Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
6896791 | European Journal of Operational Research | 2015 | 36 Pages |
Abstract
Various preference-based multi-objective evolutionary algorithms have been developed to help a decision-maker search for his/her preferred solutions to multi-objective problems. In most of the existing approaches the decision-maker preferences are formulated either by mathematical expressions such as the utility function or simply by numerical values such as aspiration levels and weights. However, in some sense a decision-maker may find it easier to specify preferences visually by drawing rather than using numbers. This paper introduces such a method, namely, the brushing technique. Using this technique the decision-maker can specify his/her preferences easily by drawing in the objective space. Combining the brushing technique with one existing algorithm PICEA-g, we present a novel approach named iPICEA-g for an interactive decision-making. The performance of iPICEA-g is tested on a set of benchmark problems and is shown to be good.
Related Topics
Physical Sciences and Engineering
Computer Science
Computer Science (General)
Authors
Rui Wang, Robin C. Purshouse, Ioannis Giagkiozis, Peter J. Fleming,