Decision aiding model with entropy-based subjective utility
Introduction
In the real world, it is often difficult to find an alternative that is the best with respect to all the desirable criteria. Hence, the decision-maker (DM) makes a compromise among the conflicting criteria in order to determine his best choice. Most approaches explain this compromise through the concept of utility. It is averred in [37] that the utility maximizing behaviour is the key characteristic of human decision making. For instance, a consumer evaluates the utility of various alternatives against a set of desired properties (or criteria), and chooses the one giving him the maximum utility. Similarly, an organization determines the best production technology or the most suitable supplier; a student chooses a career stream among many; an individual chooses whether to be an employee or an entrepreneur. Any decision or choice by a DM is an outcome of a complex cognitive process that broadly involves:
- •
evaluation of utility for each of the criteria, and
- •
aggregation of such utility values to arrive at the net alternative utility.
Multi criteria decision aiding (MCDA) has become an important and pervasive subfield of operations research, which gives a systematic process to aid a decision-maker (DM) in choosing the best alternative. In the recent times, MCDA has received a lot of attention in the literature [32], [43]. The evaluation of subjective utility values is of crucial importance for a MCDA approach. In this regard, the works of Grable and Lytton [20], and Clemen and Reilly [15] are notable. Both are about the identification of the value of the utility function at a discrete set of points. The utility function is then obtained through a simple interpolation. The attempts in [1], [2], [39] to elicit the underlying utility function behind a DM’s choices have received attention. However, all these approaches require an extensive interaction with the DM, which makes them difficult to automate. Moreover, these methods need an ordinal preference for each of the alternatives. This restricts their usefulness in practical applications.
Some approaches have attempted to represent the utility function in terms of the parameters of aggregation operators, such as weighted averaging or Choquet integral [14]. In [40], the weight vector of weighting averaging operator is learned, while the parameter (fuzzy measure) of Choquet integral is resolved for a few datasets in supervised learning setup in [36]. Thus, these methods deduce parameter(s) that best represent(s) a given dataset. These methods are good at explaining the choices for a given dataset, but not so helpful in our goal of inferring the utility of a particular criterion value. Secondly, these approaches are based on the pairwise comparisions of the alternatives. This is achieved either by asking a DM directly about his preferences, or indirectly inducing through the DM’s pre-given ranking of the alternatives. In both the cases, it is quite a time-consuming and cumbersome task, when the number of alternatives exceeds 5 or 6.
In a similar vein, a few interesting approaches have appeared on optimization-based utility elicitation from a given dataset [8], [10]. Another set of studies considers that the knowledge of a DM is incomplete to precisely evaluate an alternative against a criterion [7], [29], [41], [45]. Different methods have been applied to deal with the incomplete information. In [12], [13], the authors make use of a probability distribution over different utility functions, and the final choice is determined by the expected utility averaged over such a distribution. However, these studies assume to be given with a database of partially elicited utility functions, which, in practice, are quite difficult to gather.
In [11], a set of all those utility functions is constructed, which do not contradict the given information. The alternative that achieves minimum worst-case regret, in a mixed-integer programming approach, forms the best choice. Bayesian methods have been explored to quantify uncertainty about preferences in [23]. The studies [7], [16], [24] solve convex optimization problems of real-valued values to find the best choice. The utility and expected utility theory have been built upon for decision-making under uncertainty in [6] and [17], respectively. The concept of probabilistic choice model has been propounded in [38].
These methods, despite their theoretical soundness, are not convenient to solve real MCDA problems, as they suffer from three fundamental limitations :
- (i)
they do not give the utility for a particular value. In practice, a criterion value is perceived differently by different individuals, and hence each DM evaluates a different utility (desirability) for a criterion value. For instance, length of a car (say 5000 mm), would be perceived differently by different individuals. While, it might be usual (normal) for one individual , it may be considered as too grand by another. With the extant methods, it is difficult to quantify this utility value that is specific to a DM.
- (ii)
they are context-specific. A DM’s choices may completely differ with the availability of more alternatives that may be significantly better than the earlier alternatives, on offer. For instance, in the case of car selection, a new promising model often outshines the extant bestsellers. Also, quite often, a DM’s information regarding the number of criteria is often incomplete, which results in a modified utility evaluation in the face of new information. It is difficult to model such variations with most of the extant methods.
- (iii)
they do not focus on the individualistic aspects of the aggregation process. Different individuals have different tolerance for an imperfect criterion. A tolerant DM displays a degree of compensation in the determination of the net alternative utility, in the sense that a poor utility score for a criterion can be compensated by a good utility score for another criterion. This degree of compensation varies from DM to DM. A perfectionist would have the expected utility as the minimum of the criteria utility values (maximin/ANDlike), displaying a nil compensation, whereas an extreme tolerant DM with maximum compensation has his expected utility towards the maximum of the criteria utility values (maximax/ORlike).
These limitations of the extant approaches significantly restrict their application in the real world problems. Hence, we are inspired to develop a more practical and systematic utility representation approach. The paper is organized as follows. Section 2 discusses the MCDA paradigm, and builds upon the motivation and contributions of the study. In Section 3, we look deeper at the concept of entropy to explore its usefulness to represent the subjective utility of a decision-maker. Section 4 gives a method to quantify subjective utility using the concept of entropy. In Section 5, we present a MCDA approach using the proposed entropy-based model. Section 6 illustrates the usefulness of the proposed concepts through a real application. Section 7 concludes the paper.
Section snippets
MCDA Paradigm
The fundamental framework of MCDA is given in Fig. 1. The DMs can be individuals, committees, households, organizations, countries, or any similar decision-making unit, and the criteria are the observable characteristics of an alternative. The alternatives refer to the various competing options such as different products, courses, suppliers, strategies, etc. The set of alternatives is denoted by that is also referred to as choice-set.
A DM evaluates different alternatives
Subjective expected utility in the context of MCDA
The concept of subjective expected utility has been first propounded in [34], extending the works of Neumann and Morgenstern [42]. Since then, it has received a wide attention [6], [34]. It considers that there exists subjective probabilities corresponding to different possible outcomes, each of which is associated with a utility value where refers to a set of n possible outcomes (also referred to as prospects) for an uncertain alternative. The expected utility ui for
Entropy-based utilitarian model
We first motivate the use of Hanman-Anirban entropy, followed by a presentation of our model.
A novel MCDA method
In this section, based on the proposed method to deduce the subjective utility, we weave a MCDA approach to determine the best alternative.
Car-buying: A multi-criteria decision making problem
Buying a car is a typical example of multi-criteria decision making, in which a potential buyer compares several options (alternatives) to select the one that suits him the most. The buyer evaluates the various options against a set of desirable criteria, such as engine-power, mileage, after-sales service, cylinder-volume, sale-price, aesthetic, comfort etc. We give a diagrammatic representation of a general buying decision in Fig. 3. The decision data is given as the quadruple where
Conclusions
The proposed MCDA model gives a systematic approach to aid a DM in determining his best alternative. The proposed approach closely mimics a real world MCDA process, in which a DM first evaluates the subjective criteria utility values, aggregates such utility values for each alternative, and then chooses the alternative with the greatest utility. A DM’s subjective utilitarian model towards a criterion is captured through an entropic framework, considering its entire range of the values for the
References (45)
Compensative weighted averaging operators
Appl. Soft. Comput.
(2015)- et al.
An elicitation of utility for quality of life under prospect theory
J. Health Econ.
(2016) - et al.
A new entropy function for feature extraction with the refined scores as a classifier for the unconstrained ear verification
J. Electr. Syst. Inf. Technol.
(2017) - et al.
Constraint-based optimization and utility elicitation using the minimax decision criterion
Artif. Intell.
(2006) - et al.
Expected utility theory without the completeness axiom
J. Econ. Theory.
(2004) - et al.
Financial risk tolerance revisited: the development of a risk assessment instrument
Financ. Serv. Rev.
(1999) - et al.
Question selection for multi-attribute decision-aiding
Eur. J. Oper. Res.
(2003) Fuzzy entropy and conditioning
Inf. Sci.
(1986)- et al.
Utility representation of an incomplete and nontransitive preference relation
J. Econ. Theory
(2016) On measures of fuzziness
Fuzzy Sets Syst.
(1989)
Entropy-optimal weight constraint elicitation with additive multi-attribute utility models
Omega (Westport)
Deriving rankings from incomplete preference information: a comparison of different approaches
Eur. J. Oper. Res.
A chi-square method for priority derivation in group decision making with incomplete reciprocal preference relations
Inf. Sci.
Maximum entropy utility
Ope. Res.
Is there one unifying concept of utility?an experimental comparison of utility under risk and utility over time
Manage. Sci.
Rough information set and its applications in decision making
IEEE Trans. Fuzzy Syst.
Representing uncertainty with information sets
IEEE Trans. Fuzzy Syst.
A definition of subjective probability
Ann. Math. Stat.
Decision making under uncertainty when preference information is incomplete
Manage. Sci.
A parameter-free elicitation of the probability weighting function in medical decision analysis
Manage. Sci.
Utilities as random variables: density estimation and structure discovery
Proceedings of the Sixteenth conference on Uncertainty in artificial intelligence (UAI-00)
Making rational decisions using adaptive utility elicitation
Proceedings of the Seventeenth National Conference on Artificial Intelligence (AAAI-00), Austin, TX
Cited by (11)
An entropy framework for randomness and fuzziness
2024, Expert Systems with ApplicationsA framework for the evaluation and selection of shoreline surface washing agents in oil spill response
2021, Journal of Environmental ManagementInformation theory and player archetype choice in Hearthstone
2021, Information SciencesCitation Excerpt :Measuring the amount of randomness displayed in the meta over time enables appreciation of how balanced active archetypes are, how effective recent changes were, and ultimately helps characterise the state of the Hearthstone system. Past examples of Shannon entropy offering insights into complex systems include: Miranskyy et al. [28] who used entropy measures to understand and compare rare events in defective software; Cao et al. [6] who developed Shannon entropy-based measures on graphs to understand the underlying complexity of graph families; and Aggarwal [1] who recently applied generalised Shannon entropy measures to provide decision support in the face of multiple criteria that were often conflicting. Associated with information entropy is the concept of system criticality [21], sometimes referred to as the edge of chaos [35].
Deep Rényi entropy graph kernel
2021, Pattern RecognitionCitation Excerpt :For example, George et al. [15] provided a comprehensive analysis, making the most of the entropy rate created by a Markov chain over a connected graph of order n and subject to a prescribed stationary distribution. Aggarwal et al. [16] presented an entropy-based method to model a decision-maker’s subjective utility for a criterion value. Zhou et al. [17] showed an ave-entropy based weight assignment process considering the risk preference of decision making to deal with manifold attribute decision making problems, where indefinite subjective judgments such as belief distributions are included.
On modeling ambiguity through entropy
2023, International Transactions in Operational ResearchAssessing Potential of Organizations with Fuzzy Entropy
2023, Operations Research Forum