运筹与管理 ›› 2017, Vol. 26 ›› Issue (3): 86-91.DOI: 10.12005/orms.2017.0061

• 理论分析与方法探讨 • 上一篇    下一篇

基于NSGAII的协同产品开发项目自动协商决策

黄敏镁1, 袁际军2, 曹亮3   

  1. 1.华南师范大学 公共管理学院,广东 广州 510006;
    2.广东财经大学 金融学院,广东 广州 510320;
    3.中国电信股份有限公司广东分公司 销售及渠道拓展事业部,广东 广州 510180
  • 收稿日期:2015-04-19 出版日期:2017-03-25
  • 作者简介:黄敏镁(1979-),女,副教授,博士,研究方向:运营系统建模与智能优化;袁际军(1975-),男,副教授,博士,研究方向:大规模定制、供应链金融;曹亮(1980-),男,硕士,研究方向:企业管理。
  • 基金资助:
    国家自然科学基金青年科学基金项目(71102146);国家自然科学基金面上项目(71272084);教育部人文社会科学研究青年基金资助项目(大规模定制下时变客户需求的产品配置动态更新优化与管理应用研究:12YJC630278);广东财经大学校级科研项目(11YB63001)

NSGAII Based Automated Negotiation in Collaborative Product Development Project

HUANG Min-mei1, YUAN Ji-jun2, CAO Liang3   

  1. 1.School of Public Administration, South China Normal University, Guangzhou 510006, China;
    2.School of Finance, Guangdong University of Finance & Economics, Guangzhou 510320, China;
    3.Sales and Distribution Service Department, Guangdong Provincial Corporation CHINA TELECOM, Guangzhou 510180, China
  • Received:2015-04-19 Online:2017-03-25

摘要: 供应链环境下的协同产品开发项目中,核心企业与合作企业之间,需要对项目工期、预算和质量进行协商。科学高效的协商决策方法有助于协同产品开发项目参与各方实现双赢的协商目标。采用多Agent系统建立了协商问题的框架,构建了协商Agent以自身效用最大化和与对方建议相似度最大化为目标的多目标反建议模型;提出了基于NSGAII的自动协商决策算法,算法采用了正整数和小数混合的实数编码方式,并在遗传操作中增加了约束限制,剔除不可行个体。计算机仿真实验结果表明了算法对提高协商成功率和协商效果的有效性,说明了协商Agent可接受的最低效用值对于协商结果的影响。

关键词: 供应链管理, 自动协商, 多目标优化, 协同产品开发, NSGAII, 多Agent系统

Abstract: The supply chain core enterprise needs to negotiate with cooperative enterprises about the schedule, budget and quality of collaborative product development project. Scientific and efficient negotiation decision method is helpful for the cooperative product development project participants to achieve a win-win objective. The framework of negotiation problem with multi-agent system is established, the multi-objective counter proposal model of the negotiation agent to both maximizes its own utility and similarity with the opponent’s proposal is constructed, and the automatic negotiation decision algorithm based on NSGAII is proposed. The positive integer and decimal mixed real number coding is used in NSGAII, and non feasible individual is removed by adding constraints in the genetic operations. The experimental results show that the algorithm can improve the success rate and the joint utility of negotiation. It also illustrates the impact of the lowest acceptable utility of negotiation agent on negotiation result.

Key words: supply chain management, automated negotiation, multi-objective optimization, collaborative product development, NSGAII, multi-agent system

中图分类号: