【作者】 李喆；

【导师】 杜纲；

【作者基本信息】 天津大学 ， 工业工程， 2007， 博士

【摘要】 产品族设计是大规模定制中的关键问题。以往的研究,对于产品族主要功能参数及其关联的从属功能参数的设计多采取各自独立的方式而忽略其间的主从关联关系,缺乏对各功能参数的协同设计的研究,特别是具有模糊参数的情形。模糊多层次优化问题由于其广泛的实际应用背景而成为近年来研究的热点,但由于其求解的难度而主要停留在理论层面。本文针对这一现实背景,对面向产品族协同优化设计的模糊双层规划模型与方法进行了系统的研究,主要工作如下:第一,基于对目前产品族参数设计的研究综述,分析并指出了研究中的不足;提出了基于主从关联结构的产品族协同优化设计的理念,将具有主从结构的产品族协同优化设计问题分为两类:全局-局部协同优化和关键-从属协同优化。第二,对于全局-局部协同优化和关键-从属协同优化两类问题,研究了其概念模型和数学模型建立的全过程,建立了包括市场需求分析、产品族架构、功能指标确定、柔性与模糊性分析等的分析模型和数学模型,探讨了大规模定制下的定制参数、平台参数、产品平台、产品族等的规划设计方法。第三,研究了模型的解析型算法,分别针对具有模糊目标与模糊约束的模糊双层规划以及具有模糊系数的模糊双层规划模型构建了求解方法,并给出产品族设计的数值算例验证了其可行性。第四,研究了模型的智能型算法,分别建立了基于遗传算法和粒子群算法的模糊双层规划求解方法,给出了数值算例。第五,将所研究的模型和方法应用于具有模糊参数的机械产品族优化设计,建立了产品族协同优化设计的模糊双层规划模型。最后对全文工作做了总结,并对今后的进一步研究做了展望。更多还原

【Abstract】 Product family design is the critical problem of mass customization. In previous literatures, methods of product family design for main function parameters and correlative subordinate function parameters are adopted respectively, so that the leader-follower relationship between them and the research on cooperative design with function parameters are neglected, especially with fuzzy parameters. It is a hotspot that how to solve the fuzzy multilevel optimization problems, because of its comprehensive realistic application, however, it has been focus on the academic meaning because of the difficulty to seek solution. Aiming at the realistic background, systemic research on fuzzy bilevel optimization models and methods faced to hierarchy optimization design of product family are made. The main work of this paper is as the followings:Firstly, the deficiencies in the study based on summarizing the current research on product family design are pointed out, and the cooperative optimization design problems of product family with principal-subordinate structure are divided into two species: the global-local cooperative optimization and the pivotal-subordinative cooperative optimization.Secondly, aiming at two problems of the global-local and the pivotal- subordinative cooperative optimization, the process of up building their notional models and mathematic models are researched, analysis models and mathematic models are set up, including analyzing market demand, constructing product family, ascertaining function index, analyzing flexibility and fuzziness and so on. Moreover, the programming design methods about customization parameters, flat parameters, product flat, product family in mass customization are discussed.Thirdly, analytical solution to the models are researched and the method of resolving problems are constructed, aiming at fuzzy bilevel optimization models with fuzzy objects and fuzzy bilevel programming and fuzzy bilevel optimization models with fuzzy parameters, respectively. Numeral example of product family design is put forward to prove it feasible.Fourthly, intelligent algorithms of the models are researched, the solution to fuzzy bi-level programming based on genetic algorithm and Particle Swarm Optimization are constructed, respectively, and then numeral example are put forward.Fifthly, the models and methods studied in this paper are applied to mechanical product family optimization design with fuzzy parameters, furthermore, fuzzy bi-level optimization models of product family cooperative optimization design are set up.Finally, the total work of the paper is summarized and the vista of further research is made.更多还原