基于服务质量的若干排队系统的博弈分析

发布时间：2018-04-26 12:20

本文选题：服务质量 + 博弈分析　；参考：《北京交通大学》2017年博士论文

【摘要】：在排队系统中,服务质量是反映顾客满意度的重要指标.服务台的故障及休假均能影响服务质量.基于服务质量,本文从经济学角度考虑了可修排队、休假排队及医疗服务系统等若干排队系统的博弈问题.首先,本文将可修排队系统与重试排队系统相结合,分析了同质顾客与异质顾客的策略行为.其次,在不同信息水平下,研究了带有工作休假的离散排队系统和带门策略的双向排队系统中顾客的策略行为.最后,本文研究了医疗服务系统中的博弈分析及最优预算分配策略.本文共分为五章.第一章简述了排队博弈分析的研究背景及现状,并给出了本文所需的基本概念及理论知识.第二章分析了可修重试排队系统中的博弈问题.2.1节讨论了带单删除机制的常数重试排队系统中顾客的策略行为,得到了可见与不可见情形下顾客的均衡策略及社会最优策略.直观地认为,服务台故障会降低顾客的服务质量,导致进入系统的顾客数减少.然而,数值实例显示,不同信息水平下顾客对单位时间内发生故障次数的偏好不一样.2.2节用可修重试排队系统刻画了感知错误下的认知用户在频谱共享中的策略行为,服务台故障看作是授权用户的到达.其中,认知用户发生感知错误会降低系统的服务质量.本节重点分析了感知错误对认知用户策略行为的影响,通过征收入场费的方式使得认知用户按照社会最优策略采取行动,从而充分利用有限的认知无线电资源.2.3节给出了可修的经典重试排队系统中异质顾客的博弈分析,数值实例表明,单位时间内顾客的等待花费不同时,服务商最大利润与社会收益最大值不再相等,顾客的异质性增加了顾客间的负外部经济效应.第三章研究了离散时间与连续时间休假排队系统中的博弈问题.3.1节得到了单重工作休假策略下离散排队系统中的顾客均衡策略,分析了不同信息水平对顾客策略的影响.另外通过数值实例分析了休假时间对顾客均衡行为的影响.3.2节利用门策略控制下的双向排队系统讨论了'乘客-出租车'中的博弈问题,该门策略表明当出租车的数量达到最大阈值时停止进入系统,低于一定阈值时重新开始进入系统.在此门策略下,得到了三种不同信息水平(完全不可见、几乎不可见和完全可见情形)下到达乘客的均衡进队策略与社会最优进队策略.数值实例表明,在完全可见情形下,顾客的均衡阈值策略不仅依赖于门策略,也依赖于乘客与出租车到达率的大小关系.当出租车的停车数量比较多时,带有门策略的双向排队系统能够在一定条件下提高系统的社会收益.第四章基于博弈理论,通过权衡系统的服务速率与服务质量研究了医疗服务系统中的预算分配策略,分别得到了两层和三层医疗服务系统中病人的均衡策略及唯一最优分配策略.而且对于旨在通过预算分配策略提高服务速率与服务质量的不同层次的医疗服务系统,上述结果具有鲁棒性.从社会管理者(政府)角度出发,其分析结果为公共医疗服务系统中预算分配问题提供了合理的建议.第五章总结了本论文的成果与创新点,并给出了未来的研究方向.
[Abstract]:In the queuing system, the quality of service is an important indicator of customer satisfaction. The failure and vacation of the service table can affect the quality of service. Based on the quality of service, this paper considers the game problems of a number of queuing systems, such as repairable queues, vacation queues and medical service systems from the economic point of view. The trial queuing system is combined to analyze the strategic behavior of homogeneity customers and heterogeneous customers. Secondly, under different information levels, this paper studies the customer strategy behavior in the discrete queuing system with work vacation and the two-way queuing system with gate strategy. Finally, this paper studies the game analysis and optimal budget allocation in the medical service system. This paper is divided into five chapters. In the first chapter, the research background and current situation of queuing game analysis are briefly described, and the basic concepts and theoretical knowledge are given in this paper. The second chapter analyzes the game problem.2.1 in retrial retrial queuing system, and discusses the customer strategy behavior in a constant retrial queuing system with single deletion mechanism. The customer equilibrium strategy and social optimal strategy under visible and invisible conditions are considered. It is intuitively believed that the service table failure can reduce the customer's service quality and lead to the reduction of the number of customers entering the system. However, the numerical example shows that the preference of the customer to the number of failures per unit time under different information levels is not the same as.2.2 repairable. The trial queuing system depicts the strategy behavior of the cognitive users in the spectrum sharing under the perceived error. The service desk failure is regarded as the arrival of the authorized users. In this section, the cognitive users' perception errors will reduce the system's service quality. This section focuses on the analysis of the effect of perception errors on the cognitive user strategy behavior and through the acquisition of admission fees. In order to make the cognitive users act according to the social optimal strategy, the game analysis of the heterogeneous customers in the repairable classic retrial queuing system is given full use of the limited cognitive radio resource.2.3 section. The numerical example shows that the maximum profit and the maximum social income value of the service provider are different when the customer's waiting cost is different within the unit time. The customer heterogeneity increases the negative external economic effect between customers. In the third chapter, the game problem.3.1 in the discrete time and continuous time vacation queuing system is studied. The customer equilibrium strategy in the discrete queuing system under the single work vacation strategy is obtained, and the influence of the different information level on the customer strategy is analyzed. A numerical example is given to analyze the effect of vacation time on customer equilibrium behavior..3.2 section uses a two-way queuing system controlled by gate strategy to discuss the game problem in the "passenger taxi". The strategy shows that when the number of taxis reaches the maximum threshold, it stops entering the system and begins to enter the system when the threshold is below a certain threshold. Under the strategy, three different information levels (completely invisible, almost invisible and fully visible) are obtained to reach the balanced queue strategy and the social optimal team strategy. The numerical example shows that the customer's equilibrium threshold strategy depends not only on the gate strategy but also on the passenger and taxi arrival rates in the completely visible case. When the number of taxi parking is much more, the two-way queuing system with gate strategy can improve the social income of the system under certain conditions. The fourth chapter studies the budget allocation strategy in the medical service system by weighing the service rate and quality of service based on game theory, and obtains two and three layers respectively. The balance strategy and the only optimal allocation strategy in the medical service system. And the results are robust to the different levels of medical service system aimed at improving service rate and quality of service through the budget allocation strategy. From the perspective of the social Manager (government), the results of the analysis are in the public health service system. The fifth chapter summarizes the achievements and innovations of this paper, and gives directions for future research.

【学位授予单位】：北京交通大学
【学位级别】：博士
【学位授予年份】：2017
【分类号】：F224.32

【相似文献】