Multi-Objective Optimization Using Evolutionary Algorithms
Multi-Objective Optimization Using Evolutionary Algorithms
Hi-index | 0.00 |
This paper proposes the interactive system that can help humans to understand the trade-off relationship of Pareto optimal solutions (e.g., good products from a certain aspect) in multi-dimensional space. For this purpose, the following two methods are proposed from the viewpoint of the number of evaluation criteria which should be considered by a user at one time: (i) the two fixed evaluation criteria are employed to evaluate the solutions; and (ii) some evaluation criteria selected by a user (i.e., the number of the evaluation criteria is varied by a user) are employed to evaluate them. To investigate the effectiveness of our proposed system employing either of two methods, we conduct human subject experiments on the motor selection problem and have revealed the following implications: (i) the proposed system based on the two fixed evaluation criteria contributes to helping users to find better motors in terms of all the evaluation criteria, while (ii) the proposed system based on the selected evaluation criteria is more effective to help users to understand Pareto optimal solutions when more evaluation criteria need to be considered.