Multicriteria choice based on criteria importance methods with uncertain preference information

Multicriteria choice methods are developed by applying methods of criteria importance theory with uncertain information on criteria importance and with preferences varying along their scale. Formulas are given for computing importance coefficients and importance scale estimates that are “characteristic” representatives of the feasible set of these parameters. In the discrete case, an alternative with the highest probability of being optimal (for a uniform distribution of parameter value probabilities) can be used as the best one. It is shown how such alternatives can be found using the Monte Carlo method.     

On the probability of a rational outcome for generalized social welfare functions on three alternatives

In Kalai (2002) [10], Kalai investigated the probability of a rational outcome for a generalized social welfare function (GSWF) on three alternatives, when the individual preferences are uniform and independent. In this paper we generalize Kalai's results to a broader class of distributions of the individual preferences, and obtain new lower bounds on the probability of a rational outcome in several classes of GSWFs. In particular, we show that if the GSWF is monotone and balanced and the distribution of the preferences is uniform, then the probability of a rational outcome is at least 3/4, proving a conjecture raised by Kalai. The tools used in the paper are analytic: the Fourier-Walsh expansion of Boolean functions on the discrete cube, properties of the Bonamie-Beckner noise operator, and the FKG inequality.     

On Method for Uncertain Multiple Attribute Decision Making Problems with Uncertain Multiplicative Preference Information on Alternatives

The uncertain multiple attribute decision making (UMADM) problems are investigated, in which the information about attribute weights is known partly and the attribute values take the form of interval numbers, and the decision maker (DM) has uncertain multiplicative preference information on alternatives. We make the decision information uniform by using a transformation formula, and then establish an objective-programming model. The attribute weights can be determined by solving the developed model. The concept of interval positive ideal point of alternatives (IPIPA) is introduced, and an approach based on IPIPA and projection to ranking alternatives is proposed. The method can avoid comparing and ranking interval numbers, and can reflect both the objective information and the DMs subjective preferences.     

考虑混合信息表征和属性交互的目标导向新产品开发方案选择方法

将目标值融入到新产品开发方案选择中,考虑方案属性值达成目标值的情况,有助于企业选择更具竞争力的产品开发方案。针对属性值和目标值的混合信息表征以及属性交互的问题,提出基于目标导向决策分析和k-可加模糊测度的新产品开发方案选择方法。首先,考虑目标值和属性值表示为区间值、模糊数、语言值等混合信息的情形,将其转化为概率密度;结合属性的三种偏好,利用目标导向决策分析计算属性值达成目标值的概率。其次,基于属性交互方向和强度等信息,利用最小方差法识别k-可加模糊测度,进而利用Choquet积分算子集结各属性的目标达成概率作为产品开发方案选择的依据。最后,将方法应用于大型集成电路测试仪的开发方案选择,验证了有效性。     

An interactive decision procedure with multiple attributes under risk

Consider a finite set of alternatives under risk which have multiple attributes. MARPI is an interactive computer-based procedure to find an efficient choice in the sense of linear expected utility. The choice is based on incomplete information about the decision maker's preferences which is elicited and processed in a sequential way. The information includes qualitative properties of the multivariate utility function such as monotonicity, risk aversion, and separability. Further, in case of an additively separable utility function, bounds on the scaling constants are elicited, and preferences (not necessarily indifferences) between sure amounts and lotteries are asked from the decision maker. The lotteries are Bernoulli lotteries generated by MARPI using special strategies. At every stage of the procedure the efficient set of alternatives is determined with respect to the information elicited so far.The procedure has been fully implemented on a PC. The paper exhibits the basic ideas of MARPI and some details of its implementation.     

Variable block search with prior information

Consider the problem of optimizing an unknown, unimodal, univariate function by direct function evaluation. This article replaces the classic criterion for evaluating search strategies, minimax, by one we feel to be more practical, minimum expected final interval of uncertainty. This latter measure of effectiveness is useful when the experimenter has some prior knowledge of the underlying function and can specify a probability distribution on the location of the optimum.We show here that for a large class of search strategies under the uniform distribution, minimax is equivalent to minimum expected final interval length, i.e., the uniform distribution represents a ‘worst-case’ prior. In addition, computational procedures are develooed for determining the optimal search strategy for arbitrary a priori distributions. Several examples are included.     

Systemic decision making in AHP: a Bayesian approach

Systemic decision making is a new approach for dealing with complex multiactor decision making problems in which the actors’ individual preferences on a fixed set of alternatives are incorporated in a holistic view in accordance with the “principle of tolerance”. The new approach integrates all the preferences, even if they are encapsulated in different individual theoretical models or approaches; the only requirement is that they must be expressed as some kind of probability distribution. In this paper, assuming the analytic hierarchy process (AHP) is the multicriteria technique employed to rank alternatives, the authors present a new methodology based on a Bayesian analysis for dealing with AHP systemic decision making in a local context (a single criterion). The approach integrates the individual visions of reality into a collective one by means of a tolerance distribution, which is defined as the weighted geometric mean of the individual preferences expressed as probability distributions. A mathematical justification of this distribution, a study of its statistical properties and a Monte Carlo method for drawing samples are also provided. The paper further presents a number of decisional tools for the evaluation of the acceptance of the tolerance distribution, the construction of tolerance paths that increase representativeness and the extraction of the relevant knowledge of the subjacent multiactor decisional process from a cognitive perspective. Finally, the proposed methodology is applied to the AHP-multiplicative model with lognormal errors and a case study related to a real-life experience in local participatory budgets for the Zaragoza City Council (Spain).     

A multiple criteria decision analytic approach for evaluating acid rain policy choices

Multi-Attribute Utility Theoretic Approach is adopted to analyze the problem of acid rain policy choices for multiple impacted groups (agents) under two conditions: exact and imprecise preference information. The analysis involves the setting up of attributes and feasible policy alternatives. Estimates of acid rain impacts are obtained from secondary data. The emphasis is more on framework which allows explorations of compromises and the identification of acceptable impact areas than on exact results. The strongest message of this paper is that even without specifying exact cardinal preferences some dominance relationships can be established between policies in a real-world setting.     

Fuzzy Mathematical Programming: A Unified Approach Based On Fuzzy Relations

Fuzzy mathematical programming problems (FMP) form a subclass of decision - making problems where preferences between alternatives are described by means of objective function(s) defined on the set of alternatives. The formulation a FMP problem associated with the classical MP problem is presented. Then the concept of a feasible solution and optimal solution of FMP problem are defined. These concepts are based on generalized equality and inequality fuzzy relations. Among others we show that the class of all MP problems with (crisp) parameters can be naturally embedded into the class of FMP problems with fuzzy parameters. We also show that the feasible and optimal solutions being fuzzy sets are convex under some mild assumptions.     

The Analytic Hierarchy Process in an uncertain environment: A simulation approach

Traditionally, decision makers were forced to converge ambiguous judgments to a single point estimate in order to describe a pairwise relationship between alternatives relative to some criterion for use in the Analytic Hierarchy Process (AHP). Since many circumstances exist which make such a convergence difficult, confidence in the outcome of an ensuing AHP synthesis may be reduced. Likewise, when a group of decision makers cannot arrive at a consensus regarding a judgment, some members of the group may simply lose confidence in the overall synthesis if they lack faith in some of the judgments. The AHP utilizes point estimates in order to derive the relative weights of criteria, sub-criteria, and alternatives which govern a decision problem. However, when point estimates are difficult to determine, distributions describing feasible judgments may be more appropriate. Using simulation, we will demonstrate that levels of confidence can be developed, expected weights can be calculated and expected ranks can be determined. It will also be shown that the simulation approach is far more revealing than traditional sensitivity analysis.     

When conflict induces the expression of incomplete preferences

Multicriteria conflict arises in pairwise comparisons, where each alternative outperforms the other one on some criterion, which imposes a trade-off. Comparing two alternatives can be difficult if their respective advantages are of high magnitude (the attribute spread is large). In this paper, we investigate to which extent conflict in a comparison situation can lead decision makers to express incomplete preferences, that is, to refuse to compare the two alternatives, or to be unable to compare them with confidence. We report on an experiment in which subjects expressed preferences on pairs of alternatives involving varying conflicts. Results show that depending on whether the participants are allowed to express incomplete preferences or not, attribute spread has a different effect: a large attribute spread increases the frequency of incomparability statements, when available, while it increases the use of indifference statements when only indifference and preference answers are permitted. These results lead us to derive some implications for preference elicitation methods involving comparison tasks.     

Hybrid bootstrap for mapping quantitative trait loci

The hybrid bootstrap uses resampling ideas to extend the duality approach to interval estimation for a parameter of interest when there are nuisance parameters. The confidence region constructed by the hybrid bootstrap may perform much better than the parametric bootstrap region in situations where the data provide substantial information about the nuisance parameter, but limited information about the parameter of interest. We apply this method to estimate the location of quantitative trait loci (QTL) in interval mapping model. The conditional distribution of quantitative traits, given flanked genetic marker genotypes is often assumed to be the mixture model of two phenotype distributions. The mixing proportions in the model represent the recombination rate between a genetic marker and quantitative trait loci and provides information about the unknown location of the QTL. Since recombination events are unlikely, we will have less information about the location of the QTL than other parameters. This observation makes a hybrid approach to interval estimation for QTL appealing, especially since the necessary distribution theory, which is often a challenge for mixture models, can be handled by bootstrap simulation.     

兼顾效益和成功率的区间数决策矩阵的一种综合排序方法

目前的区间数的决策矩阵排序方法都只根据效益型准则进行的,而没有考虑到方案实施的成功率这个重要准则。本文提出了区间数的可行度概念,在此基础上给出了基于可行度的区间数决策矩阵的排序方法;然后提出了一种兼顾效益和成功率的区间数决策矩阵排序的综合评价方法,该方法能融合多方面的决策信息。文中给出了一个实例说明本法的有效性和可行性。     

A hybrid inexact-stochastic water management model

An inexact-stochastic water management (ISWM) model is proposed and applied to a case study of water quality management within an agricultural system. The model is based on an inexact chance-constrained programming (ICCP) method, which improves upon the existing inexact and stochastic programming approaches by allowing both distribution information in B and uncertainties in A and C to be effectively incorporated within its optimization process. In its solution process, the ICCP model (under a given pi level) is first transformed into two deterministic submodels, which correspond to the upper and lower bounds for the desired objective function value. This transformation process is based on an interactive algorithm, which is different from normal interval analysis or best/worst case analysis. Interval solutions, which are feasible and stable in the given decision space, can then be obtained by solving the two submodels sequentially. Thus, decision alternatives can be generated by adjusting decision variable values within their solution intervals. The obtained ICCP solutions are also useful for decision makers to obtain insight regarding tradeoffs between environmental and economic objectives and between increased certainties and decreased safeties (or increased risks). Results of the case study indicate that useful solutions for the planning of agricultural activities in the water quality management system have been obtained. A number of decision alternatives have been generated and analyzed based on projected applicable conditions. Generally, some alternatives can be considered when water quality objective is given priority, while the others may provide compromises between environmental and economic considerations. The above alternatives represent various options between environmental and economic tradeoffs. Willingness to accept low agricultural income will guarantee meeting the water quality objectives. A strong desire to acquire high agricultural income will run into the risk of violating water quality constraints.     

Reasoning with various kinds of preferences: logic,non-monotonicity,and algorithms

As systems dealing with preferences become more sophisticated, it becomes essential to deal with various kinds of preference statements and their interaction. We introduce a non-monotonic logic distinguishing sixteen kinds of preferences, ranging from strict to loose and from careful to opportunistic, and two kinds of ways to deal with uncertainty, either optimistically or pessimistically. The classification of the various kinds of preferences is inspired by a hypothetical agent comparing the two alternatives of a preference statement. The optimistic and pessimistic way of dealing with uncertainty correspond on the one hand to considering either the best or the worst states in the comparison of the two alternatives of a preference statement, and on the other hand to the calculation of least or most specific “distinguished” preference orders from a set of preference statements. We show that each way to calculate distinguished preference orders is compatible with eight kinds of preferences, in the sense that it calculates a unique distinguished preference order for a set of such preference statements, and we provide efficient algorithms that calculate these unique distinguished preference orders. In general, optimistic kinds of preferences are compatible with optimism in calculating distinguished preference orders, and pessimistic kinds of preferences are compatible with pessimism in calculating distinguished preference orders. However, these two sets of eight kinds of preferences are not exclusive, such that some kinds of preferences can be used in both ways to calculate distinguished preference orders, and other kinds of preferences cannot be used in either of them. We also consider the merging of optimistically and pessimistically constructed distinguished preferences orders.     

On the genus distributions of wheels and of related graphs

We are concerned with families of graphs in which there is a single root-vertex ofunbounded valence, and in which, however, there is a uniform upper bound for the valences of all the other vertices. Using a result of Zagier, we obtain formulas and recursions for the genus distributions of several such families, including the wheel graphs. We show that the region distribution of a wheel graph is approximately proportional to the sequence of Stirling numbers of the first kind. Stahl has previously obtained such a result for embeddings in surfaces whose genus is relatively near to the maximum genus. Here, we generalize Stahl’s result to the entire genus distributions of wheels. Moreover, we derive the genus distributions for four other graph families that have some similarities to wheels.     

Simplification of joint confidence regions for the parameters of the Pareto distribution

The Pareto distribution is an important distribution in statistics, which has been widely used in economics to model the distribution of incomes. Separate interval estimations for parameters of the Pareto distribution have been well established in the literature. For a type-II right censored sample, Chen (Metrika 44:191–197, 1996) proposed a joint confidence region for the parameters. Wu (Comput Stat Data Anal 52:3779–3788, 2008) derived a joint confidence region for any type-II censored sample, but its computation is difficult. Both Wu’s and Chen’s results are simplified in this paper, and some errors are corrected. The methodology used in this article can be applied to other distributions, as long as the underlying distribution can be transformed to the standard exponential distribution in a simple way.     

Classifying efficient alternatives in SMAA using cross confidence factors

Stochastic multicriteria acceptability analysis (SMAA) is a family of methods for aiding multicriteria group decision making. These methods are based on exploring the weight space in order to describe the preferences that make each alternative the most preferred one. The main results of the analysis are rank acceptability indices, central weight vectors and confidence factors for different alternatives. The rank acceptability indices describe the variety of different preferences resulting in a certain rank for an alternative; the central weight vectors represent the typical preferences favouring each alternative; and the confidence factors measure whether the criteria data are sufficiently accurate for making an informed decision.In some cases, when the problem involves a large number of efficient alternatives, the analysis may fail to discriminate between them. This situation is revealed by low confidence factors. In this paper we develop cross confidence factors, which are based on computing confidence factors for alternatives using each other’s central weight vectors. The cross confidence factors can be used for classifying efficient alternatives into sets of similar and competing alternatives. These sets are related to the concept of reference sets in Data Envelopment Analysis (DEA), but generalized for stochastic models. Forming these sets is useful when trying to identify one or more most preferred alternatives, or suitable compromise alternatives. The reference sets can also be used for evaluating whether criteria need to be measured more accurately, and at which alternatives the measurements should be focused. This may cause considerable savings in measurement costs. We demonstrate the use of the cross confidence factors and reference sets using a real-life example.