Abstract: In the process of cloud manufacturing mode,postponement strategy is still an effective strategy to reduce supply chain risks.Existing studies on postponement are often based on fixed product architecture,with limited attention to the coupling among product family configuration,postponement contracting decisions and consumer behaviors in cloud manufacturing.Thus,a hierarchical optimization approach was proposed,and a three-level hierarchical optimization model with a single manufacturer as the leader,multiple distributors as the followers and multiple consumers as the sub-followers is formulated.The optimization objective of the manufacturer and each distributor was to maximize the net profit,and the optimization objective of each consumer group was to maximize the utility value.Analytical solutions were developed incorporating a nested genetic algorithm to solve the model.A case study of laser printer product family postponement production was reported to illustrate the feasibility of the proposed approach.Some important managerial insights were obtained from the sensitivity analyses of the proposed postponement cost and preference parameters,and the objective value situation of each decision maker in different regions was obtained by the sensitivity analysis of the union of the two parameters.

Key words: cloud manufacturing, postponement strategy, product family, consumer behaviors, three-level hierarchical optimization

摘要： 在当今云制造模式的进程中,延迟策略仍然是企业降低供应链风险的一种有效策略。然而,当前对延迟的大多数研究往往是基于事先已经固定好的产品架构,且较少关注到在面向云制造模式下的产品族配置、延迟承包决策以及消费者行为之间存在的内在固有耦合关系。因此,提出对这三者的一种主从优化方法,并构建了一个以单个制造商为主、多个分销商为从,以及多个消费者群体为子从的三层主从优化模型。其中,制造商和每个分销商的优化目标均是最大化自身的净利润,而每个消费者群体的优化目标则是最大化自身的效用值。采用分析法和嵌套式的遗传算法求解该模型,以激光打印机产品族延迟生产案例验证方法的可行性。最后,通过对提出的延迟成本和偏好两个参数的灵敏度分析实验得出了一些重要管理启示,并通过对联合这两个参数的灵敏度分析实验获得了各决策主体的收益区域分布范围。

关键词: 云制造, 延迟策略, 产品族, 消费者行为, 三层主从优化