【作者】 吴军；

【导师】 杜纲；

【作者基本信息】 天津大学 ， 管理科学与工程， 2020， 博士

【摘要】 开放制造旨在基于开放资源的原则,以开放的、协作的和分布式的方式生产产品,它不但促进了不同企业之间的合作,而且支持在整个价值链上的知识和服务的共享和交换。延迟策略旨在将产品分化推迟到供应链末端。开放制造使得在产品设计阶段便可对更多的延迟活动进行选择匹配和优化,并且延迟策略的成功实施,也必须要在产品设计的早期阶段就考虑到它的应用。然而,当前对延迟的研究主要集中在供应链管理上,且往往是基于事先已经固定好的产品架构,对产品设计与延迟实现之间的内在交互影响关系关注较少。本文强调这种内在交互影响关系,并基于开放制造环境提出了一种动态延迟设计问题,它主要指将产品架构与延迟实现进行关联协同设计。而在该过程中必然会涉及多主体间的关联决策与优化问题,这种关联决策往往具有主从结构关系。因此,本文提炼出开放制造环境下动态延迟设计的主从关联优化问题,具体如下:(1)研究产品族设计与延迟实现之间的主从关联优化问题,并同时考虑到供应商决策的影响,相应地,提出产品族设计与供应链延迟决策的主从关联优化问题。通过构建在产品族设计与供应链延迟决策间的主从交互决策框架,提出了以制造商为主、多个供应商和延迟承包商为从的双层优化模型。采用分析法和遗传算法求解,应用于电动汽车产品族延迟案例,并对模型参数进行了灵敏度分析。(2)进一步考虑到每个延迟产品模块对延迟类型及相应承包商可能会有不同的最优选择,提出产品族设计与延迟类型及方案选择的主从关联优化问题。通过分析它们之间内在的交互决策框架,构建了以产品族设计为主、延迟类型及方案选择为从的0-1整数非线性双层规划模型。采用嵌套遗传算法来求解,并应用于智能冰箱产品族案例,最后将结果与传统两阶段法和集成法的结果进行了比较。(3)进一步考虑到延迟承包商的分包或关联配套行为,提出产品族架构与延迟承包及子承包决策的三层主从关联优化问题。构建了以制造商为主者、多个承包商为从者及多个子承包商为子从者的三层动态博弈框架,基于Stackelberg博弈理论建立了三层博弈理论模型。结合分析法和嵌套遗传算法求解,应用于电动汽车产品族延迟案例,并对提出的延迟成本和需求两个参数进行了灵敏度分析。(4)考虑到延迟实现中的不同延迟方式和生产技术等对环境产生的影响不同,提出绿色产品族设计与低碳延迟实现的主从关联优化问题。构建了以绿色产品族设计为上层、低碳延迟实现为下层的主从交互机制,提出0-1整数非线性的双层优化模型。采用嵌套遗传算法求解,应用于绿色电动汽车延迟案例,并将结果分别与两阶段法、集成法以及不考虑环保的主从关联优化法的结果进行了比较。更多还原

【Abstract】 Open manufacturing aims to produce products in an open,collaborative and distributed manner based on open source principles.It promotes collaboration across different firms and supports the sharing and exchange of knowledge and services throughout the value chain.The postponement strategy aims to postpone the differentiation of products to the end of supply chain until the customization-specific demand is received.Open manufacturing environment enables more postponement activities to be selected and matched in the product design stage.In practice,successful implementation of postponement strategy in open manufacturing must incorporate its concept as early as possible in the product design stage.However,existing studies on postponement have mainly focused on supply chain management,with limited attention to the inherent coupling of product design and postponement fulfillments in an open manufacturing environment,which entails the necessity of joint decision making for product architecture design and postponement fulfillments of open manufacturing activities.This dissertation emphasizes such a joint decision-making process and formulates the Dynamic Postponement Design(DPD)problems that consider an undefined product architecture that interacts with the postponement fulfillments in an open manufacturing environment.It will inevitably involve multi-agent decision-making and optimization problems in DPD process,which often have a leader-follower structure.Thus,this dissertation formulates the following four Hierarchical Joint Optimization(HJO)problems for DPD in open manufacturing:(1)We firstly research the HJO problem of product family design and postponement fulfillment,and the influences of related suppliers’ decision-makings on the optimization problem are considered at the same time.Correspondingly,the HJO problem of product family design and supply chain postponement decision is proposed.A hierarchical interactive decision-making framework between product family design and supply chain postponement decision is constructed.An HJO model with a single manufacturer as the leader and multiple suppliers and contractors as the followers is proposed.Analytical solutions are developed incorporating a genetic algorithm for the proposed HJO model.A case study of electric vehicle postponement is performed to illustrate the effectiveness of the proposed methodology.(2)As the most appropriate postponement type and postponement scheme(contractor)for each postponed product module may be different,the HJO problem of product family design and the selections of postponement type and scheme is proposed.To tackle the HJO problem based the DPD mode,an HJO model based on Stackelberg game theory is developed.The HJO model deploys a bilevel mixed 0-1 nonlinear programming decision structure to reveal the inherent coupling of product design and the selections of postponement type and scheme.To solve the HJO model,a nested genetic algorithm is developed.A case study of smart refrigerator product family design and postponement is reported to demonstrate the feasibility and potential of the proposed HJO approach,and two experiments are designed to compare the obtained results with those results derived from the traditional all-in-one and two-stage methods.(3)Based on the subcontracting decisions or associated supporting behavior of the postponement contractors,we formulate the three-level HJO problem of postponement contracting with subcontracting for product family architecting.A three-level dynamic non-cooperative game framework is constructed with a single manufacturer as the leader,multiple postponement contractors as the followers,and multiple subcontractors as the sub-followers.The interaction among these different stakeholders is modeled as a non-linear,mixed-integer,three-level game-theoretic model based on the Stackelberg game theory.Analytical solutions are developed incorporating a nested genetic algorithm.A practical case study of postponement contracting decisions in an electric vehicle company is reported to verify the feasibility and potential of the proposed HJO approach.(4)Considering that the different postponement fulfillment modes and production technologies have different impacts on the environment,we propose the HJO problem of green product family design and low-carbon postponement fulfillment.A dynamic interactive decision framework with green product family design as the upper level optimization and low-carbon postponement fulfillment as the lower level optimization is constructed.A mixed-integer nonlinear HJO model is developed to characterize the leader-follower decision making process.A nested genetic algorithm is developed and implemented to solve the HJO model.A case study of electric vehicles demonstrates the feasibility and potential of the proposed optimization problem by HJO approach.更多还原