Decision Analysis for Management Judgment / Edition 3 available in Paperback
- Pub. Date:
|Product dimensions:||6.73(w) x 9.69(h) x 1.06(d)|
About the Author
Paul Goodwin is Senior Lecturer in Management Science in the Management School at the University of Bath. His research interests focus on the role of management judgment in forecasting and decision-making, and he has published in key journals in the field. He is an Associate Editor of the International Journal of Forecasting and a member of the editorial board of the Journal of Behavioral Decision Making. He has consulted with firms such as British Telecom and South Western Electricity.
George Wright is a psychologist with an interest in the role and validity of judgment in decision making and forecasting. He is especially interested in the use of management science- and behaviorally-based methods to improve decision-making. He has published in key journals, and is the Editor of the Journal of Behavioral Decision Making. He has consulted with firms such as IBM, ICL, NEC, Petronas, Jordan Ministry of Planning, and the Scottish Football Association.
Read an Excerpt
Decision Analysis for Management Judgment
By Paul Goodwin George Wright
John Wiley & SonsISBN: 0-470-86108-8
Imagine that you are facing the following problem. For several years you have been employed as a manager by a major industrial company, but recently you have become dissatisfied with the job. You are still interested in the nature of the work and most of your colleagues have a high regard for you, but company politics are getting you down, and there appears to be little prospect of promotion within the foreseeable future. Moreover, the amount of work you are being asked to carry out seems to be increasing relentlessly and you often find that you have to work late in the evenings and at weekends.
One day you mention this to an old friend at a dinner party. 'There's an obvious solution,' he says. 'Why don't you set up on your own as a consultant? There must be hundreds of companies that could use your experience and skills, and they would pay well. I'm certain that you'd experience a significant increase in your income and there would be other advantages as well. You'd be your own boss, you could choose to work or take vacations at a time that suited you rather than the company and you'd gain an enormous amount of satisfaction from solving a variety of challenging problems.'
Initially, you reject the friend's advice as being out of the question, but as the days go by the idea seems to become more attractive. Over theyears you have made a large number of contacts through your existing job and you feel reasonably confident that you could use these to build a client base. Moreover, in addition to your specialist knowledge and analytical ability you have a good feel for the way organizations tick, you are a good communicator and colleagues have often complimented you on your selling skills. Surely you would succeed.
However, when you mention all this to your spouse he or she expresses concern and points out the virtues of your current job. It pays well - enough for you to live in a large house in a pleasant neighborhood and to send the children to a good private school - and there are lots of other benefits such as health insurance and a company car. Above all, the job is secure. Setting up your own consultancy would be risky. Your contacts might indicate now that they could offer you plenty of work, but when it came to paying you good money would they really be interested? Even if you were to succeed eventually, it might take a while to build up a reputation, so would you be able to maintain your current lifestyle or would short-term sacrifices have to be made for long-term gains? Indeed, have you thought the idea through? Would you work from home or rent an office? After all, an office might give a more professional image to your business and increase your chances of success, but what would it cost? Would you employ secretarial staff or attempt to carry out this sort of work yourself? You are no typist and clerical work would leave less time for marketing your services and carrying out the consultancy itself. Of course, if you failed as a consultant, you might still get another job, but it is unlikely that it would be as well paid as your current post and the loss of self-esteem would be hard to take.
You are further discouraged by a colleague when you mention the idea during a coffee break. 'To be honest,' he says, 'I would think that you have less than a fifty-fifty chance of being successful. In our department I know of two people who have done what you're suggesting and given up after a year. If you're fed up here why don't you simply apply for a job elsewhere? In a new job you might even find time to do a bit of consultancy on the side, if that's what you want. Who knows? If you built up a big enough list of clients you might, in a few years' time, be in a position to become a full-time consultant, but I would certainly counsel you against doing it now.'
By now you are finding it difficult to think clearly about the decision; there seem to be so many different aspects to consider. You feel tempted to make a choice purely on emotional grounds - why not simply 'jump in' and take the risk? - but you realize that this would be unfair to your family. What you need is a method which will enable you to address the complexities of the problem so that you can approach the decision in a considered and dispassionate manner.
This is a personal decision problem, but it highlights many of the interrelated features of decision problems in general. Ideally, you would like to maximize your income, maximize your job security, maximize your job satisfaction, maximize your freedom and so on, so that the problem involves multiple objectives. Clearly, no course of action achieves all of these objectives, so you need to consider the trade-offs between the benefits offered by the various alternatives. For example, would the increased freedom of being your own boss be worth more to you than the possible short-term loss of income?
Second, the problem involves uncertainty. You are uncertain about the income that your consultancy business might generate, about the sort of work that you could get (would it be as satisfying as your friend suggests?), about the prospects you would face if the business failed and so on. Associated with this will be your attitude to risk. Are you a person who naturally prefers to select the least risky alternative in a decision or are you prepared to tolerate some level of risk?
Much of your frustration in attempting to understand your decision problem arises from its complex structure. This reflects, in part, the number of alternative courses of action from which you can choose (should you stay with your present job, change jobs, change jobs and become a part-time consultant, become a full-time consultant, etc.?), and the fact that some of the decisions are sequential in nature. For example, if you did decide to set up your own business should you then open an office and, if you open an office, should you employ a secretary? Equally important, have you considered all the possible options or is it possible to create new alternatives which may be more attractive than the ones you are currently considering? Perhaps your company might allow you to work for them on a part-time basis, allowing you to use your remaining time to develop your consultancy practice.
Finally, this problem is not yours alone; it also concerns your spouse, so the decision involves multiple stakeholders. Your spouse may view the problem in a very different way. For example, he or she may have an alternative set of objectives than you. Moreover, he or she may have different views of the chances that you will make a success of the business and be more or less willing than you to take a risk.
The role of decision analysis
In the face of this complexity, how can decision analysis be of assistance? The key word is analysis, which refers to the process of breaking something down into its constituent parts. Decision analysis therefore involves the decomposition of a decision problem into a set of smaller (and, hopefully, easier to handle) problems. After each smaller problem has been dealt with separately, decision analysis provides a formal mechanism for integrating the results so that a course of action can be provisionally selected. This has been referred to as the 'divide and conquer orientation' of decision analysis.
Because decision analysis requires the decision maker to be clear and explicit about his or her judgments it is possible to trace back through the analysis to discover why a particular course of action was preferred. This ability of decision analysis to provide an 'audit trail' means that it is possible to use the analysis to produce a defensible rationale for choosing a particular option. Clearly, this can be important when decisions have to be justified to senior staff, colleagues, outside agencies, the general public or even oneself.
When there are disagreements between a group of decision makers, decision analysis can lead to a greater understanding of each person's position so that there is a raised consciousness about the issues involved and about the root of any conflict. This enhanced communication and understanding can be particularly valuable when a group of specialists from different fields have to meet to make a decision. Sometimes the analysis can reveal that a disputed issue is not worth debating because a given course of action should still be chosen, whatever stance is taken in relation to that particular issue. Moreover, because decision analysis allows the different stakeholders to participate in the decision process and develop a shared perception of the problem it is more likely that there will be a commitment to the course of action which is eventually chosen.
The insights which are engendered by the decision analysis approach can lead to other benefits. Creative thinking may result so that new, and possibly superior, courses of action can be generated. The analysis can also provide guidance on what new information should be gathered before a decision is made. For example, is it worth undertaking more market research if this would cost $100 000? Should more extensive geological testing be carried out in a potential mineral field?
It should be stressed, however, that over the years the role of decision analysis has changed. No longer is it seen as a method for producing optimal solutions to decision problems. As Keeney points out:
Decision analysis will not solve a decision problem, nor is it intended to. Its purpose is to produce insight and promote creativity to help decision makers make better decisions.
This changing perception of decision analysis is also emphasized by Phillips:
... decision theory has now evolved from a somewhat abstract mathematical discipline which when applied was used to help individual decision-makers arrive at optimal decisions, to a framework for thinking that enables different perspectives on a problem to be brought together with the result that new intuitions and higher-level perspectives are generated.
Indeed, in many applications decision analysis may be deliberately used to address only part of the problem. This partial decision analysis can concentrate on those elements of the problem where insight will be most valuable.
While we should not expect decision analysis to produce an optimal solution to a problem, the results of an analysis can be regarded as being 'conditionally prescriptive'. By this we mean that the analysis will show the decision maker what he or she should do, given the judgments which have been elicited from him or her during the course of the analysis. The basic assumption is that of rationality. If the decision maker is prepared to accept a set of rules (or axioms) which most people would regard as sensible then, to be rational, he or she should prefer the indicated course of action to its alternatives. Of course, the course of action prescribed by the analysis may well conflict with the decision maker's intuitive feelings. This conflict between the analysis and intuition can then be explored. Perhaps the judgments put forward by the decision maker represented only partially formed or inconsistent preferences, or perhaps the analysis failed to capture some aspect of the problem.
Alternatively, the analysis may enable the decision maker to develop a greater comprehension of the problem so that his or her preference changes towards that prescribed by the analysis. These attempts to explain why the rational option prescribed by the analysis differs from the decisionmaker's intuitive choice can therefore lead to the insight and understanding which, as we emphasized earlier, is the main motivation for carrying out decision analysis.
Applications of decision analysis
The following examples illustrate some of the areas where decision analysis has been applied.
Improved strategic decision making at Du Pont
The Du Pont chemical company has used influence diagrams (see Chapter 6) and risk analysis (Chapter 7) throughout the organization to create and evaluate strategies. The analysis has allowed them to take into account the effect on the value of the business of uncertainties such as competitors' strategies, market share and market size. Among the many benefits of the approach, managers reported that it enhanced team building by providing a common language for sharing information and debate. It also led to a commitment to action so that the implementation of the selected strategy was likely to be successful. One application alone led to the development of a strategy that was expected to enhance the value of the business by $175 million.
Structuring decision problems in the International Chernobyl Project
Four years after the accident at the Chernobyl nuclear power plant in 1986 the International Chernobyl Project was undertaken at the request of the Soviet authorities. Decision analysis was used in the project to evaluate countermeasure strategies (for example, relocation of some of the population, changes in agricultural practice and decontamination of buildings). The use of SMART (Chapter 3) in decision conferences (Chapter 12) enabled groups of people from a wide variety of backgrounds - such as ministers, scientists and regional officials - to meet together to structure the decision problem. They were thus able to clarify and elucidate the key issues associated with the strategies, such as the number of fatal cancers which they would avert, their monetary costs, the extent to which they could reduce stress in the population and their public acceptability. By using decision analysis it was possible to evaluate the strategies by taking into account all these issues, regardless of whether they were easily quantified or capable of being measured on a monetary scale.
Selecting R&D projects at ICI Americas
At ICI Americas, decisions on which research and development projects should be allocated resources were complicated by a number of factors. These included the large number of projects (53 on one occasion) that were proposed, the sparseness of information and uncertainty associated with these projects and the sequential nature of the decisions. For example, if a selected project is a technical success should a decision then be made to develop it commercially? Simple and transparent decision analysis models, using decision trees (Chapter 6), were used to provide logic and consistency to the selection process, allowing a variety of criteria, such as risk, speed of development and sales level, to be taken into account. The models were easily able to clarify the sequences of decisions to managers and allowed uncertainties to be explicitly addressed. Managers judged the process to be superior to the use of intuition or checklists which are often used to select research projects.
Excerpted from Decision Analysis for Management Judgment by Paul Goodwin George Wright Excerpted by permission.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
Table of ContentsDecisions Involving Multiple Objectives.
How People Make Decisions Involving Multiple Objectives.
Introduction to Probability.
Decision Making Under Uncertainty.
Decision Trees and Influence Diagrams.
Applying Simulation to Decision Problems.
Revising Judgments in the Light of New Information.
Biases in Probability Assessment.
Methods for Eliciting Probabilities.
Decisions Involving Groups of Individuals.
Resource Allocation and Negotiation Problems.
Decision Framing and Cognitive Inertia.
Scenario Planning: An Alternative Way of Dealing with Uncertainty.
The Analytic Hierarchy Process.
Alternative Decision-Support Systems.
Suggested Answers to Selected Questions.
Suppliers of Computer Software.
Most Helpful Customer Reviews
The book has provided insight into the most mysterious aspects of human psychology. These aspects matter when crcial management decisions have to be taken for meeting the organisational goals.