Risks in DP Shuttle Tanker Offloading Operations

Dr Haibo Chen 29 May 2013

CeSOS Highlights in 10 Years!

Main Contributors (1999 – 2013)

Dr. Haibo Chen Scandpower Inc., Lloyd’s Register Beijing, China Dr. Sverre Haver, Mr. Harald Kleppestø Mr. Kjell Larsen Statoil, Norway Capt. Helge Samuelsen Mr. Arve Lerstad Ship Modelling and Simulation Centre Trondheim, Norway

Prof. Torgeir Moan CeSOS, Dept. of Marine Technology NTNU, Trondheim, Norway Dr. Jan Erik Vinnem Preventor AS, Stavanger Mr. Kåre Breivik Sevan Marine Arendal, Norway

We would like to thank Mr. Kjell Helgøy from Teekay for information on DP safety and incidents, and Mr. Torbjørn Hals from Kongsberg Maritime for information on DP control system.

What are offloading concepts: late 1970s - Present

• Indirect offloading: – A small offloading platform – Submerged offloading systems such as

OLS, SAL, STL

• Direct offloading: – Tandem/alongside from ship-shaped units

(weather vane), typical 80 m distance – Normally with hawser and hose connections

What are risks: Collision and Oil Spill

• DP shuttle tanker position loss

• Towards installation: Collision risk – DP shuttle tanker collision had occurred on Emerald

FSU, Gryphon FPSO, Captain FPSO, Schiehallion FPSO, Norne FPSO, Njord FSU and several loading platforms in the past 15 years in the North Sea.

– The highest shuttle tanker speed at contact had reached 2.4 knots.

– The impact energy involved in these collision ranges from a few MJ to around 100 MJ. In one of the collision incident flare tower supporting structure located on the stern of FPSO suffered damage.

• Away from installation: Hose rupture and oil spill risk

Has oil spill happened before?

• On the 20th August 1980: One fatality on the bow of a shuttle tanker performing offloading. The hose was ruptured after hawser breakage and timely shutdown of crude offloading was not achieved. The crude oil pumped to the shuttle tanker bow area and caught fire. (station-keeping problems)

• On the 12th December 2007: Statfjord A loading buoy A rupture in the hose resulted in around 4,400 m3 of crude oil being pumped into the sea. (hose integrity problems)

Collision Frequency Model (1999 – 2003)

• Practical frequency model between DP shuttle tanker and FPSO – P(drive-off): Frequency of shuttle tanker drive-off forward which has the

potential to cause collision.

– P(Failure of recovery | drive-off): Failure probability of recovery initiated from shuttle tanker, given a shuttle tanker drive-off forward.

• Identify vulnerable situations for drive-off: surging and yawing • Recognize human recovery element and investigate how to improve

( ) ( ) ( )offdriverecoveryofFailurePoffdrivePcollisionP −×−=

The modelling work was summarized in Journal of Reliability Engineering and System Safety 84 (2004) 169-186.

Drive-off Forward: Causes and Frequency

• DP2 shuttle tanker generic drive-off forward frequency is around 0.02 to 0.05 per year (assume 24 hours operation, 1 offloading per week) – Incidents information from 1980s to 2009. Estimated DP shuttle tanker

offloading hours based on industrial statistical sources. – Assumptions related to DP2 vs. DP1 shuttle tankers

• Main causes

– Controllable pitch propeller, e.g. failure in pitch control, or feedback – Position reference systems, e.g. common mode failure of DARPS – Sensors, e.g. failure of one wind sensor, hawser tension sensor, or draught

sensor – DP software, e.g. software error, hidden bugs – Human operators, e.g. human errors by DP operator

Vulnerable Situation: Relative Motions in Offloading

• Tanker is vulnerable to drive-off when relative motions between FPSO and tanker is excessive.

Surging Yawing

FPSO Hawser

Wind, Wave, Current

Hose

FPSO Tanker

Tanker

80 m

60 m

Normal

Surging

FPSO Hawser

Wave, Current

Hose

Tanker

Tanker

Heading Deviation FPSO

Wind

Fishtailing

Normal

Yawing

Hs = 5.5 m loading / 4.5 m connection

SIMO Simulation Work (1999 – 2003)

• Objectives: – Predict how likely excessive surging and yawing will happen. – Identify effective measures to reduce these occurrences.

• Simulation work in cooperation with Statoil and Marintek

– SIMO time-domain simulation of FPSO and tanker – Vessel and environmental data provided by Statoil – Validation by full scale motion measurements on FPSO and tanker

• Recommendations :

– FPSO should minimize surge motion, and tanker should avoid following a moving target on FPSO stern for positioning.

– Operational coordination on FPSO and tanker for mean heading. – FPSO should minimize yaw motion.

The modelling work was summarized in Journal of Offshore Mechanics and Arctic Engineering, August 2004, Vo.126, 235-242.

DP Operator Recovery Actions

FPSO Stern

(1) (2) (3)

FPSO Stern FPSO Stern

Primary

Secondary

Time Constraint for Successful Intervention

To stop tanker within a short separation distance in drive-off scenario, recovery has to be initiated very early.

Big tanker mass vs. Short distance Propulsion response & effects

Separation Distance (m) 50 80 150

Time window for recovery (s) 37 53 81

How much time is needed by a human operator?

Modelling of Operator Intervention

Reference: Wickens’ model & Step-ladder model for human information processing stages

Information time: 0-Ta

Decision time: Ta-Td

Execution time: Td-T1

Drive-off Initiation

T1 0 Time

Action Initiation

DECISION

State Evaluation Task Formulation

INFORMATION

Observation Detection

Data Action EXECUTION

Muscle Command

First Abnormal Signal

Drive off Confirmation

Ta Td

Investigate Operator Action Time

• Incidents indicated that the action initiation time range from one to two minutes.

• Expert judgment by Simulator instructor – Simulator training with experienced tanker DP operators – An average 29 s in INFORMATION stage and 56 s in DECISION &

EXECUTION stages, and in total 85 s for action initiation

Information Stage Decision and Execution Stages

Time (sec) 0 - 10 10 - 20 20 - 30 30 - 50 0 - 20 20 - 30 30 - 60 60 - 90

No. out of 100 times Training 10 20 20 50 0 20 30 50

Probability 0.1 0.2 0.2 0.5 0.0 0.2 0.3 0.5

DP Operator Questionnaire Survey in 2002

• 10 captains and 7 DP officers • 16 Feedbacks of time estimation, 1093 tandem offloadings

experience behind. • Questionnaire formulated based on human action model

Reasonable Time Needed To Initiate Recovery Action

0

30

60

90

120

150

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Feedback ID

Tim

e (s

ec) Execution Time

Decision TimeInformation Time

Simulator Observation of DP Operator Emergency Intervention (2003 – 2006) • DP operator reaction time to drive-off in simulator • 66 records in 2003-2006. • Human action vs. probability curve fitted, representing best available

knowledge.

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0 20 40 60 80 100 120 140 160 180 200

Prob

abili

ty o

f Rea

ctio

n w

ithin

Tim

e

Time (s)

DP Operator Reaction Time in Drive-off Scenario (2003-2006, 66 simulator observations)

Direct Offloading Development (2007 – 2013) • Shuttle tanker offloading operations

from fixed or geostationary floating offshore installations in the North Sea.

• Offloading from Kristin Platform, 2007, 12-days

• Offloading from geostationary Sevan FPSOs in the Hummingbird/Chestnut fields in the North Sea

• Offloading from Sevan FPSO in the Goliat field in the Barents Sea (2015)

What’s new in direct offloading?

• Separation distance 250 m

• Weathervane with inherent safe heading philosophy – Heading pivot point – Min. 150 m no entry zone

• No hawser between

installation and shuttle tanker

150

m

250 m

Illustration drawing: The size and distance are not to scale.

Heading pivot point

hose

Wind

Fixed / Geostationary Offshore Installations

Risks in Offloading: Collision and Oil Spill

• DP shuttle tanker position loss!

• Towards installation = Collision risk

• Away from installation = Hose rupture and oil spill risk

150

m

250 m

Illustration drawing: The size and distance are not to scale.

Heading pivot point

hose

Wind

Fixed / Geostationary Offshore Installations

Zone 1Zone 2

Zone 1Zone 2

The 29th, 30th Int. Conference on Ocean, Offshore and Arctic Engineering (OMAE): - OMAE2010-21185 for collision risk analysis - OMAE2011-50344 for oil spill risk analysis

Drive-off with Potential for Collision

Wind / Wave

Drive-off into Collision Zone

Installation

Drive-off No Collision

Illustration drawing: The size and distance are not to scale.

Inherent Operational Safety Barriers

• Shuttle tanker positioning strategy = less than 1 hour drive-off collision risk exposure time vs. 20 hours.

• 250 m distance = Time window for recovery action by tanker DP operator is 3 minutes vs. 1 minute.

DP Operator Reaction Time in Drive-off Scenario(2003-2006, 66 simulator observations)

0.0

0.2

0.4

0.6

0.8

1.0

0 20 40 60 80 100 120 140 160 180 200

Time (s)

Prob

abili

ty o

f Rea

ctio

n w

ithin

Ti

me

Collision Frequency: Direct vs. Tandem Offloading

• The collision frequency in the direct offloading (6.43·10-5 per year) is much lower than the equivalent tandem offloading from FPSO (1.62·10-2 per year).

4.66E-05

1.20E-05

4.82E-06

1.17E-02

3.04E-03

1.21E-03

1.62E-04

4.05E-05

1.0E-06

1.0E-05

1.0E-04

1.0E-03

1.0E-02

1.0E-01

0-22 MJ 22-49 MJ 49-87 MJ 87-136 MJ >136 MJ

Freq

uenc

y per

Year

Direct of f loading f rom geostationary FPSO

Tandem of f loading f rom ship-shaped FPSO

negligible negligible

Collision Impact Energy (relevant to FPSO damage)

Premises: Offloading once per

week, 24 hours operational time.

Summary

• Collision frequency model with “human element” into equation. – Study of failure prone situation of drive-off: relative motions – Investigate human action time under emergency situations

• Simulator observation of human action and timing

• Direct offloading from geostationary FPSO/installation: from concept

to real operation.

References • Chen, H. and Moan, T.: "Human Intervention of Tanker Drive-off in Tandem Offloading Operation",

The 2nd International Conference on Human Factors in Ship Design and Operation, RINA HQ, London, UK, 2002

• Chen, H. and Moan, T.: "FPSO - Shuttle Tanker Collision Risk Reduction", OMAE2003-37108, Proceedings of the 22nd OMAE Conference, Cancun, Mexico, 2003

• Chen, H. and Moan, T.: "Probabilistic Modeling and Evaluation of Collision between Shuttle Tanker and FPSO in Tandem Offloading", Journal of Reliability Engineering and System Safety, Vol. 84 (2004) 169-186

• Chen, H., Moan, T., Haver, S., and Larsen, K.: "Prediction of Relative Motions and Probability of Contact between FPSO and Shuttle Tanker in Tandem Offloading Opera-tion", Journal of Offshore Mechanics and Arctic Engineering, Vol. 126, 235-242, August 2004.

• Chen, H., Moan, T. and Vinnem, J. E.: “Safety of shuttle tanker offshore loading operations with emphasis on the human barrier”, European Safety & Reliability Conference (ESREL2007), 24-27 June 2007.

• Chen, H. and Moan, T.: “Human element in the safety modeling of offshore marine operations”, Proc. of the Marine Operations Specialty Symposium 2008, Singapore, 5-7 March 2008.

• Chen, H., Lerstad, A. and Moan, T.: “Probabilistic Evaluation of Collision between DP Shuttle Tanker and Geostationary FPSO in Direct Offloading”, OMAE2010-21185, Proc. of the ASME 29th International Conference on Ocean, Offshore and Arctic Engineering, Shanghai, China, June 6-11, 2010.

• Chen, H., Moan, T., Breivik K., Lerstad, A.: “Analysis of Oil Spill Risk in DP Shuttle Tanker Direct Offloading Operations”, OMAE2011-50344, Proc. of the ASME 30th International Conference on Ocean, Offshore and Arctic Engineering, Rotterdam, The Netherlands, June 19-24, 2011.

For more information, please contact:

Dr. Haibo Chen Managing Director Scandpower Asia Operations

T +86 138-0132-0200 E hch@scandpower.com W www.scandpower.com w www.lr.org