Abstract:  Based on design structure matrix， the process coupling analyses were difficult to effectively solve the multi-agent decision optimization problems in the change of complex product designs， such as the coupling of process correlation， the readjust of designs， and the reallocation of design resources. Moreover， the theory of decision optimization with the agent as a completely rational agent deviated greatly from the specific practices. An optimization method of complex product design change decision was proposed considering the psychological perception of the agent under process correlation. The prospect theory was used to construct the complex product design change decision optimization model with the objective function of minimizing the psychological disturbance of the design agent repeated execution and the delay period of design change. A thing and season-based adaptive multi-objective dynamic scheduling algorithm  was proposed by using two-layer coding method combining process and design agent. At last， the rationality and validity of the above decision model and method were verified by the comparison experiments and result analyses of the example simulation. 

Key words: complex product, design change decision optimization, process correlation, prospect theory, agent psychological perception

摘要： 基于设计结构矩阵的工序耦合分析难以有效地解决复杂产品设计变更中工序关联解耦、设计再调整、设计资源再配置等多主体决策优化问题，且将主体作为完全理性行为人的决策优化理论与具体实践亦偏差较大。对此，提出了工序关联下考虑主体心理感知的复杂产品设计变更决策优化方法。运用前景理论构建以设计主体对工序重复执行和设计变更延期心理感知扰动最小为目标函数的复杂产品设计变更决策优化模型；采用工序与设计主体相结合的双层编码方法，提出了基于事件与周期的自适应多目标动态调度算法；最后，通过算例仿真的对比实验及结果分析，验证上述决策模型与方法在不同情况下的合理性和有效性。

关键词: 复杂产品, 设计变更决策优化, 工序关联, 前景理论, 主体心理感知