Disutility Entropy in mMulti-attribute Utility Analysis


Suetcue M.

COMPUTERS & INDUSTRIAL ENGINEERING, vol.169, 2022 (Journal Indexed in SCI) identifier identifier

  • Publication Type: Article / Article
  • Volume: 169
  • Publication Date: 2022
  • Doi Number: 10.1016/j.cie.2022.108189
  • Title of Journal : COMPUTERS & INDUSTRIAL ENGINEERING
  • Keywords: Multiattribute Utility Function, Cross Derivative, Shannon Entropy, Cumulative Residual Entropy, Maximum Entropy, Decision making with partial information, DISTRIBUTIONS, INFORMATION

Abstract

In this paper, we present an alternative formulation of utility entropy called disutility entropy. Previous entropy measurements in the literature use utility density function in maximum entropy formulations. Also, in most of the cases, the sign of the cross derivative of utility functions makes it impossible to apply utility entropy for more than one attribute cases. To simplify entropy measurement and relieve some burden of this task, in this paper, we present how to use multiattribute utility functions in utility entropy formulation. We show the applicability of our proposed approach and how to apply the disutility entropy approach with given constraints to singleattribute, bi-attribute, and multiattribute utility functions. Therefore, the usefulness and feasibility of the proposed method in multiattribute utility theory field is improved. We finally discuss and interpret the application of maximum disutility entropy through several examples to illustrate the new proposed approach.