节点文献

基于随机需求的配送中心设备配置与库存控制问题研究

Research on the Equipment Configuration and the Inventory Control in the Distribution Center Based on the Random Demand

【作者】 付延冰

【导师】 陈治亚;

【作者基本信息】 中南大学 , 交通运输规划与管理, 2010, 博士

【摘要】 现代物流业是现代经济的重要组成部分,是我国工业化进程中最为经济合理的综合服务模式。作为现代物流系统的重要环节,配送中心受到广泛关注,许多企业组建了自己的配送中心,第三方物流配送中心纷纷成立,铁路等传统运输方式也在积极进行现代物流配送中心建设的相关研究和探索。配送中心连接着供货点和需求点,是两者之间的桥梁,优化配送中心设备配置,控制库存费用,降低运营成本,提高服务水平,对于整个物流系统都具有非常重要的作用。本文以对配送中心日常作业流程的细致分析为基础,以有效配置配送中心基本作业设备、提高作业效率和降低营运成本为目标,在总结国内外相关研究的基础上,充分考虑配送中心日常作业中存在诸多不确定因素的特点,运用随机过程理论和计算机仿真技术,选择性研究配送中心的两个基础问题,在研究二级库存协同控制问题时考虑了设备配置对配送中心服务过程的影响。本文具体研究内容包括以下几个方面:研究了采取整车直接运输策略时配送中心基本配送作业设备的配置问题。对于先到先服务作业顺序下的设备配置问题,运用串联排队的理论计算先到先服务规则下配送中心交货周期的平均长度,以对客户配送服务的准时率等为约束条件,以成本最低为优化目标建立了基本配送作业设备配置的模型;对于有优先权服务作业顺序下的设备配置问题,根据配送中心各配送服务相关部门作业的特点,对客户的优先服务方式进行了分配,以此为基础计算了配送中心对不同等级客户交货周期的平均长度,建立了基于排队论的基本配送作业设备配置模型。对于两种作业顺序下的设备配置问题均采用枚举法求解最优值。在Matlab语言环境下根据系统运行情况直接开发了两个仿真程序分别用于模拟先到先服务和有优先权服务两种作业顺序下配送中心基本配送服务作业过程,建立了采取整车直接运输策略时配送中心基本配送作业设备配置的仿真模型,并进行优化求解。研究了采取零担运输策略时配送中心基本配送作业设备的配置问题。对于即时发车条件下的设备配置问题,分析了配送中心货物拣选任务的完成情况和配送中心内部待运货物数量的变化情况,推导出配送中心作业平稳的充分和必要条件,计算了即时发车条件下配送中心交货周期的平均长度,以成本最低为优化目标,建立了基于排队论的基本配送服务作业设备配置模型;对于不满载发车条件和满载发车条件下的设备配置问题,运用排队论中批量服务的理论分别计算了配送中心交货周期的平均长度,建立了基本配送作业设备配置的模型。对于三种发车条件下的设备配置问题均采用枚举法求解最优值。在Matlab语言环境下根据系统运行情况直接开发了三个仿真程序以分别模拟三种发车条件下配送中心基本配送服务作业过程,建立了采取零担运输策略时配送中心基本配送作业设备配置的仿真模型,并进行优化求解。研究了需求连续的点库存控制问题。以库存总成本最小为优化目标,构建了不考虑服务水平约束的连续型点库存控制模型。在求解过程中,分别针对均匀分布、指数分布和正态分布等三种常见的提前期需求情况,计算了每个订货周期内的平均缺货量,以安全因子作为中间变量,推导出不同需求条件下最优订货量和最优订货点等指标的计算公式;以成本最小为优化目标,将对客户的服务水平作为约束条件之一,构建了考虑服务水平约束的连续型点库存控制模型,将需求量服务水平作为固定数值对模型进行求解,以安全因子作为中间变量,推导出均匀分布、指数分布和正态分布等三种需求条件下最优订货量和最优订货点等指标的计算公式。研究了考虑配送中心配送服务作业过程的二级库存协同控制问题。本文所研究的二级库存协同控制系统由一个配送中心和若干个零售商组成,配送中心对零售商需求的反应时间不仅受配送中心库存策略的影响还受到配送中心配送作业能力和配送需求到达密集程度的影响。通过对系统作业情况的细致分析,以成本最小为优化目标构建了二级库存协同控制的仿真模型,开发了仿真程序,并运用改进的遗传算法对问题进行优化求解。

【Abstract】 Modern logistics is the important component of the modern economics, and the most economical and reasonable integrated service mode. As the important link in the logistics system, distribution center attracted a wide spread, and a lot of enterprise set up their own distribution center, and many third party logistics distribution centers are established, and the traditional transportation ways like as the railway are engaged in the research of building the modern distribution center. Distribution center connect the suppliers and the consumers, which is the bridge. It’s very useful for the whole logistics system to optimize the equipment configuration, to control the inventory cost, to decrease the operating cost and to improve the service level of the distribution center.Based on the careful analysis of the routine operating flow of the distribution center and the relative research at home and abroad, and in order to effectively allocate the basic equipment in the distribution center, to improve operation efficiency, and to decrease operation cost, and considering the character that there are a lot of uncertain factors in the routine work of the distribution center, this paper selectively researches several basic problems in the process of construct and the routine work in the distribution center by the stochastic process theory and computer simulation. The research contents include as following:Firstly, this paper researches the problem how to optimize the equipment configuration based on direct shipping strategy. For the problem of the equipment configuration under the order of first come first serve, this paper calculates the average time length of delivery cycle of the distribution center by the tandem queuing, and constructs the model to optimize the equipment configuration based on the queuing theory with the goal of minimizing the total cost, and with the restrictive condition to delivery the freight on time. For the problem of the equipment configuration under the order of priority service, this paper sets the service ways for different customs according to the operating features of different sectors of the distribution center, and calculates the average time length of delivery cycle of the distribution center, and constructs the model to optimize the equipment configuration based on the queuing theory with the goal of minimizing the total cost. For the equipment configuration under the two orders of serve, this paper designs the search steps based on the enumeration method to find the optimal value. This paper exploits two simulation programs to simulate the delivery process of the distribution center under the order of first come first serve and the order of priority service according to their actual operation situation by matlab language, and constructs their simulation models and then optimizes their solutions.Secondly, this paper researches the problem how to optimize the equipment configuration based on package transportation strategy. For the problem of the equipment configuration under the condition of instant departure, this paper analyzes the situation to pick goods and the change of the quantity of the goods waiting for delivery, and deduces the necessary and sufficient condition of operation stationary of the distribution centre, and calculates the average time length of delivery cycle of the distribution center, and constructs the model to optimize the equipment configuration based on the queuing theory with the goal of minimizing the total cost. For the problems of the equipment configuration under the condition of the no full load departure and the full load departure, this paper calculates the average time length of delivery cycle of the distribution center, and constructs the model to optimize the equipment configuration based on the queuing theory. For the equipment configuration under the three conditions of departure, this paper designs the search steps based on the enumeration method to find the optimal value. This paper exploits three simulation programs to simulate the delivery process of the distribution center under these condition of departure according to their actual operation situation by matlab language, and constructs their simulation models and optimizes their solutions.Thirdly, this paper researches the spots inventory control problem with the continuous demand. This paper constructs the continuous inventory control model without the service level constraint with the goal of minimizing the total inventory cost. In the process of solving the problems, according to three common demand types in the lead time such as the uniform distribution, the exponential distribution and the normal distribution, we calculate the average backorder quantity in one order cycle, and with the intermediate variable of safety factor, we deduce the calculation formulas of these indexes such as optimal order point and optimal order quantity. This paper constructs the continuous inventory control model with the service level constraint with the goal of minimizing the total inventory cost. We look the demand quantity service level as the known parameter, and with the intermediate variable of safety factor, we deduce the calculation formulas of these indexes such as optimal order point and optimal order quantity under the demand condition such as the uniform distribution, the exponential distribution and the normal distribution.Finally, this paper researches the two echelon inventor control problem considering the delivery process. The two echelon inventor control system is composed by one distribution center and some retailers, and the response time is affected not only by the inventory control strategy of the distribution center but also by the performance ability of the distribution center and the arrival intensive degree of the distribution demand. Based on the careful analysis of the operating situation of the system, constructs the inventory control simulation model with the service level constraint and with the goal of minimizing the total cost, and exploits the simulation programs and optimizes their solutions with the genetic algorithm.

  • 【网络出版投稿人】 中南大学
  • 【网络出版年期】2010年 11期
  • 【分类号】F224;F253.4
  • 【被引频次】12
  • 【下载频次】1546
  • 攻读期成果
节点文献中: 

本文链接的文献网络图示:

本文的引文网络