Abstract: To solve the problem of comprehensive optimization of conflicts between quality and cost in the process of collaborative design,the quality cost multi-objective optimization problems was transformed into multi-objective game decision problems under dynamic alliances by introducing the game theory,and a game model that based on historical data was established.The game effect matrix formed by information indicated that due to the incompleteness of historical data,the multiple regression analysis was adopted to complete the game utility matrix,and a game model of conflict resolution based on historical data information was proposed.The game matrix was solved through Shapley value algorithm.The allocation solution of conflict was used to represent the best quality and cost comprehensiveness.To seek the contribution of each game player and verify the reliability of equilibrium results obtained by Shapley values,a quantitative assessment method for the contribution of Shapley values was proposed.An assembly process of a lathe spindle was used as an example to demonstrate the feasibility and effectiveness of the proposed method.

Key words: collaborative design, conflict resolution, multiple regression, game model, historical data, incompleteness

摘要： 为解决需方参与的协同设计过程中存在的质量与成本综合优化冲突问题,引入博弈论,将协同设计过程中质量成本多目标优化问题转换为动态联盟下的多目标博弈决策问题,并构建博弈模型。博弈模型亦可由历史数据信息构成的博弈效应矩阵表示,由于历史数据具有残缺性,采用多元回归分析将博弈效用矩阵完备化,提出一种基于历史数据信息的冲突消解博弈模型;然后通过Shapley值算法求解博弈矩阵,根据求解得到的冲突关系分配方案表示“质量和成本综合性最优”的一种贡献度估计,为了寻求各博弈方的贡献度,验证Shapley值所求均衡结果的可靠性,提出一种针对Shapley值贡献度的定量评估方法。最后以某车床主轴装配过程为例,验证了该方法的可行性和有效性。

关键词: 协同设计, 冲突消解, 多元回归, 博弈模型, 历史数据, 残缺性