Abstract: In the globalization process,postponement strategy is still an effective way for enterprises to cope with the diversity and uncertainty of consumer needs.Most of the current researches on postponement are often based on a fixed product family architecture at the outset,with limited attention to the inherent interaction between Product Family Architecture Design (PFAD) and Supply Chain Postponement Decisions (SCPD).Therefore,a hierarchical interactive optimization approach is proposed.By constructing the internal hierarchical interactive evaluation mechanism,a nonlinear bilevel programming model with PFAD as the upper-level optimization and SCPD as the lower-level optimization is formulated.As the leader,the designer designs the product family architecture and selects the type of postponement product module to maximize the customer utility per unit cost.The suppliers,manufacturers,postponement contractors,and distributors play the roles of followers and optimize the postponement manufacturing process of the product family individually with the goal of minimizing their operational costs.A bi-level nested genetic algorithm is designed to solve the proposed optimization model.A case study of smart refrigerator product family postponement manufacturing is introduced to verify the feasibility of the proposed hierarchical interactive optimization approach.Some managerial insights are obtained from the sensitivity analysis on the parameter θ in the MNL choice rule.

Key words: product family architecture, supply chain postponement, hierarchical interactive optimization, nonlinear bilevel programming model, nested genetic algorithm

摘要： 在当今市场全球化的进程中,延迟策略是企业应对顾客需求多样化和不确定性的一种有效方式。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族架构,且较少关注到在产品族架构设计与整条供应链延迟决策之间存在的内在交互影响关系,因此,提出对二者的一种主从交互优化方法。通过构建在它们之间的主从交互决策机制,建立了一个以产品族架构设计为主、供应链延迟决策为从的非线性双层规划模型。模型上层是开发商设计产品族架构和决策其中的延迟产品模块类型,从而最大化单位成本的顾客效用;下层由多个供应商、多个制造商、多个延迟承包商、以及多个分销商分别优化产品族的延迟制造过程,并以各自的运营成本最小化为目标。针对模型求解的复杂性,设计了一种嵌套遗传算法进行求解。以智能冰箱产品族延迟生产案例验证所提优化模型和求解算法的可行性,并通过对MNL选择规则中的参数θ进行灵敏度分析实验得出了一些管理启示。

关键词: 产品族架构, 供应链延迟, 主从交互优化, 非线性双层规划模型, 嵌套遗传算法