Abstract: To coordinate preference conflicts among customers and designers from different fields in the process of product collaborative creative design, the original cause of conflict was analyzed, binary linguistic analysis was used to integrate agents'evaluation information with multi-granularity. Then, a multi-agent conflict coordination model, which aiming at maximizing the satisfaction degree of agents, was constructed. Furthermore, an improved adaptive particle swarm optimization algorithm was proposed to find the optimal solution to the model. In this algorithm, inertia weight was dynamically changed according to current evolution speed, diversity and iterations of the swarm, which provided the algorithm with effective dynamic adaptability. Finally, the feasibility and validity of the proposed model and algorithm were validated through an example of mobile phone design optimization.

Key words: collaborative creative design, customer involvement, multi-agent, conflict coordination, two-tuple linguistic, particle swarm optimization

摘要： 在分析协同产品创新设计过程中客户、多领域设计人员等多角色主体间众多冲突原因的基础上，为协调各设计主体在设计目标上的偏好冲突，提出了运用二元语义分析来集结各主体多粒度评价语言信息的方法，并以满意度最大为优化目标建立了多主体冲突协调模型。针对该模型，提出了改进的自适应粒子群算法进行优化求解。算法中，粒子的惯性权值随当前粒子群的进化速度和多样性动态改变，从而使算法具有动态自适应性。通过某手机设计优化实例验证了所提模型与算法的可行性和有效性。

关键词: 协同创新设计, 客户参与, 多主体, 冲突协调, 二元语义, 粒子群优化