The Research On Stochastic Service System With Finite Buffers

In real life,a lot of production systems and service systems are the stochastic service system.In general,the customer's waiting time is limited in the buffer,and we can control the waiting time by setting the appropriate buffer size.It is necessary to study the stochastic service system with finite buffer size.At present,the research on stochastic service systems with finite buffer size focuses on the analysis of various performance measures of the system.In order to achieve better resource allocation and reduce the system operation cost,this paper analyzes two stochastic service systems.The corresponding optimized model is established.The corresponding optimal control strategy is obtained with an objective to minimize system operation cost.The first stochastic service system with finite buffer size has two tandem servers.Customers arrive to the system according to a Poisson process and enter the fixed server.Service time is deterministic in the model.The second stochastic service system with finite buffer size has multiple servers.Customers arrive to the system according to batch Markov arrival process and enter any idle server.Service time follows phase type distribution.For the tandem stochastic service system with finite buffer and a Poisson arrival process,the stochastic breakdowns and system maintenance are considered.Various system operation costs are analyzed,and a model is established to minimize the system cost function.The optimal buffer size is derived.For the stochastic service system with finite buffer size and multiserver,customers arrive to the system according to a batch Markov arrival process.Firstly,the steady-state probability distribution and the performance measures of the system are derived by analyzing the transfer of system states.A model is established by combing the energy consumption and system performance measures.Finally,the optimal number of servers is derived to minimize the system operation cost.