随机生产服务系统最优服务能力分配策略  

Optimal Service Capacity Allocation in a Stochastic Production Service System

作　　者：王康周[1] 江志斌[2] 

机构地区：[1]兰州大学管理学院,兰州730000 [2]上海交通大学工业工程与管理系,上海200030

出　　处：《系统管理学报》2016年第4期767-774,共8页Journal of Systems & Management

基　　金：国家自然科学基金重点项目(70932004);中央高校基本科研业务费专项资金资助项目(15LZUJBWZY121)

摘　　要：针对由生产设施和服务中心组成的柔性生产服务系统,研究服务能力在不同订单之间的分配问题。生产设施生产两类产品,服务中心基于对应的产品为两类顾客提供不同的服务,服务每个顾客订单需要消耗对应的产品。生产和服务时间均服从指数分布,需求过程为泊松过程。因服务能力有限,管理者在任何时刻须决定将服务能力分配给哪一类订单。将系统的最优服务能力分配问题转化为一个连续时间马尔科夫决策过程。通过解析证明和分析,发现该随机系统的最优服务策略为形式非常简洁的指标型优先权服务策略,但又不同于传统服务系统的最优服务策略。最后,分别对最优服务策略与分散决策下的服务策略、相反优先权服务策略以及动态协同下的服务策略进行对比,说明了最优服务能力分配策略的优越性和有效性。The service capacity allocation problem of a production-service system is considered.The system consists of one flexible manufacturing facility and one flexible service center.The manufacturing facility produces two types of items in the first stage,then the service center fulfills two types of orders by the dedicated items in the second stage,each order depletes one unit of corresponding item.Production and service times follow exponential distribution.Customer demands of both types arrive according to independent Poisson processes.The service capacity are limited,thus the manager needs to decide service allocation between two types of items at any time.We formulate the problem as a continuous-time Markov decision process and use this formulation to characterize the structure of the optimal scheduling policies.We show that the optimal service policy is an index type priority policy in concise form,but the policy is different from the optimal policy for the classical service system.Lastly,the policy is compared with several other policies（decentralized policy,contrary priority policy,and coordinated dynamic policy）to show the superiority and effectiveness of the optimal service policy.

关 键 词：生产服务系统 库存服务系统 最优能力分配 马尔科夫决策过程 服务型制造 

分 类 号：F224[经济管理—国民经济] F273