but also failures that may be present in the whole chain of responsibility.

Chapter 17

INVESTIGATION OF MARINE CASUALTIES

(OTHER THAN VERY SERIOUS CASUALTIES) AND MARINE INCIDENTS

17.1 A marine safety investigation should be conducted into marine casualties (other than very serious marine casualties �– which are addressed in chapter 6 of this Code) and marine incidents, by the flag State of a ship involved, if it is considered likely that a marine safety investigation will provide information that can be used to prevent marine casualties and marine incidents in the future. 17.2 Chapter 7 contains the mandatory requirements for determining who the marine safety investigating State(s) are for a marine casualty. Where the occurrence being investigated in accordance with this chapter is a marine incident, chapter 7 should be followed as a recommended practice as if it referred to marine incidents.

Page 208: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

MSC-MEPC.3/Circ.2 ANNEX Page 16

I:\CIRC\MSC-MEPC\3\2.doc

Chapter 18

FACTORS THAT SHOULD BE TAKEN INTO ACCOUNT WHEN SEEKING AGREEMENT UNDER CHAPTER 7 OF PART II

18.1 When the flag State(s), a coastal State (if involved) or other substantially interested States are seeking to reach agreement, in accordance with chapter 7 of Part II on which State or State(s) will be the marine safety investigating State(s) under this Code, the following factors should be taken into account: .1 whether the marine casualty or marine incident occurred in the territory, including

territorial sea, of a State; .2 whether the ship or ships involved in a marine casualty or marine incident

occurring on the high seas, or in the exclusive economic zone, subsequently sail into the territorial sea of a State;

.3 the resources and commitment required of the flag State and other substantially

interested States; .4 the potential scope of the marine safety investigation and the ability of the

flag State or another substantially interested State to accommodate that scope; .5 the need of the investigator(s) carrying out a marine safety investigation to access

evidence and consideration of the State or States best placed to facilitate that access to evidence;

.6 any perceived or actual adverse effects of the marine casualty or marine incident

on other States; .7 the nationality of the crew, passengers and other persons affected by the marine

casualty or marine incident.

Chapter 19

ACTS OF UNLAWFUL INTERFERENCE 19.1 If in the course of a marine safety investigation it becomes known or is suspected that an offence is committed under article 3, 3bis, 3ter or 3quarter of the Convention for the Suppression of Unlawful Acts Against the Safety of Maritime Navigation, 1988, the marine safety investigation Authority should immediately seek to ensure that the maritime security Authorities of the State(s) concerned are informed.

Page 209: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

MSC-MEPC.3/Circ.2 ANNEX Page 17

I:\CIRC\MSC-MEPC\3\2.doc

Chapter 20

NOTIFICATION TO PARTIES INVOLVED AND COMMENCEMENT OF AN INVESTIGATION

20.1 When a marine safety investigation is commenced under this Code, the master, the owner and agent of a ship involved in the marine casualty or marine incident being investigated, should be informed as soon as practicable of: .1 the marine casualty or marine incident under investigation; .2 the time and place at which the marine safety investigation will commence; .3 the name and contact details of the marine safety investigation Authority(ies); .4 the relevant details of the legislation under which the marine safety investigation

is being conducted; .5 the rights and obligations of the parties subject to the marine safety investigation;

and .6 the rights and obligations of the State or States conducting the marine safety

investigation. 20.2 Each State should develop a standard document detailing the information in paragraph 20.1 that can be transmitted electronically to the master, the agent and the owner of the ship. 20.3 Recognizing that any ship involved in a marine casualty or marine incident may continue in service, and that a ship should not be delayed more than is absolutely necessary, the marine safety investigating State(s) conducting the marine safety investigation should start the marine safety investigation as soon as is reasonably practicable, without delaying the ship unnecessarily.

Chapter 21

CO-ORDINATING AN INVESTIGATION

21.1 The recommendations in this chapter should be applied in accordance with the principles in chapters 10 and 11 of this Code.

Page 210: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

MSC-MEPC.3/Circ.2 ANNEX Page 18

I:\CIRC\MSC-MEPC\3\2.doc

21.2 The marine safety investigating State(s) should ensure that there is an appropriate framework within the State for: .1 the designation of investigators to the marine safety investigation including an

investigator to lead the marine safety investigation; .2 the provision of a reasonable level of support to members of the marine safety

investigation; .3 the development of a strategy for the marine safety investigation in liaison with

other substantially interested States; .4 ensuring the methodology followed during the marine safety investigation is

consistent with that recommended in resolution A.884(21), as amended; .5 ensuring the marine safety investigation takes into account any recommendations

or instruments published by the Organization or International Labour Organization, relevant to conducting a marine safety investigation; and

.6 ensuring the marine safety investigation takes into account the safety management

procedures and the safety policy of the operator of a ship in terms of the ISM Code.

21.3 The marine safety investigating State(s) should allow a substantially interested State to participate in aspects of the marine safety investigation relevant to it, to the extent practicable. 21.3.1 Participation should include allowing representatives of the substantially interested State to: .1 interview witnesses; .2 view and examine evidence and make copies of documents; .3 make submissions in respect of the evidence, comment on and have their views

properly reflected in the final report; and .4 be provided with the draft and final reports relating to the marine safety

investigation*. 21.4 To the extent practical, substantially interested States should assist the marine safety investigating State(s) with access to relevant information for the marine safety investigation. To the extent practical, the investigator(s) carrying out a marine safety investigation should also be afforded access to Government surveyors, coastguard officers, ship traffic service operators, pilots and other marine personnel of a substantially interested State. 21.5 The flag State of a ship involved in a marine casualty or marine incident should help to facilitate the availability of the crew to the investigator(s) carrying out the marine safety investigation.

* The reference to �‘extent practical�’ may be taken to mean, as an example, that co-operation or participation is

limited because national laws make it impractical to fully co-operate or participate.

Page 211: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

MSC-MEPC.3/Circ.2 ANNEX Page 19

I:\CIRC\MSC-MEPC\3\2.doc

Chapter 22

COLLECTION OF EVIDENCE

22.1 A marine safety investigating State(s) should not unnecessarily detain a ship for the collection of evidence from it or have original documents or equipment removed unless this is essential for the purposes of the marine safety investigation. Investigators should make copies of documents where practicable. 22.2 Investigator(s) carrying out a marine safety investigation should secure records of interviews and other evidence collected during a marine safety investigation in a manner which prevents access by persons who do not require it for the purpose of the investigation. 22.3 Investigator(s) carrying out the marine safety investigation should make effective use of all recorded data including voyage data recorders if fitted. Voyage data recorders should be made available for downloading by the investigator(s) carrying out a marine safety investigation or an appointed representative. 22.3.1 In the event that the marine safety investigating State(s) do not have adequate facilities to read a voyage data recorder, States with such a capability should offer their services having due regard to the: .1 available resources; .2 capabilities of the read-out facility; .3 timeliness of the read-out; and .4 location of the facility.

Chapter 23

CONFIDENTIALITY OF INFORMATION 23.1 States should ensure that investigator(s) carrying out a marine safety investigation only disclose information from a marine safety record where: .1 it is necessary or desirable to do so for transport safety purposes and any impact

on the future availability of safety information to a marine safety investigation is taken into account; or

.2 as otherwise permitted in accordance with this Code*.

* States recognize that there are merits in keeping information from a marine safety record confidential where it

needs to be shared with people outside the marine safety investigation for the purpose of conducting the marine safety investigation. An example is where information from a marine safety record needs to be provided to an external expert for their analysis or second opinion. Confidentiality would seek to ensure that sensitive information is not inappropriately disclosed for purposes other than the marine safety investigation, at a time when it has not been determined how the information will assist in determining the contributing factors in a marine casualty or marine incident. Inappropriate disclosure may infer blame or liability on the parties involved in the marine casualty or marine incident.

Page 212: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

MSC-MEPC.3/Circ.2 ANNEX Page 20

I:\CIRC\MSC-MEPC\3\2.doc

23.2 States involved in marine safety investigation under this Code should ensure that any marine safety record in its possession is not disclosed in criminal, civil, disciplinary or administrative proceedings unless: .1 the appropriate authority for the administration of justice in the State determines

that any adverse domestic or international impact that the disclosure of the information might have on any current or future marine safety investigations is outweighed by the public interest in the administration of justice; and

.2 where appropriate in the circumstances, the State which provided the

marine safety record to the marine safety investigation authorizes its disclosure. 23.3 Marine safety records should be included in the final report, or its appendices, only when pertinent to the analysis of the marine casualty or marine incident. Parts of the record not pertinent, and not included in the final report, should not be disclosed. 23.4 States need only supply information from a marine safety record to a substantially interested State where doing so will not undermine the integrity and credibility of any marine safety investigation being conducted by the State or States providing the information. 23.4.1 The State supplying the information from a marine safety record may require that the State receiving the information undertake to keep it confidential.

Chapter 24

PROTECTION FOR WITNESSES AND INVOLVED PARTIES 24.1 If a person is required by law to provide evidence that may incriminate them, for the purposes of a marine safety investigation, the evidence should, so far as national laws allow, be prevented from admission into evidence in civil or criminal proceedings against the individual.

Examples of where it may be appropriate to disclose information from a marine safety record in criminal, civil,

disciplinary or administrative proceedings may include:

1 where a person the subject of the proceedings has engaged in conduct with the intention to cause a destructive result; or

2 where a person the subject of the proceedings has been aware of a substantial risk that a destructive result

will occur and having regard to the circumstances known to him or her it is unjustifiable to take the risk.

Page 213: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

MSC-MEPC.3/Circ.2 ANNEX Page 21

I:\CIRC\MSC-MEPC\3\2.doc

24.2 A person from whom evidence is sought should be informed about the nature and basis of the investigation. A person from whom evidence is sought should be informed, and allowed access to legal advice, regarding: .1 any potential risk that they may incriminate themselves in any proceedings

subsequent to the marine safety investigation; .2 any right not to self-incriminate or to remain silent; .3 any protections afforded to the person to prevent the evidence being used against

them if they provide the evidence to the marine safety investigation.

Chapter 25

DRAFT AND FINAL REPORT

25.1 Marine safety investigation reports from a marine safety investigation should be completed as quickly as practicable. 25.2 Where it is requested, and where practicable, the marine safety investigating State(s) should send a copy of a draft marine safety investigation report for comment to interested parties. However, this recommendation does not apply where there is no guarantee that the interested party will not circulate, nor cause to circulate, publish or give access to the draft marine safety investigation report, or any part thereof, without the express consent of the marine safety investigating State(s). 25.3 The marine safety investigating State(s) should allow the interested party 30 days or some other mutually agreed time to submit their comments on the marine safety investigation report. The marine safety investigating State(s) should consider the comments before preparing the final marine safety investigation report and where the acceptance or rejection of the comments will have direct impact on the interests of the interested party that submitted them, the marine safety investigating State(s) should notify the interested party of the manner in which the comments were addressed. If the marine safety investigating State(s) receives no comments after the 30 days or the mutually agreed period has expired, then it may proceed to finalize the marine safety investigation report*. 25.4 Where it is permitted by the national laws of the State preparing the marine safety investigation report, the draft and final report should be prevented from being admissible in evidence in proceedings related to the marine casualty or marine incident that may lead to disciplinary measures, criminal conviction or the determination of civil liability. 25.5 At any stage during a marine safety investigation interim safety measures may be recommended.

* See chapter 13 where provisions with respect to providing interested parties with reports on request may

alternatively be included as a mandatory provision.

Page 214: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

MSC-MEPC.3/Circ.2 ANNEX Page 22

I:\CIRC\MSC-MEPC\3\2.doc

25.6 Where a substantially interested State disagrees with the whole or a part of a final marine safety investigation report, it may submit its own report to the Organization.

Chapter 26

RE-OPENING AN INVESTIGATION 26.1 Marine safety investigating State(s) which have completed a marine safety investigation, should reconsider their findings and consider re-opening the investigation when new evidence is presented which may materially alter the analysis and conclusions reached. 26.2 When significant new evidence relating to any marine casualty or marine incident is presented to the marine safety investigating State(s) that have completed a marine safety investigation, the evidence should be fully assessed and referred to other substantially interested States for appropriate input.

__________

Page 215: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

Chapter 4 Witnesses and Interviews Collecting Data Collecting data is a critical part of the investigation. The detailed information collected by the accident investigation team is the foundation for the entire investigation, including the analyses and conclusions. These in turn become the basis for identifying preventive measures to preclude recurrences, Consequently, it is important to ensure that all relevant information is collected and that the information is accurate. Gathering and analyzing information is an interdependent process that takes place throughout the first three weeks of the investigation cycle. As preliminary analysis is conducted on the initial evidence, gaps will become apparent, requiring the team to collect additional evidence. Generally, many data collection and analysis iterations occur before the team can be certain that all pertinent evidence has been gathered and analyses are finalized. Three key types of evidence are collected during the investigation:

Human or testamentary evidence includes witness statements and observations.

Physical evidence is matter related to the accident (e.g., equipment, parts, debris, hardware, and other physical items).

Documentary evidence includes paper and electronic information, such as records, reports, procedures, and documentation.

Collecting evidence can be a lengthy, time-consuming, and piecemeal process. Witnesses may provide sketchy or conflicting accounts of the accident. Physical evidence may be badly damaged or completely destroyed. Documentary evidence may be minimal or difficult to access. Thorough investigation requires that team members be diligent in pursuing evidence and adequately explore leads, lines of inquiry, and potential causal factors until they gain a sufficiently complete understanding of the accident. The process of collecting data is iterative. Preliminary analysis of the initial evidence identifies gaps that will direct subsequent data collection. Generally, many data collection and analysis iterations occur before the team can be certain that all analyses can be finalized, The process of data collection also requires a tightly coordinated, interdependent set of activities on the part of several investigators.

16

Page 216: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

TIP It maybe helpful for the lead investigator to designate one team member to oversee evidence collection and to maintain a master list of evidence collected to date. The process of pursuing evidentiary material involves:

Collecting human evidence (locating and interviewing witnesses) Collecting physical evidence (identifying, documenting, inspecting, and preserving relevant matter) Collecting documentary evidence Examining organizational concerns, management systems, and line management oversight Preserving and controlling evidence.

Collecting Human Evidence Human evidence is often the most insightful and also the most fragile. Witness recollection declines rapidly in the first 24 hours following an accident or traumatic event. Therefore, witnesses should be located and interviewed immediately and with high priority. As physical and documentary evidence is gathered and analyzed throughout the investigation, this new information will often prompt followup questioning. Locating Witnesses Principal witnesses and eyewitnesses are identified and interviewed as soon as possible. Principal witnesses are persons who were actually involved in the accident; eyewitnesses are persons who directly observed the accident or the conditions immediately preceding or following the accident. General witnesses are those with knowledge about the activities taking place prior to or immediately after the accident (the previous watch, for example). Prompt arrival on scene by team members and expeditious interviewing of witnesses helps ensure that witness statements are as accurate, detailed, and authentic as possible. Table 6-1 lists sources that investigators can use to locate witnesses. Conducting Interviews Witness testimony is an important element in determining facts that reveal causal factors. It is best to interview principal witnesses and eyewitnesses first, because they often provide the most useful details regarding what happened. If not questioned promptly, they may forget important details.

17

Page 217: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

Preparing for Interviews Much of the investigation�’s fact-finding occurs in interviews. Therefore, to elicit the most useful information possible from interviewees, interviewers must be well prepared and have clear objectives for each interview. Interviews can be conducted after the team has established the topical areas to be covered in the interviews and after the lead investigator has reviewed with the board the objectives of the interviews and strategies for obtaining useful information. Table 6-2 provides guidelines for interview People�’s memories, as well as their willingness to assist an investigative board, can be affected by the way they are questioned. Based on the availability of witnesses, team members�’ time, and the nature and complexity of the accident, the lead investigator and team members must determine who to interview, in what order, and what interviewing techniques to employ. Some methods that previous accident investigation boards have found successful are described below. TIP A witness interview is not an interrogation. Investigators should convey the sense of a cooperative, informal meeting. Individual Versus Group Interviews. Depending on the specific circumstances and schedule of an accident investigation, investigators may choose to hold either individual or group interviews. Generally, principal witnesses and eyewitnesses are interviewed individually to gain independent accounts of the event. However, a group, interview may be beneficial in situations where a work crew was either involved in or witness to the accident. Moreover, time may not permit interviewing every witness individually, and the potential for gaining new information from every witness may be small. Sometimes, group interviews can corroborate testimony given by an individual, but not provide additional details. The team should use their collective judgment to determine which technique is appropriate. Advantages and disadvantages of both techniques are listed in Table 4-3. These considerations should be weighed against the circumstances of the accident when determining which technique to use.

18

Page 218: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

Table 4-3 Group and individual interviews have different advantages.

Individual Interviews Group Interviews

Advantages Obtain independent stories Obtain individual perceptions Establish one-to-one rapport

More time-efficient All interviewees supplement story; may get more complete picture Other People serve as �“memory joggers�”

Disadvantages More time-consuming May be more difficult to schedule all witnesses

Interviewees will not have independent stories More vocal members of the g4roup will say more and thus may influence those who are quieter �“Group Think�” may develop; some individual details may get lost Contradictions in accounts may not be revealed.

Interviewing: Do�’s and Don�’ts. Table 4-4 lists actions that promote effective interviews, and Table 6-5 lists actions to avoid while conducting interviews.

Table 4-4. Interviewing Do�’s.

Create a Relaxed Atmosphere Introduce yourself and shake hands. Be polite, patient, and friendly. Treat witnesses with respect.

Prepare the Witness

Describe the investigation�’s purpose: to prevent accidents, not to assign blame Explain that witnesses may be interviewed more than once. Stress how important the facts given during interviews are to the overall investigative process.

Record Information

Rely on a court reporter to provide a detailed record of the interview. Note crucial information immediately in order to ask meaningful followup questions.

Ask Questions Establish a line of questioning and stay on track during the interview. Ask the witness to describe the accident in full before asking a structured set of questions. Let witnesses tell things in their own way; start the interview with a statement such as �“Would you please tell me about...?�” Ask several witnesses similar questions to corroborate facts. Aid the interviewee with reference points; e.g., �“How did the lighting compare to the lighting in this room?�” Keep an open mind; ask questions that explore what has already been stated by others

19

Page 219: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

in addition to probing for missing information. Use visual aids, such as photos, drawings, maps, and graphs to assist witnesses. Be an active listener, and give the witness feedback; restate and rephrase key points. Ask open-ended questions that generally require more than a �“yes�” or �“no�” answer. Observe and note how replies are conveyed (voice inflections, gestures, expressions, etc.).

Close the Interview

End on a positive note; thank the witness for his/her time and effort. Allow the witnesses to read the interview transcript and comment if they so desire. Encourage the witness to contact the board with additional information or concerns. Remind the witness that a follow-up interview may be conducted.

Table 4-5. Interviewing Don�’ts.

DO NOT rush the witness while he/she is describing the accident or answering questions. DO NOT judge, display anger, refute, threaten, intimidate, or blame the witness.

DO NOT suggest answers.

DO NOT make promises that cannot be kept (for example, unrestricted confidentiality).

DO NOT use inflammatory words (�“violate, �” �“kill, �” �“lie, �” �“stupid, �” etc.).

DO NOT omit questions during the interview because you think you already know the answer.

DO NOT ask questions that suggest an answer, such as �“Was the odor like rotten eggs?�”

It is important to create a comfortable atmosphere in which interviewees are not a rushed to recall their observations. Interviewees should be told that they area part of the investigation effort and that their input will be used to prevent future accidents and not to assign blame. Before and after questioning, interviewees should be notified that follow-up interviews are a normal part of the investigation process and that further interviews do not mean that their initial statements are suspect. Also, they should be encouraged to contact the team whenever they can provide additional information or have any concerns. Interviewees should be aware of whether the information that they provide during the investigation may or may not be precluded from release to the public. Following these guidelines will help ensure that witness statements are provided freely and accurately, subsequently improving the quality and validity of the information obtained.

20

Page 220: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

Use of an Interpreter Preferably an interpreter will be supplied by a government approved interpretation service. However, sometimes one must hire an interpreter on scene. Local embassies or consulates and universities are good sources to inquire about interpreter availability. If necessary, contact a commercial interpreter firm and arrange for an interpreter to travel to the scene. This is costly but, without adequate interpreter services, the investigation cannot be properly conducted. When using an interpreter in interviewing it is important that the interpreter be fluent in the language and dialect spoken by the witness. The interpreter must also have a proper command of the language of the investigator. The interpreter must be able to grasp technical marine terms, and it may be necessary to arrange a prior meeting and/or have a list of common nautical terms available so that the interpreter has time to research the appropriate translation. The interpreter must be able to pass to the witness the information, as well as reflect the attitude and manner of expression you wish to convey. Further, the interpreter must be able to recognize any idiosyncrasies in the answers a witness may give and bring them to your attention, along with the reply. The witness should generally be seated in a chair opposite you with the interpreter in between but slightly to one side, so that the interpreter may conveniently face either the investigator or the witness as the conversation flows. Questions should be directly to the witness using the first person. The questioner should not refer to the witness in the third person, or ask the interpreter to "ask him" or "tell him" anything. Further, attempt to keep questions short. However, should it be necessary to pose a lengthy question, instruct the interpreter to translate the question in �“bite size�” pieces. In such instances, explain to the interpreter that you will pause occasionally to allow the interpreter an opportunity to translate incremental portions of the question. An interpreter should:

1.) merely act as a vehicle for accurately interpreting and passing information back and forth between you and the witness.

2) imitate your voice inflection and gestures as much as possible. 3) not carry on a conversation with the witness, other than directed by you. 4) pass on faithfully everything the witness said, including trivial remarks and

exclamations. 5) not evaluate the conversation him or her self.

Using an interpreter complicates an interview and can often more than double the time it take to complete the interview. Such interviews can be successful if they are well planned and

21

Page 221: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

controlled. At the conclusion of an interview, when the witness has left, it may be worthwhile asking the interpreters assessment of the witness. Evaluating the Witness�’s State of Mind Occasionally, a witness�’s state of mind may affect the accuracy or validity of testimony Provided. In conducting witness interviews, investigators should consider:

The amount of time between the accident and the interview. People normally forget 50 to 80 percent of the details in just 24 hours. Contact between this witness and others who may have influenced how this witness recalls the events. Signs of stress, shock, amnesia, or other trauma resulting from the accident. Details of unpleasant experiences are frequently blanked from one�’s memory.

Investigators should note whether an interviewee displays any apparent mental or physical distress or unusual behavior; it may have a bearing on the interview results. These observations can be discussed and their impact assessed with other members of the team. Investigators should also be aware of cultural differences that may be expected and the sub-cultures that may be on board a ship, particularly those with multi-national crewing. Issues of status and loss of face may be encountered from time to time. There may be a tendency for an interviewee to provide answers that he/she thinks the interviewer wants to hear, or a tendency to agree, just out of politeness. In other cases, the use of English or other common language may have different meanings or inferences. Under any of these circumstances an interpreter may prove really useful.

22

Page 222: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

Chapter 5 Physical Evidence Collecting Physical Evidence TIP To ensure consistent documentation, control, and security, it may be useful to designate a single team member or the administrative coordinator to be in charge of handling evidence. The investigative team proceeds in gathering, cataloging, and storing physical evidence from all sources as soon as it becomes available. The procedures for access to, and the controlling of, evidence maybe subject to National Legal requirements which vary from country to country. The most obvious physical evidence related to an accident or accident scene often includes solids such as:

Equipment Tools Materials Hardware Pre- and post-accident positions of accident-related elements Scattered debris Patterns, parts, and properties of physical items associated with the accident.

Less obvious but potentially important physical evidence includes fluids (liquids and gases). Ships use a multitude of fluids, including chemicals, fuels, hydraulic control or actuating fluids, and lubricants. Analyzing such evidence can reveal much about the operability of equipment and other potentially relevant conditions or causal factors. Care should be taken if there is pathogenic contamination of physical evidence (e.g., blood); such material may require autoclaving or other sterilization. Specialized technicians experienced in fluid sampling should be employed to help the team collect and analyze fluid evidence. If required, expert analysts should be requested to perform tests on the fluids and report results to the team.

23

Page 223: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

High speed vessel collisions or accidents involving explosions by result in an accident scene that is contaminated with human blood, body fluids or tissue remains. Upon entering such a scene, the investigative team must take proper precautions to protect itself from exposure to bloodborne pathogens. When handling potential bloodborne pathogens, universal precautions such as those listed in Appendix I, Chapter 5 should be observed to minimize potential exposure. All human blood and body fluids should be treated as if they are infectious. The precautions listed should be implemented for all potential exposures. Exposure is defined as reasonable anticipated skin, eye, mucous membrane, or parenteral contact with blood or other potentially infectious materials. Physical evidence should be systematically collected, protected, preserved, evaluated, and recorded to ultimately determine how and why failures occurred and whether use, abuse, misuse, or nonuse was a causal factor. Documenting Physical Evidence Evidence should be carefully documented at the time it is obtained or identified. The Accident Investigation Physical Evidence Log can help investigators document and track the collection of physical evidence. In a multi-investigator team investigation, the use of an evidence log will prevent several investigators asking for the same piece of evidence, thereby avoiding duplication of effort. Additional means of documenting physical evidence include sketches, maps, photographs, and videotape. Sketching and Mapping Sketching and mapping the position of debris, equipment, tools, and injured persons may be initiated by the team as soon as it arrives on scene. Position maps convey a visual representation of the scene immediately after an accident. Evidence may be inadvertently moved, removed, or destroyed, especially if the accident scene can only be partially secured. Therefore, sketching and mapping should be conducted immediately after recording initial witness statements. Precise scale plottings of the position of elements can subsequently be examined to develop and test accident causal theories. Photographing and Videotaping Physical Evidence Photography is a valuable and versatile tool in accident investigation. Photos or videos can identify, record, or preserve physical accident evidence that cannot be effectively conveyed by words or collected by any other means. Photographic coverage should be detailed and complete, including standard references to help establish distance and perspective. Videotapes should cover the overall accident scene, as well as specific locations or items of significance. A thorough videotape allows the team to minimize trips to the accident scene. This may be important if the scene is difficult to access or if it presents hazards.

24

Page 224: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

Good photographic coverage of the accident is essential, even if photographs or video stills will not be used in the investigation report. However, if not taken properly, photographs and videos can easily misrepresent a scene and lead to false conclusions or findings about an accident. Therefore, whenever possible, accident photography and videotaping should be performed by professionals. Photographic techniques that avoid misrepresentation, such as the inclusion of rulers and particular lighting, may be unknown to amateurs but are common knowledge among professional photographers and videographers. Even if photos are taken by a skilled photographer, the investigation team should be prepared to direct the photographer in capturing certain important perspectives or parts of the accident scene. Photographs of evidence and of the scene itself should be taken from many angles to illustrate the perspectives of witnesses and injured persons. In addition, team members may wish to take photos for their own reference, If available, digital photography will facilitate incorporation of the photographs into the investigation report. However, if this is not practical, high-quality 35mm photographs can be scanned for incorporation in the report. As photos are taken, a log should be completed noting the scene/subject, date, time, direction, and orientation of photos taken, as well as the photographer�’s name and camera settings. Inspecting Physical Evidence Following initial mapping and photographic recording, a systematic inspection of physical evidence can begin. The inspection involves:

Surveying the involved equipment, vehicles, structures, etc., to ascertain whether there is any indication that component parts were missing or out of place before the accident

Noting the absence of any parts of guards, controls, or operating indicators (instruments, position indicators, etc.) among the damaged or remaining parts at the scene

Identifying as soon as possible any equipment or parts that must be cleaned prior to examination or testing and transferring them to a laboratory or to the care of an expert experienced in appropriate testing methodologies

Noting the routing or movements of records that can later be traced to find missing components

Preparing a checklist of complex equipment components to help ensure a thorough survey.

These observations should be recorded in notes and photographs so that investigators avoid relying on their memories. Some investigators find a small cassette tape recorder

25

Page 225: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

useful in recording general descriptions of appearance and damage; however, the potential failure of a recorder, inadvertent tape erasure, and limitations of verbal description suggest that verbal recorded descriptions should be used in combination with notes, sketches, and photographs. Removing Physical Evidence Following the initial inspection of the scene, investigators may need to remove items of physical evidence. To ensure the integrity of evidence for later examination, the extraction of parts must be controlled and methodical. The process may involve simply picking up components or pieces of damaged equipment, removing bolts and fittings, cutting through major structures, or even recovering evidence from beneath piles of debris. Before evidence is removed from the accident scene, it should be photographed and its position noted on an appropriate sketch of the scene. Remember, once it has been moved, it will never be able to be returned to exactly the same position that it occupied before it was moved. It should then be carefully packaged and clearly identified. The readiness team or a pre-assembled investigator�’s kit can provide general-purpose cardboard tags or adhesive labels for this purpose. Equipment or parts thought to be defective, damaged, or improperly assembled should be removed from the accident scene for technical examination, If improper assembly is suspected, investigators should direct that the part or equipment be photographed and otherwise documented as each subassembly is removed. Items that have been fractured or otherwise damaged should be packaged carefully to preserve surface detail. Delicate parts should be padded and boxed. Both the part and the outside of the package should be labeled. Greasy or dirty parts can be wrapped in foil and placed in polyethylene bags or other nonabsorbent materials for transport to a testing laboratory, command center, or evidence storage facility. If uncertainties arise, subject matter experts can advise the board regarding effective methods for preserving and packaging evidence and specimens that must be transported for testing. When preparing to remove physical evidence, these guidelines should be followed:

Normally, extraction should not start until witnesses have been interviewed, since visual reference to the accident site can stimulate one�’s memory

Extraction and removal or movement of parts should not be started until position records (measurements for maps and photographs) have been made

Be aware that the accident site maybe unsafe due to dangerous materials or weakened structures

Locations of removed parts can be marked with orange spray paint or wire-staffed marking flags; the marking flags can be annotated to identify the part removed and to allow later measurement

26

Page 226: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

Care during extraction and preliminary examination is necessary to avoid defacing or distorting impact marks and fracture surfaces

The lead investigator and team members should concur when the parts extraction work can begin, in order to assure that board members have completed all observations requiring an intact accident site.

Collecting Documentary Evidence Documentary evidence can provide important data and should be preserved and secured as methodically as physical evidence. This information might be in the form of logbooks, equipment readouts, course recorder traces, licenses, documents, certificates, papers, photos, videotape, magnetic tape, or electronic media, either at the site or in files at other locations. Some work/process/system records are retained only for the workday or the week. Electronic data is often stored in a memory buffer and is overwritten as new data is acquired. Once an accident has occurred, the investigator must work quickly to collect and preserve these records so they can be examined and considered in the analysis. In some cases it may be necessary to obtain the services of a suitably competent translator. Accident investigation preplanning should include procedures for identifying records to be collected, as well as the people responsible for their collection. Because records are not always located at the scene of the accident, and some documents may be overlooked in the preliminary collection of evidence. Documents often provide important evidence for identifying causal factors of an accident. This evidence is useful for:

Thoroughly examining the policies, standards, and specifications that molded the environment in which the accident occurred

Indicating the attitudes and actions of people involved in the accident

Revealing evidence that generally is not established in verbal testimony.

Documentary evidence generally can be grouped into four categories:

Management control documents that communicate management expectations of how, when, where, and by whom work activities are to be performed

Records that indicate past and present performance and status of the work activities, as well as the people, equipment, and materials involved

27

Page 227: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

Reports that identify the content and results of special studies, analyses, audits, appraisals, inspections, inquiries, and investigations related to work activities

Follow-on documentation that describes actions taken in response to the other types of documentation.

Collectively, this evidence gives important clues to possible underlying causes of errors, malfunctions, and failures that led to the accident. Analysis of documents may involve two major aspects, cross checking documents from different sources that contain the same information or scientific analysis. Analysis could include cross checking the bridge movement or "bell" book with the engine room records. It cannot be emphasized enough that contemporaneous records, those made at the time, are of value, fair copies of log books, e.g. the scrap log copied out in a fair hand are of limited value. Of greater value is the cross checking of ship's records with external sources such as VTS tapes, harbor control tapes or log books, cargo terminal records, police records, customs records, or even TV or radio recordings. Investigators must keep an open mind and think latterly asking "who else may have similar information". Photocopies. Investigators should be sensitive to the possibility that photocopies of documents may not truly depict the original document. Erasures and/or the use of �“white out�” correction liquids, which may be apparent on the original document, may not show up on a photocopy of the document. Further, as in the case of logbooks, entire pages may be removed. If the investigator does not examine the original document, he will not know for sure that the photocopy provided him is, in fact, a true and accurate copy. Before photocopies of documents are accepted, the investigator should compare the copy with the original to assure that there have been no alterations to the original. Marine Documents. A list of maritime documentation that may be collected or reviewed during a marine accident investigation can be found on Appendix II of Chapter 5. The list while lengthy, is far from complete. The specific documents needed by the investigator will vary depending on the type of accident. International Safety Management System Accident investigations must thoroughly examine organizational concerns, management systems, and line management oversight processes to determine whether deficiencies in these areas contributed to causes of the accident. The investigation team should consider the full range of management systems through all levels of management in accordance with the International Safety Management (ISM) Code. It is important to note that this focus should not be directed toward individuals.

28

Page 228: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

The ISM Code documentation should be inspected as a matter of routine. It is important to ensure that the procedures in the code are adhered to. The ship operator�’s �“Documentation of Compliance�” is valid for 5 years, subject to annual verification. The ship�’s �“Safety Management Certificate�” is valid for 5 years subject to periodical verification by the administration. All aspects of the code are important to an investigator and include but are not confined to the following.

Training (ISM Code 6.3) Passage planning and procedures with pilot embarked (ISM Code ?) Information and language of ISM Code.(ISM Code 6.6) Plans, instructions, check lists for the safety of the ship and pollution

prevention.(ISM Code 7.0) Emergency preparedness (ISM Code 8.0) Reporting non-conforming incidents (ISM Code 9.1) Corrective Action (9.2) Maintenance (ISM Code 10.1) Critical equipment (ISM Code 10.3) Documentation (ISM Code 10.1) Record of internal audits (ISM Code 12.3)

If there was a departure from the code it is important to identify the non-conformity to establish whether the departure was consistent with reasonable decision making (see Course 1.3.4). Depending upon the incident it may also be necessary to check the ship�’s reporting of �“non-conforming incidents�”(ISM Code 9.1) and the management receipt of such records and subsequent action, which may include a record of corrective action (ISM Code 9.3). Preserving and Controlling Evidence Preserving and controlling evidence are essential to the integrity and credibility of the investigation. Security and custody of evidence are necessary to prevent its alteration or loss and to establish the accuracy and validity of all evidence collected. The point of contact is responsible for assuring that a chain of custody is established for all evidence removed from the accident scene before the board arrives. The board chairperson is responsible for establishing an evidentiary custody protocol to ensure that all evidence is well documented at the accident scene and carefully controlled when it is removed and stored after the board arrives. Evidence control procedures similar to the following guidelines will help assure that evidence is not adulterated, corrupted, or lost and that subsequent engineering tests, if conducted, and other analytical results are valid.

29

Page 229: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

Evidence should be photographed and/or videotaped in its original location immediately following the accident, provided it does not interfere with rescue or amelioration activities.

A log should be maintained stating the location, date, and time that photos and videos are taken. The Accident investigation Photographic Lag Sheet can be used for this purpose. Avoid using photographic attachments that digitally record the date and time on the negative because these images become a permanent part of the photo and may obscure evidence or important details in the photo or video. The computerized/printed date on the back of photos provided by film processors should be used in conjunction with, not in lieu of, a photo log, because the date on photos gives the day the film was processed, not the day the photos were taken.

Board members should prepare and sign an inventory of all evidentiary items collected, including statements regarding:

Lists of items removed from the scene Date and time items were removed from the scene Person who removed items from the scene Location where those items will be stored.

Evidence should be controlled by signature transfer (signatures of the recipient and the person relinquishing custody) and made available only to those who need to examine and use the evidence during the accident investigation. The Accident Investigation Physical Evidence Log Form may be used for this purpose.

Secure storage should be obtained immediately, and access to evidence controlled throughout the investigation.

Access to the room or suite of offices used by the investigation board should be restricted. No one other than board members, advisors, and support staff should have access to the board�’s office space; this includes janitorial staff.

The board chairperson should determine the disposition of evidence at the conclusion of the investigation.

Documentary evidence can easily be over-looked, misplaced, or taken. Documents can be altered, disfigured, misinterpreted, or electronically corrupted. Computer software and disks can be erased by exposure to magnetic fields. As with other evidence collected during the investigation, documentary evidence should be collected, inventoried (logged), controlled, and secured (in locked containers, if necessary.)

30

Page 230: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

APPENDIX I - Chapter 5 Universal Precautions to Prevent Contact with Bloodborne Pathogens On December 6, 1991, the U.S. Occupational Safety and Health Administration (OSHA) issued the regulation called �“Occupational Exposure to Bloodborne Pathogens (BBP)," found in Title 29, Section 1910.1030 of the Code of Federal Regulations. The standard covers those occupations having a high potential for exposure to bloodborne pathogens, including law enforcement, emergency response, and accident investigation personnel. Individuals covered by this standard should observe Universal Precautions to prevent contact with human blood, body fluids, tissues and other potentially infectious materials. Universal Precautions require that employees treat all human blood, body fluids, or other potentially infectious materials to be infectious for hepatitis B virus (HBV), human immunodeficiency virus (HIV), and other bloodborne pathogens. Appropriate protective measures to be taken to avoid direct contact with these materials include:

Use barrier protection at all times.

Prohibit eating, drinking, smoking, or applying makeup at the accident scene/mass disaster.

Use gloves when there may be hand contact with blood or other potentially infectious materials. Gloves should always be worn as if there are cuts, scratches, or other breaks in the skin. In some instances where there is heavily contaminated material, the use of double gloves is advisable for additional protection.

Change gloves when contaminated or as soon as feasible if torn, punctured, or when their ability to function as a barrier is compromised.

Always wash hands after removal of gloves or other personal protective equipment (PPE). The removal of gloves and other PPE should be performed in a manner which will not result in the contamination of unprotected skin or clothing.

Wear safety goggles, protective facemasks or shields, or glasses with side shields to protect from splashes, sprays, spatters, or droplets of blood or other potentially infectious materials. These same precautions must be taken when collecting dried stains for laboratory analyses.

Use disposable items, such as gloves, coveralls, shoe covers, etc., when potentially infectious materials are present.

Place contaminated sharps (e.g. broken glass, needles, knives, etc.,) in appropriate leak-proof, close-able, puncture-resistant containers when these sharps are to be discarded, transported, or shipped. If transported or shipped, containers should be appropriately labeled.

DO not bend, recap, remove, or otherwise handle contaminated needles or other sharps.

Use a protective device, such as a CPR mask, when performing mouth-to-mouth resuscitation.

Decontaminate all equipment after use with a solution of household bleach (diluted 1:10), 70% isopropyl alcohol, or other appropriate disinfectants.

After all evidence has been collected and the crime scene has been released, the owner or occupants of the affected property should be made aware of the potential risks from bloodborne pathogens.

Evidence containing blood or other body fluids should be completely dried before it is packaged and shipped to the laboratory for analysis. Appropriate biohazard warning labels must be affixed to the evidence container indicating that a potentially infectious material may be present.

31

Page 231: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

APPENDIX II - Chapter 5 A. Plans, Diagrams and Lists

1. General arrangement plans 2. General arrangement of engineroom machinery (elevation & plan views) 3. Shell expansion plans 4. Capacity plan 5. Main engine control system plans and description 6. Main engine fuel oil supply and return pumping/piping and tanks plans 7. Fuel oil service and transfer pumping/piping system plans and description

8. Fuel oil tank venting piping plan 9. Engineroom ventilation system plans 10. Passenger and crew ventilation system plans 11. General loading plans and procedures 12. Ullages and ullage tables 13. Bilge pumping and piping diagram and system description

14. Cargo pumping & piping plans and system description 15. Cargo tanks venting piping plans 16. IGS plans and system description 17. Ballast pumping & piping plans and description 18. Ballast tanks venting piping plans 19. Ballast tank coatings 20. Damage control plan (fire doors dampers, etc) 21. Fire detection plans and system description 22. Firemain piping & pumping system and description 23. CO piping system diagram and system description 24. Foam piping system diagram and description 25. Halon system diagram and description 26. One line electrical distribution diagram 27. List of bridge/radio room communication equipment 28. List of vessel navigation equipment

B. Statutory and Other Certificates 1. Gross tonnage/deadweight tonnage

2. Copy of the U.S. Certificate of Inspection (U.S. vessel only) 3. Copy of Cargo Ship Safety Construction Certificate 4. Copy of Cargo Ship Safety Radio Certificate 5. Copy of Cargo Ship Safety Equipment Certificate 6. Copy of MODU Certificate 7. Copy of Passenger Ship Safety Certificate 8. Copy of International Load Line Certificate 9. Copy of International Oil Pollution Prevention Certificate 10. Copy of Certificate of Class for Hull and Machinery 11. Copy of Vessel Radio Communication License

32

Page 232: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

12. Minimum Safe Manning Certificate 13. Ship�’s Certificate of Registry

14. Copy of Control Verification Certificate 15. Copy of Certificate for fire extinguishing system inspection 16. International Safety Management System (ISM) Documentation

17. Copies of Officers' licenses and STCW Certificates ratings STCW certificates

C. Charts, Log Books and Other Records

1. Chart of area of casualty 2. Bunkering records 3. Crew list with addresses 4. Name and addresses of previous master, chief mate, and chief engineer 5. Passenger list with addresses 6. Passenger boarding passes 7. Cabin assignments for passengers and crew 8. Terminal generated checklist and cargo loading/discharge data sheets 9. Vessel generated checklist and cargo loading/discharge data sheets 10. Liquid cargo data sheets 11. Analysis of cargo samples 12. Deck log (smooth and rough) 13. Cargo control room log (smooth and rough) 14. Engine log (smooth and rough) 15. Radio log (smooth and rough) 16. Boiler/main engine maintenance log 17. Original of course recorder printout at time of casualty 18. List of certificated lifeboatman 19. On-board crew conducted repair and maintenance records for one year

prior to accident 20. On-board repairs conducted by shore side company or personnel 21. Vessel repair/spare parts requisitions to company 22. Copy of last shipyard repair/survey specifications 23. Classification survey reports (annual and special and damage surveys) 24. Copy of the bridge record card 25. Port State and Flag State inspection reports (annual & drydocking) 26. Independent survey reports by insurance, towing, and/or fire/explosion

specialist. 27. Copy of Control Verification Examination Booklet 28. Passenger and crew medical log (ship's doctor/purser) 29. Shore Fire Department response records 30. Dangerous Stores Manifest 31. Trim and Stability Booklet

33

Page 233: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

D. Operating Procedures and Manuals 1. Oil transfer procedures for cargo and bunkers (fuel) 2. From the vessel's operation manuals:

main and emergency electrical power system description, engineroom control system, mooring gear on deck, cargo pumping and piping system description, ballast pumping and piping system description, steering and control system, boiler automation control system

3. List of safety manuals maintained on vessel 4. Company and vessel procedures for tank opening and entry 5. Oxygen Analyzing equipment specifications (model & type) and operating

manual 6. Combustible gas analyzer (model & type) and operating manual 7. Description of the vessel's planned maintenance system 8. Lifeboat and liferaft launching plan 9. Vessel Evacuation Plan 10. Copy of posted firefighting procedures for engine room and other spaces 11. Specific company orders to masters/chief engineers 12. Standing orders of ship master/chief mate/chief engineer. 13. Company training records for officers and crew 14. Company training and safety manuals 15. Station Bill 16. Company/vessel firefighting procedures 17. SAR data, including communication tapes from RCCs involved 18. Operating manual including stability control (MODU)

E. Miscellaneous 1. 8x10 pre-accident color photo of vessel (profile view) 2. General vessel characteristics 3. Bunker analysis from terminal and samples on vessel and at terminal 4. Fuel oil heating in tanks and through engineroom heaters 5. Type of tank gauging system 6. Previous accidents to this vessel, sister vessels, type and class 7. Loading Plan for last cargo(s) 8. Cargo regulations 9. Type of blowers (fans) used to vent tanks 10. Description of all temporary and permanent post-casualty repairs

11. List of the quantity and location of steel plating and internals (including piping) removed post-casualty prior to drydocking and at drydocking

12. Copy of next shipyard repair/survey specifications 13. Company organization chart

34

Page 234: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

Reference

Lecture  slides  were  taken  from  material  published  by  Stephenson  in:System  Safety  2000,  Van  Nostrand  Reinhold,  1991.

Page 235: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

Overview

MORT  chart  contains  approximately  1500  items  arranged  into  a  large/complex  fault  tree

Primarily  used  for  accident  investigation

Page 236: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

Purpose  of  MORT

To  provide  a  systematic  tool  to  aid  in  planning,  organizing,  and  conducting  an  in-­depth,  comprehensive  accident  investigation  to  identify  those  specific  that  are  LTA  and  need  to  be  corrected  to  prevent  the  accident  from  recurring.

Can  also  be  used  for  inspection,  audit,  or  appraisal  purposes.

Page 237: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

Symbols

Symbols  used  on  the  MORT  chart  are  fundamentally  the  same  as  those  used  on  other  analytical  trees  and  FTA.

Page 238: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

Event  Symbols

Page 239: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

Logic  Gates

Page 240: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

Transfers

Page 241: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

Abbreviations

LTA -­

DN -­

FT -­

HAP -­

JSA -­

CS&R -­

Page 242: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

Definition

Accepted  or  Assumed  Risk  -­ Very  specific  risk  that  has  been  identified,  analyzed,  quantified  to  the  maximum  practical  degree,  and  accepted  by  the  appropriate  level  of  management  after  proper  thought  and  evaluation.

Losses  from  Assumed  Risks are  normally  those  associated  with  earthquakes,  tornados,  hurricanes,  and  other  acts  of  nature.

Amelioration -­ Post-­accident  actions  such  as  medical  services,  fire  fighting,  rescue  efforts,  and  public  relations.

Page 243: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

Advantage

It  aids  the  accident  investigator  by  identifying  root  causes  of  the  accident.

Provides  a  systematic  method  of  evaluating  the  specific  control  and  management  factors  that  caused  or  contributed  to  the  accident.

Serve  as  planning  and  organizational  tool  for  the  collection  of  evidence  and  other  relevant  information.

Page 244: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

Disadvantage

Extremely  time  consuming  and  tedious  when  learning  about  and  first  using  the  MORT  chart.

This  approach  would  be  classified  as  overkill  for  most  accidents.

Page 245: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

Input  Requirements

In  a  nut-­shell,  extensive.  Detailed  information  regarding:Hardware

Facilities

Environment

Policies  &  Procedures

Personnel

Implementation  plans

Risk  assessment  programs

Project  documents,  etc...

Page 246: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

General  Approach

MORT  analysis  effort  begins  immediately  after  the  accident  occurs.

Performed  by  a  trained  investigator.

MORT  chart  is  used  as  a  working  tool  to  aid  in  the  gathering  and  storage  of  information.

General  method  for  working  through  the  chart  is  from  known  to  unknown.

The  top  of  the  chart  is  typically  addressed  very  early  in  the  investigation.

Page 247: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

Color  Coding

Page 248: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

Color  Coding  (Red)

Any  factor  or  event  found  to  be  LTA is  colored  redon  the  chart.

Should  be  addressed  in  the  accident  report  with  appropriate  recommendations  to  correct  the  deficiency.

Use  judiciously!  (Must  be  supported  by  facts)

Page 249: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

Color  Coding  (Green)

Any  factor  or  event  found  to  be  adequate is  colored  green on  the  chart.

Use  judiciously!  (Must  be  supported  by  facts)

Page 250: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

Color  Coding  (Black)

MORT  chart  is  designed  to  encompass  any  accident  situation,  therefore  not  all  parts  of  the  chart  may  be  relevant  to  the  particular  accident  that  is  being  investigated.

Any  factor  or  event  found  to  be  not  applicable is  color  coded  black (or  simply  crossed  out)  on  the  chart.

Page 251: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

Color  Coding  (Blue)

Indicates  that  the  block  has  been  examined,  but  insufficient  evidence  or  information  is  available  to  evaluate  the  block.  Suggests  to  collect  more  data.

Typically  these  are  colored  with  a  blue  dot  or  check  mark due  to  the  fact  that  they  should  change  color  prior  to  completing  the  investigation.

All  blue blocks  should  be  replaced  with  another  color  by  the  time  the  accident  investigation  is  complete.  But  this  may  not  always  be  the  case!  

Page 252: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

Since  this  is  a  working  document,  neatness  does  not  count  (to  a  point!).

Make  notes  on  the  chart  as  you  feel  necessary.

For  most  investigations,  analysis  tends  to  begin  at  the  specific  control  factors and  management  control  factors blocks  of  the  tree.

Page 253: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

Specific  Control  Factors

Tends  to  answer  questions  about  what  happened.

Addresses  the  accident  documentation  requirements.  

MORT  tens  to  answer  these  questions  in  more  detail  than  many  traditional  methods.

Page 254: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

Management  Control  Factors

Policy  LTA  -­ Not  typically  a  problem  with  major  organizations.  

Implementation  -­ Need  be  sure  the  policy  is  

Risk  Assessment  System  -­ Need  to  ensure  that  risks  are  properly  identified,  evaluated,  and  reported  to  management.  Oversights  or  omissions  can  also  be  a  problem  in  this  area.

Page 255: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

Top  Events

Page 256: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

Specific  Control  Factors

Page 257: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

Incident

Page 258: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

Barriers

Page 259: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

Targets

Page 260: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews

Management  Control  Factors

Page 261: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews
Page 262: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews
Page 263: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews
Page 264: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews
Page 265: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews
Page 266: Improving the Quality of Accident Investigation...of quality as consistency, depth, and time efficiency. To compare the quality of accident investigation methods, we conducted interviews