A Local Adjustment Method to Improve Multiplicative Consistency of Fuzzy Reciprocal Preference Relations

Date

2023-05-23

Advisors

Journal Title

Journal ISSN

ISSN

2168-2216

Volume Title

Publisher

IEEE

Type

Article

Peer reviewed

Yes

Abstract

Preferences that verify the transitivity property are usually referred to as rational or consistent preferences. Existent methods to improve the consistency of inconsistent fuzzy reciprocal preference relations (FPRs) fail to retain the original preference values because they always derive a new FPR. This article presents a new inconsistency identification and modification (IIM) method to detect and rectify only the most inconsistent elements of an inconsistent FPR. As such, the proposed IIM can be considered a local adjustment method to improve multiplicative consistency (MC) of FPRs. The case of inconsistent FPRs with missing values, i.e., incomplete FPRs, is addressed with the estimation of the missing preferences with a constrained nonlinear optimization model by the application of the IIM method. The implementation process of the proposed algorithms is illustrated with numerical examples. Simulation experiments and comparisons with existent methods are also included to show that the new method requires fewer iterations than existent methods to improve the MC of FPRs and achieves better MC level, while preserving the original preference information as much as possible than the existent methods. Thus, the results presented in this article demonstrate the correctness, effectiveness, and robustness of the proposed method.

Description

The file attached to this record is the author's final peer reviewed version. The Publisher's final version can be found by following the DOI link.

Keywords

Indexes, Numerical models, Robustness, Optimization, Nonlinear distortion, Cybernetics, Computer science

Citation

Xu, Y., Wang, Q., Chiclana, F. and Herrera-Viedma, E. (2023) A Local Adjustment Method to Improve Multiplicative Consistency of Fuzzy Reciprocal Preference Relations. IEEE Transactions on Systems, Man and Cybernetics: Systems,

Rights

Research Institute

Institute of Artificial Intelligence (IAI)