1. Business Performance

8/12/2019 1. Business Performance

1/42

STUDY UNIT ONE

BUSINESS PERFORMANCE

1.1 Total Quality Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

1.2 Balanced Scorecard . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111.3 ISO Framework . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141.4 Forecasting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181.5 Project Management Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311.6 Business Process Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 361.7 Study Unit 1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41

This study unit addresses the analysis of performance by business organizations. A pervasive

consideration is the pursuit ofqualityin all aspects of the organizations activities. The importance of

quality management has been recognized by the International Organization for Standardization,

which has issued quality assurance standards. Also crucial to successful business performance is

effective planning. The aspect of planning covered in this study unit isforecasting, including a variety

of mostly quantitative forecasting models. The study unit continues withproject management, a topicof growing importance to all types of organizations in a technology-based society. The next subunit

describesbusiness process reengineeringand its implications, concluding with an approach to the

persistent problem of bottleneck management.

Core Concepts

TQM is the continuous pursuit of quality in every aspect of organizational activities through (1) aphilosophy of doing it right the first time, (2) employee training and empowerment, (3) promotionof teamwork, (4) improvement of processes, and (5) attention to satisfaction of internal andexternal customers.

The total cost of quality (conformance and nonconformance costs) should be minimized.

Benchmarking involves analysis and measurement of key outputs against those of the bestorganizations. This procedure also involves identifying the underlying key actions and causesthat contribute to the performance difference.

The balanced scorecard is a means of implementing an organizations strategy. It connects criticalsuccess factors determined in a strategic analysis to performance measures that may be financialor nonfinancial.

ISO 9000 is a series of voluntary generic standards for establishing and maintaining a qualitymanagement system (QMS) within a company. QMS principles are (1) customer focus,(2) leadership, (3) involvement of people, (4) use of a process approach, (5) a systems approachto management, (6) continual improvement, (7) a factual approach to decision making, and(8) mutually beneficial supplier relationships.

Forecasts are the basis for business plans. They attempt to answer questions about the outcomes

of events, the timing of events, or the future value of a statistic (e.g., sales). Project management techniques are designed to aid the planning and control of large-scale

projects having many interrelated activities. A project is a temporary undertaking with specifiedobjectives that often involves a cross-functional team and working outside customaryorganizational lines.

One approach to business process analysis is reengineering (also called business processreengineering). It involves process innovation and core process redesign. Instead of improvingexisting procedures, it finds new ways of doing things.

The theory of constraints (TOC) is a short-term approach to managing bottlenecks (bindingconstraints) in production and distribution processes.

1

Copyright 2008 Gleim Publications, Inc. and/or Gleim Internet, Inc. All rights reserved. Duplication prohibited. www.gleim.com
http://www.gleim.com/http://www.gleim.com/


2/42

1.1 TOTAL QUALITY MANAGEMENT

1. Deming management. The work of W. Edwards Deming was influential worldwide ingaining acceptance of the principle that quality improvement generates economicadvances. Better quality means lower costs, higher productivity, increased market share,more business, and more employment.

a. Demings14 pointssummarize his theories about improving quality: (1) have aconstant purpose, (2) adopt a new philosophy (wiser use of all resources), (3) give upon quality by inspection (quality should be built in), (4) avoid the constant search forlowest-cost suppliers (build long-term relationships), (5) seek continuousimprovement, (6) train everyone, (7) provide real leadership, (8) drive fear out of theworkplace (employees should not be afraid to ask questions), (9) promote teamwork,(10) avoid slogans and targets (instead, use process-control tools), (11) get rid ofnumerical quotas, (12) remove barriers that stifle pride in workmanship, (13) realizethat education and self-improvement are the key, and (14) make transformationeveryones job.

2. The fundamental areas of quality are process quality and product quality.

a. Process qualityis the effectiveness and efficiency of the organizations internaloperations. Product qualityis the conformance of the organizations output withcustomer expectations. Product quality is best analyzed frommultipleperspectives: (1) attributes of the product (performance, serviceability, durability,etc.), (2) customer satisfaction, (3) conformity with specifications, and (4) value(relation of quality and price).

1) One of the dimensions of quality isconformance, or how well a product and itscomponents meet applicable standards. Traditionally, conforming productswere deemed to be those with characteristics that lie within an acceptablespecifiedrange of valuesthat includes a target value. This approach alsoregards a certain percentage of defective (nonconforming) units as acceptable.

a) The traditional approach was superseded by thezero-defects (goalpost

conformance)approach. It seeks to eliminate nonconforming output.b) An extension of this approach is therobust quality (absolute quality

conformance)concept. Its goal is to reach the target value in everycase. The purpose is to eliminate the hidden quality costs that occurwhen output varies from the target even though the units are withinspecifications.

3. Thetotal quality managementapproach, also frequently calledtotal quality (TQ), was avery significant development.

a. TQMcan increase revenues and decrease costs substantially. The following areTQMscore principlesor critical factors:

1) Emphasis on thecustomer

a) Satisfaction of external customersb) Satisfaction of internal customersc) Requirements for external suppliersd) Requirements for internal suppliers

2) Continuous improvementas a never-ending process, not a destination.

3) Engaging every employeein the pursuit of total quality because avoidance ofdefects in products or services and satisfaction of external customers requiresthat all internal customers be satisfied.

2 SU 1: Business Performance



3/42

4. TQM is acomprehensive approach. It treats the pursuit of quality as abasicorganizational functionthat is as important as production or marketing. Accordingly,TQM is a strategic weapon. Because it affects every aspect of the organization, itpermeates the organizational culture. Thus, the cumulative effect of TQMs continuousimprovement process can attract and hold customers and cannot be duplicated bycompetitors.

a. Definition. TQM is the continuous pursuit of quality in every aspect of organizationalactivities through

1) A philosophy of doing it right the first time,2) Employee training and empowerment,3) Promotion of teamwork,4) Improvement of processes, and5) Attention to satisfaction of internal and external customers.

a) TQM emphasizes thesuppliers relationship with the customerandidentifies customer needs. It recognizes that everyone in a process is atsome time a customer or supplier of someone else, either inside oroutside of the organization.

b) Thus, TQM begins with external customer requirements, identifies internalcustomer-supplier relationships and requirements, and establishesrequirements for external suppliers.

b. Organizations tend to be vertically organized, but TQM requires stronghorizontallinkages.

5. Themanagement of qualityis not limited to quality management staff, engineers, andproduction personnel.

a. Given theorganization-wide scopeof TQM and of the internal audit activity, the roleof the internal auditors is to evaluate the entire quality function. In particular, theinternal audit activity should be well qualified toperform risk assessments andpromote continuous improvement ofcontrols. The personnel involved in the

technical improvements of processes may be unqualified with regard to riskmanagement and control issues.

1) The internal audit activity should perform procedures to provideassurancethatthebasic objectivesof TQM (customer satisfaction, continuous improvement,and promotion of teamwork) are reached.

2) TQM concepts also are applicable to theoperations of the internal auditactivity itself. For example, periodic internal assessments of those operationsmay include benchmarking of the internal audit activitys practices andperformance metrics against relevant best practices of the internal auditprofession (PA 1311-1).

6. Implementation of TQMcannot be accomplished by application of a formula, and theprocess is lengthy and difficult. The following phases are typical:

a. Establishing an executive-levelquality councilof senior managers with stronginvolvement by the CEO

b. Providingquality trainingprograms for senior managers

c. Conducting aquality auditto identify improvement opportunities and identifystrengths and weaknesses compared with competitors

d. Preparing agap analysisto determine what is necessary to bridge the gap betweenthe organization and the quality leaders in its industry and to establish a database forthe development of the strategic quality improvement plan

e. Developingstrategic quality improvement plansfor the short and long term

f. Conductingemployee communication and training programs

SU 1: Business Performance 3



4/42

g. Establishingquality teamsto ensure that goods and services conform tospecifications

1) Hierarchal structure is replaced with teams of people from different specialties.This change follows from an emphasis on empowering employees andteamwork. Employees should (a) have proper training, necessary information,

and the best tools; (b) be fully engaged in the decision process; and (c) receivefair compensation. If empowered employees are assembled in teams ofindividuals with the required skills, they will be more effective than peopleperforming their tasks separately in a rigid structure. Moreover, a team is avehicle for idea sharing, which results in process improvement.

h. Creating ameasurement systemand setting goals

i. Revisingcompensation, appraisal, and recognition systems

j. Reviewing and revisingthe entire effort periodically

7. Various managementprocesses, tools, and measuresshould be adopted.

a. Policy deploymentis the systematic planning of corporate objectives and the detailedways in which organizational subunits will approach the accomplishment of theirrelated objectives. The purpose is for the objectives of the organization, its subunits,and its employees to beconsistent.

b. Quality function deploymentensures that customer requirements are translated intodesign requirements at each step inproduct development. It is an umbrellaconcept most useful in an environment in which the Plan-Do-Check-Act (PDCA)Cycle(the Deming Cycle) is used at all levels.

1) PDCA is a management by fact or scientific method approach to continuousimprovement. PDCA creates aprocess-centered environmentbecause itinvolves (a) studying the current process, (b) collecting and analyzing data toidentify causes of problems, (c) planning for improvement, and (d) deciding howto measure improvement(Plan).

2) The plan is then implemented on a small scale if possible(Do).

3) The next step is to determine what happened(Check).4) If the experiment was successful, the plan is fully implemented(Act).

5) The cycle is then repeated using what was learned from the preceding cycle.

c. Kaizenis the Japanese word for the continuous pursuit of improvement in everyaspect of organizational operations.

1) For example, akaizen budgetprojects costs based on future improvements.The possibility of such improvements must be determined, and the cost ofimplementation and the savings must be estimated.

d. Employee involvementmeans training and empowering employees to harness theircreativity for problem solving. Quality control circlesare used to obtain input fromemployees and to locate the best perspective on problem solving.

e. Suppliers managementis the careful selection of suppliers and the cultivation oflong-term relationships based on the consistent ability to meet mutual expectations.

f. Competitive benchmarkinginvolves continuously evaluating the practices of the bestorganizations and adapting processes to incorporate the best of these practices.

g. Quality trainingfamiliarizes all employees with the means for preventing, detecting,and eliminating nonquality. The educational processes are tailored to each group.

h. Reward and recognitionfor quality improvement should be group oriented. Theyshould be based on quality measures.

i. Customer retentionis a vitally important measure of service quality. Loyalcustomers spend more, refer new customers, and are less costly to service.




5/42

j. Examples of nonfinancial measures of internal performanceare (1) manufacturingcycle efficiency (value-added production time total manufacturing lead or cycletime), (2) ratio of good output to total output, (3) defects per product line, (4) thehalf-life method (time required to reduce the defect ratio by 50%), (5) new productdevelopment time, and (6) shipments per period.

k. Examples ofnonfinancial measures of customer satisfaction are (1) percentage ofdefective goods shipped, (2) customer complaints, (3) customer response time,(4) on-time deliveries, (5) survey data, and (6) market share.

l. Statistical and otherquantitative methodsare used to identify quality problems.

1) TheTaguchi quality loss functionis a measure of the departure from absolutequality conformance (robust quality). It is based on the principle that qualitylosses occur even when items are within specified limits or tolerances. Thus,any variation from a quality target for a characteristic results in hidden qualitycosts.

a) Thebasic formulais

L= k(xT)2

If: L = the quality costs per unitk = a constant based on the entitys external failure cost experiencex = actual measure of the quality variableT = target value of the quality variable

i) Theestimate of the constantis

k= c d 2

If: c = the loss at a specification limitd = the difference between the target value and the specification limit

ii) The following is thegraphof a Taguchi loss function (the quality costat the target value is $0):

2) Statistical quality controlis a method of determining whether a process, e.g., a

shipment or production run of units, lies within acceptable limits.a) It also is used to determine whether production processes are out of

control.

b) Statistical control chartsare graphic aids for monitoring the status of anyprocess subject to random variations. Originally developed to control thequality of production processes, they also have applications of directinterest to auditors and accountants, for example, (1) unit cost ofproduction, (2) direct labor hours used, (3) ratio of actual expenses tobudgeted expenses, (4) number of calls by sales personnel, and (5) totalaccounts receivable.




6/42

i) The chart consists of three horizontal lines plotted on a horizontaltime scale.

ii) Thecenter linerepresents the average or mean value for theprocess being controlled.

iii) The other two lines are theupper control limit (UCL)and thelowercontrol limit (LCL).

iv) The processes are measured periodically, and the values are plottedon the chart (X). If the value falls within the control limits, no actionis taken. If the value falls outside the limits, the process isconsideredout of control, and an investigation is made forpossible corrective action. An advantage of the chart is that itmakestrendsand cyclesvisible.

v) EXAMPLE:

Unit Cost X Mays results areout of control.

$1.05 UCL

X$1.00X

$0.95 LCLMarch April May

vi) AP chartis based on an attribute (acceptable/not acceptable) ratherthan a measure of a variable. Specifically, it shows the percentageof defects in a sample.

vii) AC chartalso is an attribute control chart. It shows defects peritem.

viii) AnR chartshows the range of dispersion of a variable, such as sizeor weight.

ix) AnX-bar chartshows the sample mean for a variable.

c) Variations in aprocess parametermay have several causes.

i) Random variationsoccur by chance. Present in virtually allprocesses, they are not correctable because they will not repeatthemselves in the same manner. Excessively narrow control limitswill result in many investigations of what are simply randomfluctuations.

ii) Implementation deviationsoccur because of human or mechanicalfailure to achieve target results.

iii) Measurement variationsresult from errors in the measurements ofactual results.

iv) Model fluctuationscan be caused by errors in the formulation of adecision model.

v) Prediction variancesresult from errors in forecasting data used in adecision model.

3) APareto diagramis a bar chart that assists managers in what is commonlycalled 80:20 analysis.

a) The80:20 rule, formulated by management theorist Joseph M. Juran,states that 80% of all effects are the result of only 20% of all causes.

b) In the context of quality control, managers optimize their time by focusingtheir effort on the sources of most problems.




7/42

i) The independent variable, plotted on the X axis, is the factor selectedby the manager as the area of interest: department, time period,geographical location, etc. The frequency of occurrence of thedefect (dependent variable) is plotted on the Y axis.

ii) The occurrences of the independent variable are ranked from highestto lowest, allowing the manager to see at a glance which areas areof most concern.

iii) EXAMPLE: The chief administrative officer wants to know whichdepartments are generating the most travel vouchers that have tobe returned to the submitter because of incomplete documentation.

4) Ahistogramdisplays a continuum for the frequency distribution of the

independent variable.a) EXAMPLE: The CAO wants to know the amount of the travel

reimbursement delayed by a typical returned travel voucher.




8/42

5) Afishbone diagram(also called a cause-and-effect diagramor anIshikawadiagram) is a total quality management process improvement technique that isuseful in studyingcausation(why the actual and desired situations differ).

a) This format organizes the analysis of causation and helps to identifypossible interactions among causes.

b) The head of the skeleton represents the statement of the problem.c) The principal classifications of causes are represented by lines (bones)

drawn diagonally from the heavy horizontal line (the spine).

d) Smaller horizontal lines are added in their order of probability in eachclassification.

e) EXAMPLE:

8. Thecosts of qualitymust be assessed in terms of relative costs and benefits. Thus, anorganization should attempt tominimize its total cost of quality. Moreover,nonquantitative factors also must be considered. For example, an emphasis on quality

improves competitiveness, enhances employee expertise, and generates goodwill.a. Conformance costsinclude costs of prevention and costs of appraisal, which are

financial measures of internal performance.

1) Preventionattempts to avoid defective output. These costs include(a) preventive maintenance, (b) employee training, (c) review of equipmentdesign, and (d) evaluation of suppliers.

2) Appraisalembraces such activities as statistical quality control programs,inspection, and testing.

b. Nonconformance costsinclude internal failure costs (a financial measure of internalperformance) and external failure costs (a financial measure of customersatisfaction).

1) Internal failurecosts occur when defective products are detected beforeshipment. Examples are scrap, rework, tooling changes, and downtime.

2) The costs ofexternal failure, e.g., warranty costs, product liability costs, andloss of customer goodwill, arise when problems occur after shipment.

a) They are even more important inserviceenterprises than inmanufacturingenvironments. Faulty goods sometimes may bereworked or replaced to a customers satisfaction, but poor service tendsto result in a loss of customers.

3) Environmental costsalso are external failure costs, e.g., fines fornonadherence to environmental law and loss of customer goodwill.




9/42

c. Acost-of-quality reportincludes most costs related to quality. However, some itemsoften are not reported, for example, lost contribution margins from poor productquality. They are opportunity costs and are not usually recorded by the accountingsystem. The result is understatement of the costs of poor quality.

1) Lost contribution margins from reduced sales, market share, and sales prices

are external failure costs that also are not usually included in a cost-of-qualityreport.

2) An example of acost of quality report is presented below:

Prevention costs $35,000Appraisal costs 5,000Internal failure costs 17,500External failure costs 9,500

Total costs of quality $67,000

d. Quality cost indicesmeasure the cost of maintaining a given level of quality, forexample, total quality costs divided by direct labor costs.

1) EXAMPLE: To continue the example, if total direct labor costs were $201,000,the quality cost index for the money was 33.3% ($67,000 $201,000).

9. Quality and productivitydo not necessarily have an inverse relationship. Therobustqualityview is that improving quality and reducing costs in each category may be possibleif the most efficient prevention methods are applied.

a. For example, selection of a supplier meeting high quality standards regarding defectrates and delivery times may drive down not only failure costs, but also theprevention and appraisal costs incurred when supplier performance was less reliable.

10. Management of timeis related to TQM.

a. Product development timeis a crucial factor in the competitive equation. A companythat is first in the market with a new product has obvious advantages.

1) Reducing development time is also important becauseproduct life cyclesare

becoming shorter.2) Companies need to respond quickly and flexibly to new technology, changes in

consumer tastes, and competitive challenges.

b. One financial measure of product development isbreakeven time. It is the time frommanagement approval of the project to the time when thecumulative present valueof cash inflows equals the cumulative present value of cash outflows.

1) The most popular method of determining breakeven time calculates the timerequired for thepresent value of the cumulative cash flows to equal zero.

a) An alternative that results in a longer breakeven time is to consider thetime required for the present value of the cumulative cash inflows to equalthe present value of all the expected future cash outflows.

c. Customer-response timeis the delay from placement of an order to delivery of thegood or service. Response time is a function oftime drivers. A change in a timedriver causes a change in the time required for an activity. Such changes reflectuncertainty about arrivals of customers in the queueand bottlenecks(points atwhich capacity is reached or exceeded).

1) Customer response time consists oforder receipt time(delay between thecustomers placement of an order and its receipt by the production facility),manufacturing lead or cycle time(delay from the orders receipt by theproduction facility to its completion), and order delivery time.

2) Manufacturing lead or cycle (throughput) time equalsorder waiting timeplusmanufacturing time.




10/42

11. Benchmarkingis a primary tool used in TQM. It is a means of helping organizations withproductivity management and business process analysis.

a. Benchmarking involves analysis and measurement of key outputs against those of thebest organizations. This procedure also involves identifying the underlying keyactions and causes that contribute to the performance difference.

1) Best practicesare recognized by authorities in the field and by customers forgenerating outstanding results. They are generally innovative technically or intheir management of human resources.

2) Benchmarking is an ongoing process that requiresquantitative and qualitativemeasurementof the difference between the performance of an activity and theperformance by the benchmark. This entity need not be a competitor.

b. The following are kinds of benchmarking:

1) Competitive benchmarkingstudies an organization in the same industry.

2) Process (function) benchmarkingstudies operations of organizations withsimilar processes regardless of industry. Thus, the benchmark need not be acompetitor or even a similar entity.

a) This method may introduce new ideas that provide a significant competitiveadvantage.

3) Strategic benchmarkingis a search for successful competitive strategies.

4) Internal benchmarkingis the application of best practices in one part of theorganization to its other parts.

c. The first phase in the benchmarking process is toselect and prioritizebenchmarking projects.

1) An organization must understand itscritical success factorsand businessenvironmentto identifykey business processes and drivers and to developparameters defining what processes to benchmark. Thecriteriafor selectingwhat to benchmark relate to the reasons for the existence of a process and itsimportance to the entitys mission, values, and strategy. These reasons relatein large part tosatisfaction of end-user or customer needs.

d. The next phase is toorganize benchmarking teams. A team organization isappropriate. It permits an equitable division of labor, participation by thoseresponsible for implementing changes, and inclusion of a variety of functionalexpertise and work experience.

1) Team membersshould have (a) knowledge of the function to be benchmarked,(b) respected positions in the organization, (c) good communication skills,(d) teaming skills, (e) motivation to innovate and to support cross-functionalproblem solving, and (f) project management skills.

2) The team must thoroughly investigate and document the organizationsinternalprocesses. The organization is a series of processes, not a fixed structure.

a) Aprocessis a network of related and independent activities joined by theiroutputs. One way to determine the primary characteristics of a process isto trace the path a request for a product or service takes through theorganization.

b) The team must develop afamily of measures that are true indicators ofprocess performance. It also must develop aprocess taxonomy, that is,a set of process elements, measures, and phrases that describes theprocess to be benchmarked.




11/42

c) The development ofkey indicators for performance measurement in abenchmarking context is an extension of the basic evaluative functionofinternal auditors. Internal auditors evaluate governance, riskmanagement, and control processes. Evaluation requires establishmentofadequate criteriaby management. In the absence of these criteria,internal auditors must work with management to develop appropriate

evaluation criteria(Standard 2120.A4).e. Researching andidentifying best-in-class performanceis often the most difficult

phase. The critical steps are

1) Setting up databases

2) Choosing information-gathering methods (internal sources, external publicdomain sources, and original research)

3) Formatting questionnaires (lists of questions prepared in advance), and

4) Selecting benchmarking partners.

f. Data analysisidentifies performance gaps, obtains an understanding of the reasonsthey exist, and prioritizes the key activities that will facilitate the behavioral andprocess changes needed. Sophisticated statistical and other methods may be

needed when the study involves many variables, testing of assumptions, or quantifiedresults.

g. Leadershipis most important in theimplementation phasebecause the team mustbe able to justify its recommendations. Moreover, the process improvement teamsmust manage the implementation of approved changes.

1.2 BALANCED SCORECARD

1. Internal auditors may conduct aperformance auditto provide assurance about theorganizationskey performance indicators. They also may conduct aconsultingengagementto design such a performance measurement system.

a. Effectivemanagement controlrequiresperformance measurementand feedback.This process affects allocation of resources to organizational subunits. It also affectsdecisions about managers compensation, advancement, and future assignments.

1) Furthermore, evaluating their performance serves to motivate managers tooptimize the measuresin the performance evaluation model. However, thatmodel may be inconsistent with the organizations model for managerialdecision making.

a) To achieve consistency, themodels should be synchronized. Forexample, if senior management wishes to maximize results over the longterm, subordinates should be evaluated over the long term.

b) Unfortunately, information systems seldom provide information on theoutcomes of individual decisions, and senior managers may be unaware

of desirable options not chosen. Moreover, performance feedbackusually applies to specific responsibility centers for specific periods. Theresult is that use of the same criteria for decision makingandmanagerial evaluationmay be difficult.




12/42


13/42

4) The scorecard should permit a determination of whether certainobjectivesarebeing achieved at the expense of others. For example, reduced spending oncustomer service may improve short-term financial results at a significant costsuggested by a decline in customer satisfaction.

e. Problems in implementationof the balanced scorecard approach include

1) Using too many measures, with a consequent loss of focus on CSFs2) Failing to evaluate personnel on nonfinancial as well as financial measures

3) Including measures that will not have long-term financial benefits

4) Not understanding that subjective measures (such as customer satisfaction) areimprecise

5) Trying to achieve improvements in all areas at all times

6) Not observing when current nonfinancial measures no longer relate to ultimatefinancial success

f. A typical balanced scorecard includes measures in four categories:

1) Financial

a) TheCSFsmay be sales, fair value of the firms stock, profits, and liquidity.

b) Measuresmay include (1) sales, (2) projected sales, (3) accuracy of salesprojections, (4) new product sales, (5) stock prices, (6) operatingearnings, (7) earnings trend, (8) revenue growth, (9) gross marginpercentage, (10) cost reductions, (11) economic value added (EVA),(12) return on investment (or any of its variants), (13) cash flow coverageand trends, (14) turnover (assets, receivables, and inventory), and(15) interest coverage.

2) Customer

a) TheCSFsmay be (1) customer satisfaction, (2) customer retention rate,(3) dealer and distributor relationships, (4) marketing and sellingperformance, (5) prompt delivery, and (6) quality.

b) Measuresmay include (1) returns, (2) complaints, (3) survey results,(4) coverage and strength of distribution channels, (5) market researchresults, (6) training of marketing people, (7) sales trends, (8) market shareand its trend, (9) on-time delivery rate, (10) service response time andeffectiveness, and (11) warranty expense.

3) Internal business processes

a) TheCSFsmay be (1) quality, (2) productivity (an input-output relationship),(3) flexibility of response to changing conditions, (4) operating readiness,and (5) safety.

b) Measuresmay include (1) rate of defects, (2) amounts of scrap andrework, (3) returns, (4) survey results, (5) field service reports,(6) warranty costs, (7) vendor defect rate, (8) cycle (lead) time, (9) labor

and machine efficiency, (10) setup time, (11) scheduling effectiveness,(12) downtime, (13) capacity usage, (14) maintenance, and (15) accidentsand their results.




14/42

4) Learning, growth, and innovation

a) TheCSFsmay be (1) development of new products, (2) promptness oftheir introduction, (3) human resource development, (4) morale, and(5) competence of the workforce.

b) Measuresmay include (1) new products marketed, (2) amount of design

changes, (3) patents and copyrights registered, (4) R&D personnelqualifications, (5) actual versus planned shipping dates, (6) hours oftraining, (7) skill set levels attained, (8) personnel turnover, (9) personnelcomplaints and survey results, (10) financial and operating results,(11) technological capabilities, (12) organizational learning, and(13) industry leadership.

g. EXAMPLE:

1) Eachobjectiveis associated with one or more measuresthat permit theorganization to gauge progress toward the objective.

2) Achievement of the objectives in eachperspectivemakes it possible to achievethe organizations objectives.

Balanced ScorecardFinancial Perspective

Objective: Increase shareholder value Measures: Increase in ordinary (common)stock price

Reliability of dividend payment

Customer Perspective

Objective: Increase customer satisfaction Measures: Greater market shareHigher customer retention ratePositive responses to surveys

Internal Business Process Perspective

Objective: Improve product quality Measures: Achievement of zero defects

Objective: Improve internal processes Measures: Reduction in delivery cycle timeSmaller cost variances

Learning, Growth, and InnovationPerspective

Objective: Increase employee confidence Measures: Number of suggestions toimprove processes

Positive responses to surveys

Objective: Increase employee competence Measures: Attendance at internal and externaltraining seminars

1.3 ISO FRAMEWORK

1. In 1987, theInternational Organization for Standardization (ISO) introducedISO 9000, afamily of 11 voluntary standards and technical reports that provide guidance forestablishing and maintaining aquality management system (QMS). The ISOs rulesspecify that its standards be revised every 5 years in light of technological and marketdevelopments. (NOTE: ISO is not an acronym. It means equal, suggesting that entitiescertified under ISO 9001:2000 have equal quality.)

a. The current standards(ISO 9000:2000)were issued in December 2000. For specificand up-to-date information, see the ISOs website (www.iso.org).




15/42


16/42

c. Management responsibility. Management (1) reviews the quality policy,(2) analyzes data about QMS performance, and (3) assesses opportunities forimprovement and the need for change.

1) Management ensures that systems exist to determine and satisfy customerrequirements.

d. Resource management. The resources needed to improve the QMS and satisfycustomer requirements must be provided.

1) Human resources(employees) should be qualified and trained for their specificjobs.

2) Theinfrastructure(physical resources needed for quality output) and the workenvironmentshould be sufficient.

e. Product realization processesresult in products or services received by customers.These processes must beplanned and controlled. Issues are (1) means of control,(2) objectives, (3) documentation and records needed, and (4) acceptance criteria.

1) Design and developmentalso must be planned and controlled.

2) Purchasingevaluates suppliers of materials and services, and the entitys

materials and products must be clearly identified at all times. Moreover,customer propertymust be protected.

3) Product qualitymust be preserved during handling, storage, and delivery.

f. Measurement, analysis, and improvement. The entity must have processes for(1) inspection, (2) testing, (3) measurement, (4) analysis, and (5) improvement.

1) Thus,customer satisfactionmust be measured.

2) Substandard materials and productsmust be controlled so that they do notreach customers.

3) Internal auditsmeasure and monitor QMS performance.

4) All employees should usecorrective and preventive action systems.

5) Work instructionsmust identify measuring and monitoring requirements.

5. Some companies are obtaining ISO 9000 certification because of fear that theEuropeanUnionwill require compliance with the standards in an attempt to restrict imports.

a. The standards are not yet mandatory except amongregulated products(for whichhealth and safety are concerns), such as medical devices, telecommunicationsequipment, and gas appliances.

b. Some customers demand that suppliers register.

c. ISO 9000 registration may be a key to remaining competitive. It makes customersmore comfortable with suppliers products and services.

d. Many companies implementing the standards uncoverinternal process and qualityimprovementsas a result. ISO 9000 forces companies to share information andunderstand who internal customers and users are.

6. Aregistrar, or external auditor, must be selected. Registrars are usually specialists withincertain Standard Industrial Classification (SIC) codes. Certification by a registrar avoids theneed for each customer to audit a supplier.

a. Following an on-site visit, the registrar, if convinced that a quality system conforms tothe selected standard, issues a certificate describing the scope of the registration.Registration is usually granted for a 3-year period.

b. Some companies have preliminary audits by official registrars.

c. All employees are subject to being audited. They must have the ability to say whatthey do and to demonstrate that they do what they say.




17/42

7. The ISO also has issued a set ofenvironmental standardsknown asISO 14000. Thesestandards are comparable in purpose to ISO 9000 but concernenvironmental qualitysystems. Although they have not been as widely adopted as the ISO 9000 standards, theymay become a necessity for conducting international business.

a. ISO 14000 establishes internationally recognized standards that will diminishbarriers

to tradeand make it easier to do business across borders.b. Some companies feel that adherence to ISO 14000 standards will reducemonitoringand inspectionby regulatory agencies.

c. A survey of managers found that failure to obtain ISO 14000 certification couldconstitute a potentialnontariff trade barrier because customers will require it.

d. At present, the main benefit of adopting ISO 14000 standards is internal. Companieslearn how well theirenvironmental management systemoperates relative to thoseof other companies.

e. Some companies have decided to seek ISO 14000 certification because they foundthat ISO 9000 was beneficial.

f. Some European countries already have environmental systems standards in place,and how these single-country standards will mesh with ISO 14000 is not clear.

However, individual countries standards are typically more strict.g. Some are concerned that regulators may usevoluntary ISO audits or self-audits as

a basis for punitive action. To allay these fears in the U.S., the EnvironmentalProtection Agency has issued new audit guidelines that are intended to avoid suchself-incrimination.

8. The scope ofISO 19011:2002extends to (a) the principles of auditing, (b) managing auditprograms, (c) conducting QMS audits and environmental management system audits, and(d) the competence of QMS and environmental management system auditors.

a. It applies to all entities that must perform internal or external audits of QMSs orenvironmental management systems or manage an audit program.

b. ISO 19011 may apply to other types of audits if due consideration is given to

identifying the competencies required of the auditors.9. ISO 10012:2003is a generic standard. It addresses the management of measurement

processes and confirmation of measuring equipment used to support compliance withrequired measures.

a. It states quality management requirements of ameasurement management system(MMS)that can be used as part of the overall management system.

b. It is not to be used as a requirement for demonstrating conformance with otherstandards. Interested parties may agree to use ISO 10012:2003 as an input forsatisfying MMS requirements in certification activities. However, other standardsapply to specific elements affecting measurement results, e.g., details ofmeasurement methods, competence of personnel, or comparisons amonglaboratories.




18/42

1.4 FORECASTING

1. Forecasts are the basis forbusiness plans, including budgets. They attempt to answerquestions about theoutcomesof events (e.g., the effect of a war involving a producer of oilon the oil market), thetimingof events (e.g., when will unemployment fall), or the futurevalue of a statistic(e.g., sales). In addition to intuition (informed judgment), many

quantitative methods are useful in projecting the future from past experience.a. Examples of forecasts include sales projections, inventory demand, cash flow, and

future capital needs.

1) Mostmodelsare used in the forecasting process. They are used to makedecisions that optimize future results.

2) Thereliabilityof the forecast should be determined before using it. No objectivemethod can determine the reliability of judgmental forecasts. When quantitativemethods are used, however, measurement of reliability is usually possible, e.g.,by calculating the standard error of the estimate.

2. Correlation analysisis used to measure the strength of the linear relationship between twoor more random variables. Correlation between two variables can be seen by plotting their

values on a single graph to form a scatter diagram. If the points tend to form a straightline, correlation is high. Otherwise, correlation is low. Correlation measures only linearrelationships.

a. If the points form a curve, several possibilities exist.

1) A linear relationship (a straight line) may be used to approximate a portion of thecurve.

2) A linear relationship exists between some other function of the independentvariable x (e.g., log x) and the dependent variable y.

3) No relationship exists.

b. Thecoefficient of correlation (r) measures the relative strength of the linearrelationship. It has the following properties:

1) The magnitude of r is independent of the scales of measurement of x and y.2) 1.0 < r < 1.0

a) A value of 1.0 indicates a perfectly inverse linear relationship betweenx and y.

b) A value of zero indicates no linear relationship between x and y.

c) A value of +1.0 indicates a direct relationship between x and y.

c. Scatter diagramsmay be used to demonstrate correlations. Each observationcreates a dot that pairs the x and y values. The collinearity and slope of theseobservations are related to the coefficient of correlation by the above-stated rules.




19/42

d. Thecoefficient of determination (r 2), or the coefficient of correlation squared, maybe interpreted as the proportion of the total variation in y that is explained oraccounted for by the regression equation.

1) It is approximately equal to 1 minus the quotient of the unexplained variationdivided by the total variation when the sample is large. The following is the

formula:

If: r 2 = the coefficient of determination = summationyi = an actual data point = a point on the regression line calculated from the sample

linear regression equationy = the mean of the observed data points

2) EXAMPLE: The assertion that new car sales are a function of disposableincome with a coefficient of correlation of .8 is equivalent to stating that 64%

(.8) of the variation of new car sales (from average new car sales) can beexplained by the variation in disposable income (from average disposableincome).

3) Because r 2 increases as the number of independent variablesincreases,regardless of whether the additional variables are actually correlated with thedependent variable, r 2 may be adjusted (reduced) to allow for this effect. If k isthe number of independent variables and n is the number of observations, theformula for adjusted r 2 is

3. Regression (least squares) analysisextends correlation to find an equation for the linearrelationship among variables. The behavior of the dependent variable is explained in termsof one or more independent variables. Thus, regression analysis determines functionalrelationships among quantitative variables.

a. Simple regressionhas one independent variable, and multiple regressionhas morethan one.

1) EXAMPLE: A dependent variable such as sales is dependent on advertising,consumer income, availability of substitutes, and other independent variables.

2) Multicollinearityis the condition in which two or more independent variables arestrongly correlated. The effect is greater uncertainty regarding the coefficient ofthe variables; that is, their standard errors increase. Multicollinearity is aconcern in multiple regression.

b. Regression analysis is used to findtrend lines in business data such as sales orcosts(time series analysis or trend analysis)and to develop models based on theassociation of variables (cross-sectional analysis, a method that is not time relatedas is trend analysis). Examples are

1) Trend in product sales2) Trend in overhead as a percentage of sales3) Relationship of direct labor hours to variable overhead4) Relationship of direct material usage to accounts payable




20/42

c. Somereasonable basisshould exist for expecting the variables to be related.

1) If they are obviously independent, any association found by regression is merecoincidence.

2) Regression does not determine causality, however. Although x and y movetogether, the apparent relationship may be caused by some other factor.

a) EXAMPLE: A strong negative correlation exists between the decline ininfant mortality and the increase in the number of senior citizens. Bothare likely to be primarily caused by a factor such as better medical care.

3) The statistical relationships revealed by regression and correlation analysis arevalidonlyfor the range of the data in the sample.

d. Thesimple regression equation is

If: y = the dependent variablea = the y-axis intercept (the fixed cost in cost functions)b = the slope of the regression line (the variable portion of the

total cost in cost functions)

x = the independent variablee = the error term

1) Assumptions of the modelare that

a) For each value of x, there is a distribution of values of y. The means ofthese distributions form a straight line. Hence, x and y are linearlyrelated.

b) The error term (e) is normally distributed with a mean or expected valueequal to zero.

i) The y-intercept (a) and the slope of the regression line (b) also havenormal distributions.

c) Errors in successive observations are statistically independent.

i) Thus, the estimators are unbiased.

ii) Autocorrelation (serial correlation)occurs when the observationsare not independent; in other words, later observations may bedependent on earlier ones.

d) The distribution of y around the regression line is constant for differentvalues of x.

i) Thus, the observations are characterized byhomoscedasticityorconstant variance. The deviation of points from the regression linedoes not vary significantly with a change in the size of theindependent variable.

Heteroscedasticityis the condition in which the variance ofthe error term is not constant.




21/42

ii) Graphically, the model is represented by a series of normaldistributions (subpopulations of y) around the regression line. Asnoted on the previous page, these subpopulations have the samevariance.

e. From linear algebra, theequation for a straight line may be stated as follows:

If: a = the y-axis intercept

b = the slope of the line

1) Regression analysis uses themethod of least squares, which minimizes thesum of the squares of the vertical distance between each observation point andthe regression line.

2) EXAMPLE: Observations are collected on advertising expenditures and annualsales for a firm.

Sales ($000,000s) Advertising ($000s)28 7114 3119 5021 60

16 35a) According to the regression equation that results from using least squares

computations, expected sales equal 4.2 plus .31 times the advertisingexpenditure.

b) The observations are graphed as follows:

f. Regression analysis is particularly valuable forbudgetingand cost accountingpurposes. For instance, it is almost a necessity for computing the fixed and variableportions of mixed costs for flexible budgeting.




22/42

g. The followingequationscan be used to determine the equation for the least squaresregression line (the equation for the line is in the form of y= a+ bx):

1) EXAMPLE: The use of the two equations can be illustrated with the followingdata based on a set of six paired observations (n = 6):

y x$ 6 2

7 35 24 18 36 2

y= $36 x= 13

xy

x

2

6 2 = 12 47 3 = 21 95 2 = 10 44 1 = 4 18 3 = 24 96 2 = 12 4

83 31

a) Substituting into the two equations gives

36 = 6a+ 13b83 = 13a+ 31b

b) Solving simultaneously for the two unknowns,

1116 = 186a + 403b1079 = 169a + 403b

37 = 17a

c) Thus,a= 2.176. Solving for b in the second original equation gives

83 = 13(2.176) + 31b83 = 28.288 + 31b

31b= 54.712b= 1.765

d) Hence, future costs can be predicted by using the following equation:

e) Alternative formulas that are ordinarily simpler to use are given below:

i) The slope may be expressed as

ii) The value of the y-intercept may be expressed as




23/42

h. Discriminant analysisis a variation of regression analysis in which independentvariables are categorical; i.e., each observation is assigned a category.

1) EXAMPLES:

a) High, medium, lowb) Good, bad

c) Male, femalei. Ascattergraphis a diagram of a set of paired observations. This method entails

drawing a line to approximate the relationship suggested by the data.

1) The scattergraph method suffers from being reliant on the judgment of theperson visualizing the line.

2) Once the line has been drawn, the equation of the line can be determined fromany two points on the line.

j. Thehigh-low methodis used to generate a regression line by basing the equation ononly the highest and lowest of a series of observations.

1) EXAMPLE: A regression equation covering electricity costs could be developedby using only the high-cost month and the low-cost month. If costs were $400

in April when production was 800 machine hours and $600 in September whenproduction was 1,300 hours, the equation would be determined as follows:

High month $600 for 1,300 hoursLow month 400 for 800 hoursIncrease $200 500 hours

Because costs increased $200 for 500 additional hours, the variable cost is$.40 per machine hour. For the low month, the total variable portion of thatmonthly cost is $320 ($.40 800 hours). Given that the total cost is $400 and$320 is variable, the remaining $80 must be a fixed cost. The regressionequation isy = 80 + .4x.

2) The major criticism of the high-low method is that the high and low points may be

abnormalities not representative of normal events.4. Time seriesor trend analysisrelies on past experience. Changes in the value of a

variable (e.g., unit sales of a product) over time may have several possible components.

a. In time series analysis, the dependent variable is regressed on time (the independentvariable).

b. Thesecular trendis the long-term change that occurs in a series. It is represented bya straight line or curve on a graph.

c. Seasonal variationsare common in many businesses. A variety of analysis methodsincludes seasonal variations in a forecasting model, but most methods make use of aseasonal index.

d. Cyclical fluctuationsare variations in the level of activity in business periods.

Although some of these fluctuations are beyond the control of the firm, they need tobe considered in forecasting. They are usually incorporated as index numbers.

e. Irregularor random variablesare any variations not included in the categoriesabove. Business can be affected by random happenings (e.g., weather, strikes, fires,etc.).




24/42

f. Thepercentage-of-salesmethod is the most widely used for sales forecasting. Itadjusts the current level of sales by a specified percentage increase or decrease.This method is a form of trend analysis that is convenient and easy to apply andintuitively appealing to managers. It is also useful for developingpro forma financialstatementsby estimating items that vary directly with sales as percentages ofexpected sales.

1) This method is based on the assumptions that most items directly correlate withsales and that current levels of all assets are optimal for current sales.

5. Exponential smoothingis a technique used to level or smooth variations encountered in aforecast. This technique also adapts the forecast to changes as they occur.

a. The simplest form of smoothing is themoving average, in which each forecast isbased on a fixed number of prior observations. Exponential smoothing is similar tothe moving average.

b. Exponential means that greater weight is placed on the most recent data, with theweights of all data falling off exponentially as the data age. The selection ofalpha(), the smoothing factor, is important because a high alpha places more weight onrecent data.

c. The equation for the forecast (F) for period t + 1 is

If: xt = the observation for period tt = the most recent period

= the smoothing factor (0 1)Ft = the forecast for period t

1) This method weights the observation for period t by and the forecast forperiod t by (1 ).

6. Learning curvesreflect the increased rate at which people perform tasks as they gainexperience. The time required to perform a given task becomes progressively shorter, but

this technique is applicable only to the early stages of production or of any new task.a. Ordinarily, the curve is expressed as a percentage of reduced time to complete a task

for each doubling of cumulative production. Research has shown learning curvepercentages to be between 60% and 80%. In other words, the time required isreduced by 20% to 40% each time cumulative production is doubled, with 20% beingcommon.

1) One common assumption made in a learning curve model is that thecumulativeaverage time per unitis reduced by a certain percentage each time productiondoubles.

a) The alternative assumption is thatincremental unit time(time to producethe last unit) is reduced when production doubles.

2) EXAMPLE: An 80% learning curve would result in the following performance forthe lots shown, when run in sequence (top to bottom).

Cumulative CumulativeNumber of Tasks Average Time per Unit

100 3.0200 2.4 (3.0 80%)400 1.92 (2.4 80%)800 1.536 (1.92 80%)

1,600 1.228 (1.536 80%)




25/42

b. Graphical Presentation

c. If the average time for 100 units in the example were 3 minutes per unit, the total timewould be 300 minutes. At an average time of 2.4 minutes for 200 units, the total timewould be 480 minutes. In other words, the additional 100 units required only180 minutes (480 300), or 1.8 minutes per unit.

7. Simulationis a technique for experimenting with logical and mathematical models using acomputer.

a. Despite the power of mathematics, many problems cannot be solved by known

analytical methods because of the behavior of the variables and the complexity oftheir interactions, e.g.,

1) Corporate planning models2) Financial planning models3) New product marketing models4) Queuing system simulations5) Inventory control simulations

b. Experimentationis neither new nor uncommon in business. Building a mockup of anew automobile, having one department try out new accounting procedures, and test-marketing a new product are all forms of experimentation. In effect, experimentationis organized trial and error using a model of the real world to obtain information priorto full implementation.

c. Modelscan be classified as either physical or abstract.

1) Physical models include automobile mockups, airplane models used for wind-tunnel tests, and breadboard models of electronic circuits.

2) Abstract models may be pictorial (architectural plans), verbal (a proposedprocedure), or logical-mathematical. Experimentation with logical-mathematicalmodels can involve many time-consuming calculations. Computers haveeliminated much of this costly drudgery and have led to the growing interest insimulation for management.

d. Thesimulation procedurehas five steps.

1) Define the objectives. The objectives serve as guidelines for all that follows.The objectives may be to aid in the understanding of an existing system (e.g.,

an inventory system with rising costs) or to explore alternatives (e.g., the effectof investments on the firms financial structure). A third type of objective isestimating the behavior of some new system, such as a production line.

2) Formulate the model. The variables to be included, their individual behavior,and their interrelationships must be defined in precise logical-mathematicalterms. The objectives of the simulation serve as guidelines in deciding whichfactors are relevant.

3) Validate the model. Some assurance is needed that the results of theexperiment will be realistic. This assurance requires validation of the model --often using historical data. If the model gives results equivalent to what actuallyhappened, the model is historically valid. Some risk remains, however, thatchanges could make the model invalid for the future.




26/42

4) Design the experiment. Experimentation is sampling the operation of asystem. For example, if a particular policy is simulated on an inventory modelfor two years, the results are a single sample. With replication, the sample sizecan be increased and the confidence level raised. The number of runs to bemade, length of each run, measurements to be made, and methods foranalyzing the results are all part of the design of the experiment.

5) Conduct the simulation -- evaluate results. The simulation should beconducted with care. The results are analyzed using appropriate statisticalmethods.

e. TheMonte Carlo techniqueis often used in simulation to generate the individualvalues for a random variable. A random number generator is used to producenumbers with a uniform probability distribution (equal likelihoods of occurrence). Thesecond step is to transform these numbers into values consistent with the desireddistribution.

1) The performance of a quantitative model may be investigated by randomlyselecting values for each of the variables in the model (based on the probabilitydistribution of each variable) and then calculating the value of the solution. Ifthis process is performed a large number of times, the distribution of resultsfrom the model will be obtained.

2) EXAMPLE: A new marketing model includes a factor for a competitorsintroduction of a similar product within 1 year. Management estimates a 50%chance that this event will happen. For each simulation, this factor must bedetermined, perhaps by flipping a coin or by putting two numbers in a hat andselecting one number. Random numbers between 0 and 1 could begenerated. Numbers under .5 would signify introduction of a similar product;numbers over .5 would indicate the nonoccurrence of this event.

f. Theadvantages of simulationare as follows:

1) Time can be compressed. A corporate planning model can show the results of apolicy for 5 years into the future, using only minutes of computer time.

2) Alternative policies can be explored. With simulations, managers can ask what-ifquestions to explore possible policies, providing management with a powerfulnew planning tool.

3) Complex systems can be analyzed. In many cases, simulation is the onlypossible quantitative method for analyzing a complex system such as aproduction or inventory system, or the entire firm.

g. Thelimitations of simulation are as follows:

1) Cost. Simulation models can be costly to develop. They can be justified only ifthe information to be obtained is worth more than the costs to develop themodel and carry out the experiment.

2) Risk of error. A simulation results in a prediction of how an actual system wouldbehave. As in forecasting, the prediction may be in error.

h. Sensitivity analysis. After a problem has been formulated into any mathematicalmodel, it may be subjected to sensitivity analysis.

1) A trial-and-error method may be adopted in which the sensitivity of the solutionto changes in any given variable or parameter is calculated.

a) The risk of the project being simulated may also be estimated.

b) The best project may be one that is least sensitive to changes inprobabilistic (uncertain) inputs.

2) Inlinear programmingproblems, sensitivity is the range within which aconstraint value, such as a cost coefficient or any other variable, may bechanged without changing the optimal solution. Shadow price is the synonymfor sensitivity in that context.




27/42


28/42

e. Thearrivalsin a queuing model occur in accordance with a Poisson process.

1) The model has the followingassumptions:

a) The probability of occurrence of an event is constant,

b) The occurrence of an event is independent of any other,

c) The probability of an occurrence is proportional to the length of the interval,and,

d) If the interval is small enough, the probability of more than one occurrenceapproaches zero.

2) ThePoisson probability distributionis used to predict the probability of aspecified number of occurrences of an event in a specified time interval, giventhe expected number of occurrences per time unit.

3) The relatedexponential distributionis used to approximate service times. Thisdistribution gives the probability of zero events in a given interval. Accordingly,it gives the probability that service time will not exceed a given length of time.

f. Formulas for a Single-Service Facility and Random Arrivals of Work Units

1) If: B = average number of work units arriving in one unit of time,T = average number of work units serviced in one unit of time (if no shortage

of work units), and

B< 1 (otherwise the queue will grow to infinite length), thenT

a) The average number of work units waiting in line or being serviced is

BN =

(T B)

b) The average number in the waiting line (Nq) is

B 2

Nq=T(T B)

c) The average waiting time before service is

NqW=

B

2) EXAMPLE: Cars arrive at a toll booth at the average rate of 3 cars per minute.The toll booth can serve (collect tolls from) 6 cars per minute on the average.

a) The average number of cars waiting in line or paying tolls at any time is

B 3N = = = 1car

(T B) 6 3

b) The average number in the waiting line, not being serviced, is

B 2 9

Nq = = = 1/2car, or one car waiting half the timeT(T B) 6(6 3)

c) The average waiting time is

Nq (1 2)W = = = 1/6minute, or10 seconds

B 3




29/42

9. A variant of intuitive forecasting isscenario analysis(scenario planning). This techniqueinvolves developing formal, written descriptions of equally likely future alternatives (usuallytwo to four). It is aqualitative procedure(possibly with some quantitative input) reflectingan understanding that future events involve many variables not susceptible toquantification.

a. Alongitudinal scenarioconcerns how future conditions will develop from currentconditions.

b. The more commoncross-sectional scenariodescribes a future state at a certainmoment in time.

c. Scenario analysis is a long-range forecasting method (often 5 years or more) that isbased onmultiple forecasts. For example, scenarios may be written about howvery favorable, normal, or unfavorable economic conditions will affect a particularmarket, product, or industry.

1) This strategic planning method is beneficial because itavoids surprise.

10. Game Theory

a. Game theory classifies games according to how many players participate and the sum

of the results. In a two-person game, if the payoff is given by the loser to the winner,the algebraic sum is zero and the game is called a zero-sum game.

1) If it is possible for both players to profit, however, it is called apositive-sumgame.

b. Mathematical models have been developed to select optimal strategies for certainsimple games.

1) For example, labor negotiations can be viewed as a two-person, nonzero-sumgame.

c. Game theorists have developed variousdecision rules.

1) Themaximax criterionis a decision rule adopted byrisk-seeking, optimisticplayers who desire the largest possible payoff and are willing to accept high

risk. The player determines (a) the payoff for each state of nature expected toarise after each possible decision, (b) the maximum payoff for each decision,and (c) the decision with the maximum payoff.

2) A player who uses theminimax criteriondetermines themaximum lossforeach decision possibility and then chooses the decision with the minimummaximum loss. This rule produces the same result as themaximintechnique,which determines the minimum payoff for each decision and then chooses thedecision with the maximum minimum payoff.

a) Minimax and maximin are conservative criteria used byrisk-averseplayers for whom the utility of a gain is less than the disutility of an equalloss.

3) Theminimax regret criterionis used by a player who wishes to minimize the

effect of a bad decision in either direction. It chooses the decision that has thelowest maximum opportunity cost (profit forgone).




30/42

4) Theinsufficient reason (Laplace) criterionmay be used when the decisionmaker cannot assign probabilities to the states of nature arising after adecision.

a) The reasoning is that, if no probability distribution can be assigned, theprobabilities must be equal, and theexpected valueis calculated

accordingly. For each decision, the payoffs for the various states ofnature are simply added, and the decision with the highest total ischosen. This criterion isrisk-neutral.

5) Anexpected value criterionmight be used by a risk-neutral player, that is, onefor whom the utility of a gain is the same as the disutility of an equal loss.

11. For decisions involvingrisk, expected valueprovides a rational means for selecting thebest alternative. The expected value of an action is found by multiplying the probability ofeach outcome by its payoff and adding the products. It is thelong-term average payofffor repeated trials. The best alternative has the highest expected value.

a. EXAMPLE: A dealer in yachts may order 0, 1, or 2 yachts for this seasons inventory.The cost of carrying each excess yacht is $50,000, and the gain for each yacht soldis $200,000.

State of Nature = Decision = Decision = Decision =Actual Demand Order 0 Order 1 Order 2

0 yachts $0 $(50,000) $(100,000)1 yacht 0 200,000 150,0002 yachts 0 200,000 400,000

1) The probabilities of the seasons demand are

Demand Pr

0 .101 .502 .40

2) The dealer may calculate the expected value of each decision as follows:

Order 0 Order 1 Order 2

$0 .1 = $0 $(50,000) .1 = $ (5,000) $(100,000) .1 = $ (10,000)0 .5 = 0 200,000 .5 = 100,000 150,000 .5 = 75,0000 .4 = 0 200,000 .4 = 80,000 400,000 .4 = 160,000

EV(0) = $0 EV(1) = $175,000 EV(2) = $225,000

a) The decision with the greatest expected value is to order two yachts.Absent additional information, the dealer should order two.

b. Perfect informationis the knowledge that a future state of nature will occur withcertainty, i.e., being sure of what will occur in the future. Theexpected value ofperfect information (EVPI)is the difference between the expected value withoutperfect information and the return if the best action is taken given perfect information.

1) EXAMPLE (continued):Best Action Expected Value

State of Nature Pr Best Action Payoff (Pr Payoff)Demand = 0 .1 Buy 0 $ 0 $ 0Demand = 1 .5 Buy 1 200,000 100,000Demand = 2 .4 Buy 2 400,000 160,000

$260,000




31/42

a) The dealer expects to make $260,000 with perfect information about futuredemand and $225,000 if the choice with the best expected value ismade. EVPI is

Expected value with perfect information $260,000Expected value without perfect information (225,000)

$ 35,000b) The dealer will not pay more than $35,000 for perfect information.

12. Well-designedsurveysusing questionnaires or interviews are often used to determinecustomer preferences, attitudes, and tastes. They also may be used to gather opinionsfrom experts.

1.5 PROJECT MANAGEMENT TECHNIQUES

1. Project management techniques are designed to aid the planning and control of large-scaleprojects having many interrelated activities.

a. Aprojectis a temporary undertaking withspecified objectivesthat often involves a

cross-functional teamand working outside customary organizational lines. Hence,interpersonal skillsare at a premium in project management because a managermay not have line authority over some team members.

b. Theproject life cycleconsists of

1) Conceptualization(setting overall objectives, budgets, and schedules)

2) Planning(obtaining resources, assigning duties, and coordinating activities)

3) Execution(monitoring, correcting, meeting expectations, and finishing theproject within time and budgetary limits)

a) The largest amount of resources are used during this stage.

4) Termination(turning the project over to the user and redistributing projectresources)

c. Project management softwareis available. Among other things, it should

1) Specify and schedule required activities2) Provide sensitivity analysis of the effects of changes in plans3) Calculate a projects critical path4) Establish priorities5) Be able to modify or merge plans6) Manage all types of project resources7) Monitor progress, including adherence to time budgets for activities

d. The following arebasic planning guidesfor projects:

1) The schedule and results are more important than the process.

2) A manager must simultaneously consider the overall objective, the deadline, andoperational details.

3) Planning is essential.

4) A challenging deadline is a strong motivator.

e. Enterprise resource planning (ERP)is relevant to project management. For adetailed outline, see Study Unit 10.




32/42

f. Example applications of project management

1) Building construction2) Research and development projects3) New product planning4) Feasibility studies

5) Audit studies6) Movie production7) Conversion to a new computer information system

2. Bar charts(bar graphs) are two-dimensional (variables plotted on x and y axes) graphicmeans of quantitative comparison. They use bars or rectangles with lengths proportional tothe measure of the data or things being compared.

a. Bar charts may be vertical (column charts) or horizontal.

b. Pareto analysisis based on the concept that about 80% of results or effects arecaused by about 20% of people or events. Thus, managers should concentrate onthe relatively few people or events that have the most significant effects. Paretodiagramsare bar charts in which the frequencies of the various types of adverseconditions that may occur in a given process are depicted.

1) For example, an auditor might construct a Pareto diagram of the types andnumbers of control deviations found in the receiving department. The tallest (orlongest) bar signifies the most common type of problem.

c. Ahistogramis a bar chart used to depict a frequency distribution.

1) It is a graph of the results of repeated measurements of a single variable, withtheclass intervalson the horizontal axis and the frequency(or relativefrequency) of occurrence on the vertical axis.

a) Whether the histogram approximates or deviates from a normal distributionis of interest for control purposes.

2) EXAMPLE: A histogram of the number of occurrences of various net cash flowsis given below.

a) Cash flows between $0 and $1,000 occurred on 5 days; cash flowsbetween $1,000 and $2,000 also occurred on 5 days; cash flows between$2,000 and $3,000 occurred on 3 days, etc.




33/42


34/42

2) Thestandard deviation of the completion time for the critical path is thesquare root of the sum of the variances (squares of the standard deviations) ofthe activity times.

a) EXAMPLE: If the critical path has two activities, and the standarddeviations of the completion times are 3 and 4, the standard deviation for

the critical path is

e. Paths that are not critical haveslack time. One advantage of PERT is that it identifiesslack time, which represents unused resources that can be diverted to the criticalpath.

f. Several techniques have been developed to include cost information in the analyses.This variation of PERT is often called PERT-Cost. It entails combining activities intowork packages to facilitate cost control. By estimating costs for each work package,a manager can develop a budget that indicates when costs should be incurred duringthe project.

g. Activity timescan be expressed probabilistically. Computer programs are available

to make the calculations and find critical paths.h. PERT analysis includes probabilistic estimates of activity completion times. Three

time estimatesare made optimistic, most likely, and pessimistic.

1) The time estimates for an activity are assumed to approximate a beta probabilitydistribution. In contrast with the normal distribution, this distribution has finiteendpoints (the optimistic and pessimistic estimates) and is unimodal; that is, ithas only one mode (the most likely time).

2) PERT approximates the mean of the beta distribution by dividing the sum of theoptimistic time, the pessimistic time, and four times the most likely time (themode) by six.

3) Thestandard deviationis approximated by dividing the difference between thepessimistic and optimistic times by six. The basis for the latter approximation isthat various probability distributions have tails that lie about plus or minus threestandard deviations from the mean. For example, 99.9% of observations in thenormal distribution are expected to lie within this range.

i. EXAMPLE: If an activity can be completed in 6 days (optimistic time), 10 days (mostlikely time), or 20 days (pessimistic time), the expected duration is 11 days {[6 + (4 10) + 20] 6}.

1) Thus, the most likely time is weighted the most heavily.2) The standard deviation is 2.33 [(20 6) 6].

j. EXAMPLE:




35/42

1) For the network on the previous page, the following are the paths and pathtimes:

Path Time (hours)Start-1-9-6 16.5Start-1-2-5-6 16.9

Start-1-2-3-4-6 20.8Start-1-3-4-6 19.3Start-1-7-8-6 17.4

2) Path Start-1-2-3-4-6 is thecritical pathbecause it has the longest time.

3) Path 1-3 takes only 5.7 hours, but the critical path events (1-2-3) take 7.2 hours.The slack time represented by path 1-3 is thus 7.2 5.7, or 1.5. Peopleassigned to path 1-3 have an extra 1.5 hours to help elsewhere.

5. Thecritical path method (CPM) was developed independently of PERT and is widely usedin the construction industry. CPM may be thought of as a subset of PERT. Like PERT, it isa network technique. Unlike PERT, it uses deterministic time and cost estimates. Itsadvantages include cost estimates plus the concept of crash efforts and costs.

a. Activity times are estimated for normal effort and crash effort. Crash timeis the timeto complete an activity, assuming that all available resources were devoted to thetask (overtime, extra crew, etc.).

b. Activity costs are also estimated for normal and crash efforts.

c. These estimates allow the project manager to estimate the costs of completing theproject if some of the activities are completed on a crash basis.

d. Thenetwork diagramis constructed in the same manner as PERT diagrams. Oncethe diagram is constructed, the critical paths are found for normal and crash times.More than one critical path may exist for each diagram.

e. Crashing the networkmeans finding the minimum cost for completing the project inminimum time.

1) EXAMPLE (CMA, adapted): Builder uses the critical path method to monitorjobs. It is currently 2 weeks behind schedule on Job #181, which is subject to a$10,500-per-week completion penalty. Path A-B-C-F-G-H-I has a normalcompletion time of 20 weeks, and critical path A-D-E-F-G-H-I has a normalcompletion time of 22 weeks. The following activities can be crashed:

Cost to Crash Cost to CrashActivities 1 Week 2 Weeks

BC $ 8,000 $15,000DE 10,000 19,600EF 8,800 19,500

Builder desires to reduce the normal completion time of Job #181 and report thehighest possible income for the year. Builder should crash Activity DE 1 week

and activity EF 1 week. Activities to be crashed should be on the critical path.Thus, activity BC should not be selected. It is not on the critical path. To crashactivity BC would not reduce the total time to complete the project. The onlyfeasible choices are DE and EF on the critical path. The total cost to crash DEand EF for 1 week each is $18,800 ($10,000 + $8,800), which is less than thecost to crash either activity for 2 weeks. Thus, DE and EF should be crashedfor 1 week each. The total cost is less than the $21,000 ($10,500 2) 2-weekdelay penalty.

f. CPM computer programs allow updating of the solution as work proceeds.




36/42


37/42


38/42

b. TOC analysis defines all costs asfixedin the short-term except direct materials costs.Accordingly, thethroughput contributionequals sales dollars minus directmaterials costs, which include materials handling costs as well as raw materials andpurchased components. This approach is a type of supervariable costingbecauseonly direct materials costs are inventoried.

1) The objective of TOC analysis is tomaximize throughput contributionand tominimize investments(defined as materials costs of all inventories, plus R&Dcosts, plus fixed assets) andother operating costs(defined as all operatingcosts other than direct materials costs necessary to earn the throughputcontribution).

c. TOC analysisidentifies the bottleneck operationthat determines the thr

1. Business Performance

Documents

Transcript of 1. Business Performance