共享模式服务性能建模与分析

发布时间：2019-04-22 09:40

【摘要】：近些年,产能过剩的资源分配不均及信息技术的高速发展,促进了共享模式在社会各行业的兴起与发展,如Uber的车共享,Airbnb的空间共享。如果我们把有限资源无限需求的现象看成一个服务系统,实际生活中存在着各种各样的服务系统。共享模式的服务系统具有这样的特点,用户在接受系统服务之后会以一定概率共享出服务资源,从而使得服务系统内部的服务资源在逐步地增加。共享模式的服务性能分析不同于传统非共享模式的服务性能分析。在对非共享模式服务系统进行性能分析时,我们常常用排队理论对服务系统进行数学建模并抽象出一系列数学特征,通过分析得到系统在达到稳定时的状态转移关系,并依此得到各项性能指标。而共享模式下服务系统的资源会随着顾客的共享行为而逐步的增加,此类情况下,用排队论进行求解时将无法得到稳定的系统状态,从而无法完成共享模式服务性能的求解。现实生活中,存在许多具有共享行为的服务系统,此类系统在提供服务的过程中,服务资源会随着时间动态变化,如何评价此类系统的服务性能就是本文要解决的问题。为解决上述问题,本文先做了相关的基础研究。接着本文在排队理论的基础上建立共享模式下的排队模型,并提出求解此模型服务性能的解决方案。接下来本文详细地探讨整个仿真分析系统的设计与实现过程,根据系统分层与功能模块对系统的设计进行了深入的解释,包括架构设计及实现过程,最后利用仿真分析系统进行仿真分析,通过对大量数据进行分析与可视化展示,得到共享模式服务系统与非共享模式服务系统的各自特点,设计并实现能够较准确地测定此系统的服务性能指标方案。并在此指标方案下完成对共享模式服务系统的系统指标的影响因素测定与分析。本文不仅完整地解决了共享模式下服务系统的服务性能评价问题,还通过仿真分析系统对此类模型做了深入充分地分析与研究。这也是本课的研究价值所在。
[Abstract]:In recent years, the uneven distribution of resources due to overcapacity and the rapid development of information technology have promoted the rise and development of sharing mode in various sectors of society, such as the sharing of cars in Uber and the sharing of space in Airbnb. If we regard the phenomenon of infinite demand for limited resources as a service system, there are all kinds of service systems in real life. The shared-mode service system has such characteristics that users will share the service resources with a certain probability after receiving the system services, so that the service resources within the service system are gradually increasing. The service performance analysis of shared mode is different from that of traditional non-shared mode. When analyzing the performance of non-shared mode service system, we often use queuing theory to model the service system and abstract a series of mathematical characteristics, and get the state transition relationship when the system reaches stability. According to this, the performance indexes are obtained. The resources of the service system in the shared mode will gradually increase with the sharing behavior of the customers. In this case, it will not be possible to obtain the stable state of the system when solving the problem by queuing theory, so the service performance of the shared mode will not be solved. In real life, there are many service systems with shared behavior. In the process of providing services, the service resources will change dynamically with time. How to evaluate the service performance of such systems is the problem to be solved in this paper. In order to solve the above problems, this paper first does some basic research. Then, based on the queuing theory, the queueing model in shared mode is established, and a solution to solve the service performance of this model is proposed. Next, this paper discusses the design and implementation of the whole simulation analysis system in detail, and explains the design of the system according to the hierarchical and functional modules of the system, including architecture design and implementation process. Finally, the simulation analysis system is used to simulate and analyze, through the analysis and visualization of a large number of data, the characteristics of shared mode service system and non-shared mode service system are obtained. Design and implement a scheme which can accurately measure the service performance index of the system. Under this index scheme, the influencing factors of the system index of shared mode service system are measured and analyzed. This paper not only solves the problem of service performance evaluation in the shared mode, but also makes a thorough and full analysis and research on this kind of model through the simulation and analysis system. This is also the research value of this lesson.
【学位授予单位】：北京邮电大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：O226;TP391.9

【相似文献】