A1 Journal article (refereed)
Comparing reference point based interactive multiobjective optimization methods without a human decision maker (2023)

Chen, L., Miettinen, K., Xin, B., & Ojalehto, V. (2023). Comparing reference point based interactive multiobjective optimization methods without a human decision maker. Journal of Global Optimization, 85(3), 757-788. https://doi.org/10.1007/s10898-022-01230-3

JYU authors or editors

Miettinen, Kaisa
Ojalehto, Vesa

Publication details

All authors or editors: Chen, Lu; Miettinen, Kaisa; Xin, Bin; Ojalehto, Vesa

Journal or series: Journal of Global Optimization

ISSN: 0925-5001

eISSN: 1573-2916

Publication year: 2023

Publication date: 24/09/2022

Volume: 85

Issue number: 3

Pages range: 757-788

Publisher: Springer

Publication country: Netherlands

Publication language: English

DOI: https://doi.org/10.1007/s10898-022-01230-3

Publication open access: Openly available

Publication channel open access: Partially open access channel

Publication is parallel published (JYX): https://jyx.jyu.fi/handle/123456789/83316

Abstract

Interactive multiobjective optimization methods have proven promising in solving optimization problems with conflicting objectives since they iteratively incorporate preference information of a decision maker in the search for the most preferred solution. To find the appropriate interactive method for various needs involves analysis of the strengths and weaknesses. However, extensive analysis with human decision makers may be too costly and for that reason, we propose an artificial decision maker to compare a class of popular interactive multiobjective optimization methods, i.e., reference point based methods. Without involving any human decision makers, the artificial decision maker works automatically to interact with different methods to be compared and evaluate the final results. It makes a difference between a learning phase and a decision phase, that is, learns about the problem based on information acquired to identify a region of interest and refines solutions in that region to find a final solution, respectively. We adopt different types of utility functions to evaluation solutions, present corresponding performance indicators and propose two examples of artificial decision makers. A series of experiments on benchmark test problems and a water resources planning problem is conducted to demonstrate how the proposed artificial decision makers can be used to compare reference point based methods.

Keywords: multi-objective optimisation; decision making; decision support systems; interactivity; machine learning

Free keywords: multicriteria optimization; interactive multiobjective optimization; learning phase; decision phase; performance comparison; reference point

Fields of science: