Social Simulation for Analyzing Product Recall Systems Using Co-Evolution Model with Price Competition

• 作者: Kazuo Furuta Taro Kanno Tetsuroh Watanabe
• 来源期刊: 数学和系统科学:英文版
• 学科: 数学
• 关键词: Multi-agent simulation artificial SOCIETY MULTI-OBJECTIVE optimization EVOLUTIONARY COMPUTATION genetic PROGRAMMING
• 年，卷（期）: 2018,（2）
• 页码范围: 25-43
• 页数: 19页
• 分类号: O1

摘要

In recent years,accidents and product recalls caused by product defects have become important problems in numerous industries worldwide.Nevertheless,most existing studies have examined product recalls using empirical approaches.To improve product recall systems,we studied social simulation using a multi-agent system with a co-evolution model.This research is important because empirical approaches are no longer adequate for complex and diverse modern societies.Discussions using quantitative and predictive approaches,including agent-based simulation,are therefore expected.For this study,we used a Layered Co-evolution Model to reflect situations of the real society using producer agents and consumer agents.Additionally,we applied multi-objective optimization techniques to introduce price competition situations into an artificial society.We conducted a simulation experiment,from which we discovered the possibilities that cost reduction for huge-scale product recalls is efficient,and that punishment of producers that conduct no product recalls can benefit consumers.We believe this work can contribute to supporting not only government staff for improving product recall systems,but also executive officers of product companies for deliberating their strategies of recall decisions.