Decomposition,Interdependence and Precision in Multiattribute Utility Measurements

Traditionally, parameters of multiattribute utility models, representing a decision maker's preference judgements, are treated deterministically. This may be unrealistic, because assessment of such parameters is potentially fraught with imprecisions and errors. We thus treat such parameters as stochastic and investigate how their associated imprecision/errors are propagated in an additive multiattribute utility function in terms of the aggregate variance. Both a no information and a rank order case regarding the attribute weights are considered, assuming a uniform distribution over the feasible region of attribute weights constrained by the respective information assumption. In general, as the number of attributes increases, the variance of the aggregate utility in both cases decreases and approaches the same limit, which depends only on the variances as well as the correlations among the single-attribute utilities. However, the marginal change in aggregate utility variance decreases rather rapidly and hence decomposition as a variance reduction mechanism is generally useful but becomes relatively ineffective if the number of attributes exceed about 10. Moreover, it was found that utilities which are positively correlated increase the aggregate utility variance, hence every effort should be made to avoid positive correlations between the single-attribute utilities. We also provide guidelines for determining under what condition and to what extent a decision maker should decompose to obtain an aggregate utility variance that is smaller than that of holistic assessments. Extensions of the current model and empirical research to support some of our behavioural assumptions are discussed. © 1997 John Wiley & Sons, Ltd.     

Evaluating the desirability of meals: an illustrative multiattribute decision analysis procedure to assess portfolios with interdependent items

This paper presents an exploratory study designed to address complex scaling problems in applying value/utility theory to measuring preferences over alternative combinations (or portfolios) of multiattributed items. This application considers meals served to nursing home residents as portfolios of food items and seeks to maximize meal appeal. We consider food interrelation and link single food measurements to overall meal measurements in the process of constructing meal desirability scales. We test the resulting scales to capture one expert's professional judgments and discuss issues raised in this specific application in the context of general portfolio evaluation problems. © 1998 John Wiley & Sons, Ltd.     

The decision tree polytope and its application to sequential decision problems

This paper describes a new mathematical programming approach to sequential decision problems that have an underlying decision tree structure. The approach, based upon a characterization of strategies as extreme points of a 0–1 polytope called the ‘decision tree polytope’, is particularly suited to the direct examination of risk-return and other tradeoffs amongst strategies. However, it can also be used for conventional utility maximization if a utility function is available. Further, the approach requires no algorithmic development—it can be implemented using commercially available algebraic modeling software and can solve large problems. A related, and already known, approach can be used for some more general Markov decision problems. © 1998 John Wiley & Sons, Ltd.     

A methodology to assess the utility of future space systems

The Air Force SPACECAST 2020 study identified and prioritized some future space systems and technologies required in the next century. This paper presents the results of research on the SPACECAST 2020 value model. The model determines and prioritizes future space systems’ utility towards controlling and exploiting space. This research identifies the assumptions and simplifications in the additive utility function and assesses modifications. The research shows that mutual utility independence of the mission areas is a reasonable assumption. Mutual utility independence allows the use of the multiplicative and multilinear utility functions. Also, the SPACECAST 2020 scoring functions used the same scoring scale and only measured the utility of future capabilities. This study makes modifications to the SPACECAST 2020 measure-of-merit scoring functions. It replaces most of these functions with either a concave, convex, linear or ‘S’ scoring function, which have expanded capability ranges to include both current and future capabilities. The modified scoring functions and alternative utility functions do not alter the SPACECAST 2020 results but do improve the credibility and future usefulness of the model. This study shows that the initial assumption of using an additive utility function is also valid. © 1997 John Wiley & Sons, Ltd.     

Attribute weighting methods and decision quality in the presence of response error: a simulation study

This paper uses a simulation approach to investigate how different attribute weighting techniques affect the quality of decisions based on multiattribute value models. The weighting methods considered include equal weighting of all attributes, two methods for using judgments about the rank ordering of weights, and a method for using judgments about the ratios of weights. The question addressed is: How well does each method perform when based on judgments of attribute weights that are unbiased but subject to random error? To address this question, we employ simulation methods. The simulation results indicate that ratio weights were either better than rank order weights (when error in the ratio weights was small or moderate) or tied with them (when error was large). Both ratio weights and rank order weights were substantially superior to the equal weights method in all cases studied. Our findings suggest that it will usually be worth the extra time and effort required to assess ratio weights. In cases where the extra time or effort required is too great, rank order weights will usually give a good approximation to the true weights. Comparisons of the two rank-order weighting methods favored the rank-order-centroid method over the rank-sum method. © 1998 John Wiley & Sons, Ltd.     

Dynamic decision making: a comparison of approaches

This paper is concerned with a specific type of problem, namely dynamic decisions, for which most techniques fail to provide adequate solutions. Here, we present two of the most promising optimization techniques, partially observable Markov decision processes (POMDP) and dynamic decision networks (DDN), while arguing which is the most suitable for this problem domain. Copyright © 2000 John Wiley & Sons, Ltd.     

Problem formulation for multi-criteria decision analysis: report of a workshop

This paper reports on a workshop on Problem Formulation in Multi-Criteria Decision Analysis held at SPUDM97. The focus of the workshop was the problem formulation phase which occurs between the analyst meeting a person with a ‘mess’ and the time he or she begins to analyse a structured problem with several alternatives scored against several attributes or criteria. The objectives were: to share experience on procedures which might be transferable between the methodologies; to demonstrate different skills used by the analyst in structuring decision problems; and to catalyse a discussion on the problem formulation phase of an analysis. Three analysts, who generally approach problems using multiattribute methods, addressed the same problem. The problem used was constructed to be realistic to three decision makers, who had been trained in the issues of concern. There were two sessions. Each analyst was assigned a decision maker and formulated the problem independently in the first session, held in parallel. They were each observed by two observers and many of the audience at the workshop. The three formulations were presented along with the comments of the observers and discussed at a second plenary session. This paper reports the three formulations and observations, remarking on the ‘tricks of the trade’ employed by the analysts in formulating the problem. The analysts also describe their thinking and their aims in adopting their approach and style of interaction. More general remarks on the process of decision analysis are also offered. © 1998 John Wiley & Sons, Ltd.     

Small‐unit water purifiers for remote military outposts: A new application of multicriteria decision analysis

During the U.S. military's first 145 years, waterborne and other infectious diseases caused more casualties than battlefield injuries, underscoring the importance of clean drinking water to military success. Ensuring drinking water safety poses special challenges to the U.S. Army Special Operations Forces, because treatment equipment used in the conventional Army is too heavy and operationally complex to deploy to remote outposts. Due to the lack of purpose‐built water purifiers, Special Forces often rely on commercial‐off‐the‐shelf systems, but guidelines on selecting optimal systems for specific deployment contexts are lacking. To fill this gap, this research examines whether multicriteria decision analysis (MCDA) can provide an operational framework for selecting water purifiers for Special Forces. Our objectives are to apply MCDA to help an officer preparing for village stability operations in Afghanistan select a water purifier and to assess whether other Army stakeholders agree with the officer's choice. In addition, because previous research has suggested that alternative MCDA methods can yield different rankings of decision alternatives, we test whether the MCDA method used is as important as who conducts the ranking. Toward these objectives, we use 4 MCDA approaches for eliciting preferences among 4 water purifiers from the deploying officer and 6 other Army stakeholders. The results show that regardless of method, the officer and 5 of the 6 stakeholders identified the same preferred alternative, although rankings for the less preferred water purifiers varied by method (intermethod rank‐order correlation = .57). In responses to survey questions, stakeholders indicated that they would support using the results to support decisions about water purifier acquisition (average support = 5.3 on a 0–6 Likert scale with 6 indicating strongest support).     

行为决策理论之父：纪念Edwards教授2周年忌辰

著名心理学家W ard Edwards对行为决策的学科形成与理论发展作出了巨大贡献。文章简要介绍了Edwards的生平,以主观期望效用模型和贝叶斯决策为核心回顾了Edwards的决策理论、主要成就及其对行为决策研究的学术贡献。文章并从行为决策领域的研究、科普、应用及人才培养等方面总结了中国行为决策研究的发展历程,以及Edwards的学术思想对中国行为决策研究的主要影响。文章展望了行为决策研究的前景,并呼吁更多的中国心理学工作者投身于行为决策研究。     

ACADEA—a decision support system for faculty performance reviews

ACADEA, a multi‐criteria decision support system for the performance review of individual faculty, is presented. Developed from the point of view of a department that is facing exogenously as well as self imposed objectives, the support system looks upon the aggregate performance of an academic department as the result of individual faculty member's multi‐criteria evaluations. Five objectives, research output, teaching output, external service, internal service and cost, are operationalized into criteria. The system is applied to a university department with 30 faculty members evaluated over a 3‐year period. The results identify promotional candidates and reveal underlying problems in managerial consistency, departmental sub‐groupings and the incentive structure. The outcomes of the support system are consistent with the position that equity in faculty governance does not necessarily imply equal loads on all tasks. Copyright © 2000 John Wiley & Sons, Ltd.     

Numerical and verbal decision analysis: comparison on practical cases

Numerical decision analysis (NDA), derived from statistical decision theory, is very well known. Verbal decision analysis (VDA), oriented towards so‐called unstructured problems, where the qualitative and uncertain factors dominate, is a newer direction in decision theory and practice. Verbal and numerical decision analyses (DAs) have been compared in an experimental setting, with groups of students. This paper presents the results of a comparison in the context of live practical tasks. Both approaches were attempted on two comparable choices, facing both Russian and US government agencies, involving a choice between oil and gas transportation options. The resulting methodological insights are generalized into a systematic comparison of the strong and weak features of each approach. Copyright © 2000 John Wiley & Sons, Ltd.     

A dual-process approach to health risk decision making: The prototype willingness model

Although dual-process models in cognitive, personality, and social psychology have stimulated a large body of research about analytic and heuristic modes of decision making, these models have seldom been applied to the study of adolescent risk behaviors. In addition, the developmental course of these two kinds of information processing, and their relation to the development of self-regulation are not well understood at this time. The current paper reviews what leading dual-process models have to say about the development of analytic and heuristic decision making, and their implications for adolescent risk behavior. In addition, it reviews research on the prototype willingness model of adolescent decision making—a dual-process model designed specifically to address non-intentional, but volitional adolescent risk behavior. It also discusses the implications of dual-process models for intervention research.     

Valuing the future: a MADA example involving nuclear waste storage

Intertemporal decision making involves decisions that have consequences that span several periods of time and often extend far into the future. The purpose of this paper is to discuss and highlight the differences associated with different evaluation methods designed to cope with the long-term impacts of a decision including discounting. The concepts and ideas are illustrated in the context of a decision about a nuclear waste facility. We show how applying different discounting methodologies can greatly affect the decision made, especially over long time periods. © 1998 John Wiley & Sons, Ltd.     

An interactive multicriteria decision model for multipurpose reservoir management: the Shellmouth Reservoir

Reservoir management is inherently multicriterial, since any release decision involves implicit trade-offs between various conflicting objectives. The release decision reflects concerns such as flood protection, generation of hydroelectric power, dilution of wastewater and heated effluent, supply of municipal, agricultural and industrial water, maintenance of environmental standards and satisfaction of recreational needs. This paper presents a framework for analysing trade-offs between several decision criteria and includes the dilution of heated effluents from downstream thermoelectric power generation in an optimization model for reservoir management. The model is formulated and analysed in an interactive multicriteria decision-making (MCDM) modelling framework. Rather than providing specific target levels or ad hoc constants in a goal-programming framework as proposed elsewhere, this multicriteria framework suggests a systematic way of evaluating trade-offs by progressive preference assessment. The MCDM model, based on a Chebyshev metric and a contracted cone approach, is learning-oriented and permits a natural exploration of the decision space while maintaining non-dominated decisions. A detailed case study of the Shellmouth Reservoir in Manitoba, Canada serves as an illustration of the model. © 1998 John Wiley & Sons, Ltd.     

个体自信度对双人决策的影响

群体决策是重要的社会现象, 个体自信度在群体决策中发挥了重要作用。本文开展了不同难度和信息交流方式下的双人决策实验, 通过分析自信度和个体决策以及决策调整行为的关系, 研究了个体自信度的交流对双人决策的影响。实验结果表明, 个体的自信度与选择的正确率高度正相关; 双人决策过程是个体根据对方的自信度和选择来不断调整自己的选择最终达成一致的过程, 并通过交互过程提高双人决策的正确率; 实验中双人决策的质量明显优于“自信度分享模型”和“更自信者主导决策模型”的预期结果, 表明群体决策不是通过分享自信度进行的贝叶斯优化整合过程, 也不是由更自信的个体完全主导的过程。     

Decision by sampling: The role of the decision environment in risky choice

Decision by sampling (DbS) is a theory about how our environment shapes the decisions that we make. Here, I review the application of DbS to risky decision making. According to classical theories of risky decision making, people make stable transformations between outcomes and probabilities and their subjective counterparts using fixed psychoeconomic functions. DbS offers a quite different account. In DbS, the subjective value of an outcome or probability is derived from a series of binary, ordinal comparisons with a sample of other outcomes or probabilities from the decision environment. In this way, the distribution of attribute values in the environment determines the subjective valuations of outcomes and probabilities. I show how DbS interacts with the real-world distributions of gains, losses, and probabilities to produce the classical psychoeconomic functions. I extend DbS to account for preferences in benchmark data sets. Finally, in a challenge to the classical notion of stable subjective valuations, I review evidence that manipulating the distribution of attribute values in the environment changes our subjective valuations just as DbS predicts.     

Sensitivity analysis with interdependent criteria for multicriteria decision making: The case of soil pollution treatment

This paper is focused on interdependences between criteria in multicriteria decision analyses. Such interdependences are usually ignored in sensitivity analyses. After a discussion on the nature of interdependences, methods are presented to deal with interdependences in the context of Monte Carlo experiments. These methods are applied in the context of soil pollution treatment alternatives. It is shown that ignoring interdependences may have a distorting effect on the results of sensitivity analysis on rankings of alternatives.     

Latent growth curve analysis with dichotomous items: Comparing four approaches

A Monte Carlo study was used to compare four approaches to growth curve analysis of subjects assessed repeatedly with the same set of dichotomous items: A two‐step procedure first estimating latent trait measures using MULTILOG and then using a hierarchical linear model to examine the changing trajectories with the estimated abilities as the outcome variable; a structural equation model using modified weighted least squares (WLSMV) estimation; and two approaches in the framework of multilevel item response models, including a hierarchical generalized linear model using Laplace estimation, and Bayesian analysis using Markov chain Monte Carlo (MCMC). These four methods have similar power in detecting the average linear slope across time. MCMC and Laplace estimates perform relatively better on the bias of the average linear slope and corresponding standard error, as well as the item location parameters. For the variance of the random intercept, and the covariance between the random intercept and slope, all estimates are biased in most conditions. For the random slope variance, only Laplace estimates are unbiased when there are eight time points.     

A review of multiobjective programming and its application in quantitative psychology

Multiobjective programming, a technique for solving mathematical optimization problems with multiple conflicting objectives, has received increasing attention among researchers in various academic disciplines. A summary of multiobjective programming techniques and a review of their applications in quantitative psychology are provided.