【作者】 陈国华；

【导师】 张根保；

【作者基本信息】 重庆大学 ， 机械工程， 2011， 博士

【摘要】 可靠性是供应链的一个基本属性,也是供应链运作的基础,可靠性是伴随不确定性问题普遍存在的客观现象。汽车供应链是一个复杂的系统过程,它具有多参与主体、跨地域和多环节等特征,使汽车供应链更容易受到来自外部环境和链上各实体内部不确定因素的影响,造成供应链出现各种问题。在这些不确定因素作用下,可靠性已成为能否确保供应链正常运行的关键要素。为简化问题及研究的方便,本文以整车制造商为核心的三级汽车制造供应链为研究对象,以可靠性工程、系统工程、供应链管理、工业工程、先进制造技术与计算机技术等多种交叉学科理论与技术为基础,以汽车供应链结构可靠性分析、汽车供应链交付可靠性研究、基于故障树分析法的汽车供应链关键因素失效诊断、汽车供应链可靠性失效模式诊断技术、汽车供应链可靠性分配技术研究等为主要内容,对汽车供应链可靠性的若干关键技术进行了深入研究。具体研究内容如下:①汽车供应链的结构可靠性分析。结构可靠性分析是一种基础性分析,是供应链可靠性研究的基础,在对供应链可靠性的定义重新进行界定、对供应链可靠性进行分类及提出供应链可靠性的若干评价指标后,深入分析了汽车供应链的组成单元,并对单元的构成方式及可靠性运用马氏理论进行了分析与改进研究,由此确定了汽车供应链结构可靠性的计算方法。重点研究了仓库的不同运作方式对汽车供应链结构可靠性的影响,运用定量分析的方法分析了设置仓库与不设置仓库汽车供应链系统可靠性的变化情况、仓库设置在汽车供应链系统上游与下游对系统可靠性的影响情况以及仓库上下游子系统对整个供应链系统的影响情况等。②为研究三种模式的汽车供应链对交付可靠性影响的系统行为,将系统动力学方法引入到汽车供应链交付可靠性行为研究中。在对交付可靠性进行精确定义后,根据三种模式的汽车供应链运作特点,对它们分别建立了系统动力学模型,然后以交付可靠性和库存变化的系统行为作为研究对象,对它们在相同参数、市场扰动以及运行策略改变等情况下进行了系统行为的仿真对比分析,得出了不同运作机制下汽车供应链在交付可靠性方面的内在运作规律。③针对汽车供应链可靠性关键因素诊断问题,在分析了汽车供应链失效的原因后,借鉴机器零部件失效诊断分析的方法,运用故障树分析的原理,将供应链诊断这一抽象问题予以具体化处理,从可靠性角度研究了具有独立失效及具有共因失效的汽车供应链关键因素的诊断技术。对于具有独立失效的汽车供应链关键因素诊断技术的研究,主要采用了蒙特卡罗仿真的方法,在构建了汽车供应链失效诊断的一般模型后,建立了仿真算法来求相关可靠性指标和诊断关键因素;对于具有共因失效的汽车供应链关键因素诊断技术的研究,详细分析了供应商与供应商之间的共因失效(Supplier-Supplier:SS共因失效)、供应商与制造商之间的共因失效(Supplier-Manufacturer:SM共因失效)、制造商与分销商之间的共因失效(Manufacturer-Distributor:MD共因失效)、分销商与分销商之间的共因失效(Distributor-Distributor:DD共因失效)以及供应商、制造商、分销商之间的共因失效(Supplier-Manufacturer-Distributor:SMD共因失效)等多种情况的关键因素诊断技术,对它们分别建立了各自的失效诊断模型,并研究了各自的算法及应用。④针对汽车供应链可靠性失效模式诊断,引入了一种新的方法-GO法来解决汽车供应链的精确计算以及汽车供应链失效模式诊断等问题。通过对导致汽车供应链系统成功的各类重要因素进行详细分析的基础上,建立了汽车供应链可靠性分析的GO模型,该模型具有通用性,能对汽车供应链可靠性进行精确计算,并能对汽车供应链系统进行失效模式的诊断及其重要度的计算,为汽车供应链系统改善与优化提供了思路。⑤针对汽车供应链可靠性分配方法问题,借鉴机器零部件的可靠性分配方法,根据可靠性工程中可靠性成本函数的规律,构建了汽车供应链可靠性预估成本函数,并结合供应链的结构特征以及前面的研究结论,分别对汽车供应链可靠性预估成本函数中的各参数值给出了具体的确定方法,然后建立了汽车供应链可靠性的分配模型,并根据汽车供应链可靠性与成本的微分性质,构建了以重要度最大单元为搜索方向逐步迭代的算法,该算法相对以前研究文献中的算法更简单、更易用。更多还原

【Abstract】 Reliability is not only a fundamental property of supply chain, but also the basis of supply chain’s operations and it is accompanied by the objective phenomenon of widespread uncertainty. Automotive supply chain is a more complex system composed by many players, with the cross-regional and multi-link features. These features will make automotive supply chain more vulnerable to external environment and the entities within the chain of adverse factors, resulting in supply Chain failure. In these uncertain factors, reliability is an important factor, which can ensure the successful operation of supply chain. To simplify problem studied in this paper and facilitate the research, taking three-tier supply chain centered on the vehicle manufacturer as the subject, using the interdisciplinary theory and technology including reliability engineering, systems engineering, supply chain management, industrial engineering, advanced manufacturing technology and computer technology, and so on, this paper studied some key technologies of reliability of automotive supply chain deeply, which includes the analysis of structure reliability of automotive supply chain, the study on delivery reliability of automotive supply chain, the research on failure diagnosis of key factors in automotive supply chain based on fault tree analysis , the study on diagnostic technologies of failure mode of the automotive supply chain, and the study on reliability allocation technologies of automotive supply chain. Specific contents are as follows:①The analysis of structure reliability of automotive supply chain. The analysis of structure reliability is a basic one, and it is the base of studying reliability of supply chain. After redefining the definition of reliability of supply chain, classifying reliability of supply chain and putting forward some reliability index of supply chain, the component units of the automotive supply chain was analyzed deeply, and the construction mode and reliability of unit was analyzed and improved by the use Markov theory, thus the method of calculating the reliability of automotive supply chain was determined. At the same time, the paper focused on warehouse’s role in promoting systemic reliability of automotive supply chain. It quantitatively analyzed the variation of supply chain system’s reliability in setting warehouse or not, the different influence on supply chain system reliability exerted by setting warehouse upstream or downstream, and the diverse effect on the whole system aroused by upstream supply chain and downstream supply chain. ②To study the system behavior of influence on delivery reliability exerted by supply chains with different modes,a new method—systems dynamics (SD) is introduced into the analysis of supply chain’s behaviors. After precisely defining delivery reliability, three SD models were set up by various automotive supply chains’characteristics,then taking delivery reliability and inventories as the object, the simulation of them was run under the circumstances of the same parameters, disturbances of market demand and operational policies shift, respectively,and we obtained some inherent operational mechanisms of delivery reliability of automotive supply chain with various operational modes.③Study on failure diagnosis of key factors in automotive supply chain. After the reason why automotive supply chain fails was analyzed, borrowing ideas from fault diagnosis method of machinery parts to embody the abstract problem in supply chain, the diagnosis technology of key factors in automotive supply chain including the diagnosis technology of independent failure and the diagnosis technology of common cause failure were studied on with the theory of fault tree analysis. The study on the diagnosis technology of independent failure in automotive supply chain is mainly by the simulation of Monte Karlo Method. After constructing the general model of diagnosing key factors in automotive supply chain,its simulation algorithm was presented to count its interrelated reliability indexes and diagnose crucial factors. Simulation example was presented to prove the method’s feasibility and validity. At the same time, we studied the diagnosis technology of common cause failure in automotive supply chain, which includes the forms of common cause failure between suppliers, between suppliers and manufactures, between manufactures and distributors, between distributors, and between suppliers and manufactures, distributors. Then we established their fault diagnosis model, studied its simulation algorithm and application examples.④Study on diagnosis of reliability failure mode of automotive supply chain. To study the accurate calculations and fault mode diagnosis of supply chain, the new idea that GO methodology is applied in reliability analysis of supply chain system is brought forward. Through analyzing detailedly every important factor leading to success of supply chain system and taking three-tier supply chain as the object,GO model of reliability analysis of supply chain is found, which is universal. It can not only caculate reliability of supply chain system in detail,but also diagnose its fault mode and caculate its importance,which provides the way for improving and optimizing supply chain system. ⑤Study on the technologies and methods of reliability allocation of automotive supply chain. Learning the method used in reliability allocation of machine parts and according to the laws of reliability cost function in reliability engineering, the reliability estimated cost function of automotive supply chain was constructed. At the same time, according to the structure feature of automotive supply chain and the conclusions of previous research,the method of determining parameter values of the reliability estimated cost function of automotive supply chain were given,and the reliability allocation model of automotive supply chain was established. According to the differential nature between reliability and cost in automotive supply chain, the step by step iterative algorithm, taking the factor with the greatest importance as the search direction,was built. Relative to previous algorithm studied in the literature, the algorithm presented in the paper is simpler, easy to use.更多还原