The acceptance and accuracy of decision analysis methods

This paper reports a field experiment which compared two approaches to decision analysis, called cases and criteria scaling, in terms of their acceptance to users and their predictive features. The case method simulates decisions likely to occur in practice. Each decision is described by particular values of the decision criteria. Experts consider the values of criteria present in each decision and make a judgment. Regression is used to correlate the criteria values with the judgments. The regression equation provides the prediction model. The criteria scaling method decomposes the decision task. Experts weight each criterion and specify how decisions change across levels of each criterion. The predictive model is made up of the sum of each criterion weight multiplied by the criterion-predictor functional relationship. Both methods were applied to build models which predicted the demands for nursing time, using patient severity indicators, for two hospital units. The case method had considerable predictive accuracy and had favorable participant reactions. Predictions made by criteria scaling overestimated needs and this method was viewed as inaccurate and hard to understand by participants. The implications of these findings are discussed.     

A CASE IN ON-LINE DECISION ANALYSIS FOR PRODUCT PLANNING1

This report presents an application of quantitative decision-analytic techniques to consider an opportunity for capital investment in a new product. Decision analysis is a well-established technique for evaluating major decisions in which substantial resources and time are available for the analysis and the problem is definitely formulated at the outset. This case study, however, illustrates how decision-analytic techniques can be used on-line to improve decision making in a situation in which time and resources are limited, and the manager's perception of the problem changes in response to feedback from the analysis. The basic decision problem was whether or not the AIL division of Cutler-Hammer, Incorporated should purchase a six-month option on a flight-safety system patent. The president of AIL had only a few weeks in which to make a decision. The analytic approach focused on a probabilistic discounted earnings model. This model was refined over a two-week period through a sequence of iterations which incorporated the results of both direct and indirect assessments of probabilities and values. The analysis affected the company's decision process by providing a vehicle for structuring the ongoing communication between AIL's president and his advisors, and by focusing their attention on the most important issues. In the end, the analysis facilitated a unanimous decision from the decision-making group not to purchase the option, but to consider participating in the license as a subcontractor at a later date. This strategy had not been considered at the outset.     

Planning with multiple objectives

Planning any task that is characterised by the presence of multiple objectives presents the decision maker with a set of special considerations which are not present in the case of decision making with one unique objective. The paper outlines the stages in the formulation of a dynamic multiple-objective objective function and introduces the criterion complex as a decision making construct particularly suited to formulating and solving problems with multiple objectives. The paper then considers some of the mathematical properties of the alternative formulations of criterion complex.     

SMALL PANELS OF EXPERTS IN DICHOTOMOUS CHOICE SITUATIONS*

This paper analyzes an expert resolution problem under an uncertain dichotomous choice situation. The experts share a common system of norms and therefore they all prefer the alternative that best suits their purpose. The selection of such an alternative is referred to as a correct choice. Our analysis of optimal decision rules for panels of independent experts is pursued for n-member decision-making bodies, n≤ 5. The suggested optimality criterion is the maximization of the probability of the panel's making the correct choice. Within our framework, this criterion is equivalent to the more common criterion of expected-utility maximization. For three-member panels of experts, the expert resolution problem is solved and illustrated by means of a medical application. For four-member panels, we list the three relevant decision rules, specify the conditions for all possible rankings of these rules, and, finally, present an extended consulting application. We conclude by listing seven relevant decision rules in the case of five-member decision-making bodies.     

CONSTRAINED DECISION CRITERIA1

The choice of a suitable decision criterion is a difficult decision for many decision makers. In general, each criterion selects a different optimal act, each having its own appealing features. This paper proposes an approach allowing the decision maker to incorporate multiple criteria into an easily solved mathematical programming model whose solution is the optimal act. Optimality is defined by the objective function used in the model. The constraints embody the decision maker's anxieties and minimal desires.     

ON THE EQUIVALENCE OF TWO VERSIONS OF THE PRESENT VALUE PROCEDURE

The central issue in equipment investment analysis is the selection of the best of two or more alternatives. A common procedure is to first assume that for each alternative there is an infinite stream of identical pieces of equipment. The present value of the associated infinite stream of revenues and costs is calculated. Then the alternative having the highest present value (where revenues are positive and costs are negative) is selected. The assumption of the infinite stream, though computationally desirable, is not very palatable to the decision maker. In this paper we show that the above procedure is equivalent to using a criterion which is definitely more appealing. The author has found this “equivalence” result to be helpful in teaching the basic concepts of the present value procedure. Also the result should be of interest to practitioners involved in equipment investment decision making.     

Decisional Conflict and User Acceptance of Multicriteria Decision-Making Aids*

Despite the development of increasingly sophisticated and refined multicriteria decision-making (MCDM) methods, an examination of the experimental evidence indicates that users most often prefer relatively unsophisticated methods. In this paper, we synthesize theories and empirical findings from the psychology of judgment and choice to provide a new theoretical explanation for such user preferences. Our argument centers on the assertion that the MCDM method preferred by decision makers is a function of the degree to which the method tends to introduce decisional conflict. The model we develop relates response mode, decision strategy, and the salience of decisional conflict to user preferences among decision aids. We then show that the model is consistent with empirical results in MCDM studies. Next, the role of decisional conflict in problem formulation aids is briefly discussed. Finally, we outline future research needed to thoroughly test the theoretical mechanisms we have proposed.     

Signaling to Partially Informed Investors in the Newsvendor Model

We analyze a signaling game between the manager of a firm and an investor in the firm. The manager has private information about the firm's demand and cares about the short‐term stock price assigned by the investor. Previous research has shown that under continuous decision choices and the Intuitive Criterion refinement, the least‐cost separating equilibrium will result, in which a low‐quality firm chooses its optimal capacity and a high‐quality firm over‐invests in order to signal its quality to investors. We build on this research by showing the existence of pooling outcomes in which low‐quality firms over‐invest and high‐quality firms under‐invest so as to provide identical signals to investors. The pooling equilibrium is practically appealing because it yields a Pareto improvement compared to the least‐cost separating equilibrium. Distinguishing features of our analysis are that: (i) we allow the capacity decision to have either discrete or continuous support, and (ii) we allow beliefs to be refined based on either the Undefeated refinement or the Intuitive Criterion refinement. We find that the newsvendor model parameters impact the likelihood of a pooling outcome, and this impact changes in both sign and magnitude depending on which refinement is used.     

An application of multiple objective linear programming to media selection

A multiple objective linear programming (MOLP) method utilizing interval criterion weights is applied to the problem of media selection. Using this technique, it is possible to analyze a problem more explicitly in terms of the several objectives inherent in many media selection situations. In order to illustrate the interval criterion weights approach, a multiple objective hierarchical media selection model is presented and its computer results discussed. In addition to being able to deal more directly with different decision criteria, a distinguishing feature of the mathematical analysis applied here is that its output enables the media-planner to be presented with a small cluster of candidate media schedules (rather than just one). Then, from this list, the media-planner should be in a position to qualitatively make a final choice as a close approximation to his optimal solution.     

LEARNING WITH DETERMINISTIC DECISION RULES

While many problems of uncertainty are commonly analyzed by means of stochastic models, under certain circumstances this may not be an appropriate approach. The latter situation arises when the decision maker knows that the uncertain variables are not generated by a stochastic process, or when he is unwilling, or unable, to compute subjective probabilities. One of the nonstochastic approaches to uncertainty is the expectational approach in which the decision maker forms deterministic expectations about the uncertain aspects of his environment. This paper is concerned with some criteria for selecting among available expectations, or anticipations functions, and the possibility of ordering them according to these criteria. This study focuses especially on the learning criterion. The discussion brings out conceptual problems in connection with the definition of learning, as well as some technical difficulties that one encounters when attempting to compare different anticipations functions from the point of view of the learning criterion. As an illustration of the issues discussed, the paper reports on the results of some simulated decision rules. These show that decision rules in which no learning takes place, and in which some information is ignored, may perform better than more sophisticated rules.     

Cost‐Benefit Analysis for Optimization of Risk Protection Under Budget Constraints

Cost‐benefit analysis (CBA) is commonly applied as a tool for deciding on risk protection. With CBA, one can identify risk mitigation strategies that lead to an optimal tradeoff between the costs of the mitigation measures and the achieved risk reduction. In practical applications of CBA, the strategies are typically evaluated through efficiency indicators such as the benefit‐cost ratio (BCR) and the marginal cost (MC) criterion. In many of these applications, the BCR is not consistently defined, which, as we demonstrate in this article, can lead to the identification of suboptimal solutions. This is of particular relevance when the overall budget for risk reduction measures is limited and an optimal allocation of resources among different subsystems is necessary. We show that this problem can be formulated as a hierarchical decision problem, where the general rules and decisions on the available budget are made at a central level (e.g., central government agency, top management), whereas the decisions on the specific measures are made at the subsystem level (e.g., local communities, company division). It is shown that the MC criterion provides optimal solutions in such hierarchical optimization. Since most practical applications only include a discrete set of possible risk protection measures, the MC criterion is extended to this situation. The findings are illustrated through a hypothetical numerical example. This study was prepared as part of our work on the optimal management of natural hazard risks, but its conclusions also apply to other fields of risk management.     

排污权价格随机情境下基于CVaR 准则的企业生产决策

面向排污权市场价格随机情境,通过提取理性企业在产品生产、污染削减、排污权交易等方面决策行为与排污权市场价格之间的关联性,推导了排污权市场价格随机情境下的企业随机利润函数;基于条件风险价值(CVaR)风险度量准则,推导了任意置信水平下随机利润的风险价值(VaR)解析表达式,建立了不同排污权预期价格和置信水平下的CVaR测度模型;基于此,深入论证了给定置信水平下实现随机利润CVaR最大化的企业最优决策及其特征,并系统分析了外部情景要素变动对最优决策及其优化结果的影响。主要结论表明：在企业分别作为排污权出售者与购买者两种情况下,随机利润CVaR的具体测度因利润下行风险存在的排污权市场价格区间不同而存在差异;基于CVaR准则,考虑决策风险的最优决策受到决策主体置信水平、排污权初始配额、排污权市场价格均值及标准差等多因素的复杂影响,进而系统性地偏离风险中性最优决策,并表现出风险规避特征;通过算例分析对所得结论进行全面验证,进一步说明该模型能较好地刻画与反映排污权交易企业决策风险及其现实决策行为。     

On Criteria of Insignificant Difference Between Two Risks

The comparison and ranking of risks is very important for safety and cost-benefit analysis. Most formats present risks in the form of probability distribution. Different ranking criteria for probability distributions are considered. It is demonstrated that significantly overlapping distributions lead to ambiguous results. For this reason, criteria of insignificant overlapping distributions are proposed. The first criterion uses the information theory approach; and the second criterion uses the statistical tests approach. Both approaches can be applied to decision theory to avoid questionable decisions based on statistically insignificant differences between two risks.     

A case-based distance method for screening in multiple-criteria decision aid

Screening is a process of multiple-criteria decision aid (MCDA) in which a large set of alternatives is reduced to a smaller set that most likely contains the best choice. We propose screening using a distance model calibrated on the basis of the decision-maker's own judgement. Viewing MCDA as preference aggregation based on consequence data, we define consequence and preference expressions (values and weights) and describe how they are aggregated. Then we define screening and explain some of its properties. Using an appropriate definition of distance, our case-based distance method screens a set of alternatives using criterion weights and a distance threshold obtained by quadratic optimization using the decision-maker's selection of alternatives from a test set. This case-based method can elicit the decision maker's preferences more expeditiously and accurately than direct inquiry. An application in water supply planning is used to demonstrate the procedure.     

Strict Uncertainty: A Criterion for Moderately Pessimistic Decision Makers*

This paper proposes a decision rule to rank actions under strict uncertainty, the available information being limited to the states of nature, the set of alternative rows, and the consequence of choosing every row if a given state occurs. This rule is suitable to moderately pessimistic individuals and social groups, these agents being neither maximax nor maximin decision makers but people who assume that the best outcome from the action will not occur. For these decision makers the paper shows the existence of a consistent weight system in which one and only one weight is attached to each state of the world under plausible conditions of domination. Most of the traditional axioms are satisfied by the proposed ranking approach. In the frame of disappointment (measured by ranges of column dispersion), the meaning of some controversial postulates used in the literature is explained. The proposed criterion is a departure from Laplace's (1825) rule and from the remaining standard criteria. Only in the special case of equal column dispersion do both Laplace's rule and the proposed weights lead to the same solution.     

R&D项目中止决策的小波网络模式识别

如何科学、合理、有效地识别和中止那些行将失败或没有前途的 R&D项目 ,对于企业成功地进行技术创新、建立和保持竞争优势 ,显然具有十分重要的理论价值和实践意义 .本文根据中止决策的实质及模式识别的基本原理 ,探讨了定量地中止 R&D项目的一般模式识别程序 .同时 ,基于小波基具有很强的自适应数据和函数变化的能力 ,进一步研究给出了一种基于小波网络的 R&D项目中止决策分析模式识别新方法 ,并在最小均方能量准则下 ,采用相应的共轭梯度学习算法求解子波函数线性组合的尺度和时延参数 ,以及神经网络的权值 .本研究不仅克服了现有 R&D项目中止决策分析技术在构造与求解模型方面的某些局限性 ,并且为R&D项目的跟踪管理提供了一种新的理论依据 ,为具体决策单位和政府有关部门对正在实施的 R&D项目作中止决策提供了一种可操作的决策方法和程序     

A Note on Comparative Ambiguity Aversion and Justifiability

We consider a decision maker who ranks actions according to the smooth ambiguity criterion of Klibanoff, Marinacci, and Mukerji (2005). An action is justifiable if it is a best reply to some belief over probabilistic models. We show that higher ambiguity aversion expands the set of justifiable actions. A similar result holds for risk aversion. Our results follow from a generalization of the duality lemma of Wald (1949) and Pearce (1984).     

Evaluating new product decision processes

The development of a new product is often a sequential-decision process with many available stages at which the product may take any one of a number of alternate courses of action. This paper identifies an efficient algorithm for solving this sequential-decision process. A computer program, involving the algorithm, has been written to solve problems containing up to one thousand stages. Input into the program includes the subjective probabilities of success for each decision branch, the cost associated with each decision stage and level, the rate for discounting all monetary values to the present plus an indicator for each decision stage and level of the desired decision criterion. The algorithm utilizes the dynamic-programming solution approach. The output of the algorithm contains the minimum, or maximum, expected discounted decision values for each stage and level of the network. The algorithm and computer program have been primarily utilized in the pharmaceutical industry. An illustrative, exemplary use of the algorithm in this industry is presented.     

Robustness of linear models in dynamic multivariate predictions

Linear aggregation models employing unit and equal weights have been shown to be superior to human decisions in a surprising range of decision situations. In addition, decisions based on these models have often been found to be superior to those based on linear regression models (LRMs). This general issue was explored for repetitive decisions in production planning. The problem considered differs in several aspects from the types of problems investigated previously: (1) the problem is dynamic rather than static; (2) a set (or vector) of interactive decisions dependent on previous decisions is required to be made, where a decision in stage t, the dependent variable, becomes an independent variable in stage t + 1; and (3) the criterion function is cost with a quadratic loss function (rather than the correlation measure of R²). Moreover, since repetitive decisions were involved, the parameters of the models were estimated using past human decisions. These were used to predict specific values of the decision variables (rather than rank order), which in turn were employed recursively to predict values of the decision variables at subsequent stages. While decisions from equal weighting rules were found to be superior to human decisions and not greatly inferior to decisions from linear regression models, decisions from unit weighting rules performed poorly. The rationale for such performance is discussed, indicating that previous theoretical and empirical research on linear weighting models is not generally applicable to dynamic, multivariate interactive decisions problems with lagged variables.