Researches On Simulation And Optimization Of Queuing Performance Based On Monte Carlo Method

Currently, queuing problems is a common phenomenon, such as the queue problem in banks and hospitals. With the rapid development of computer in recent years, the congestion of network is also becoming one of an important queue problem. Be a subjiet based on probability theory and stochastic processes, queue theory play an important role in addressing the issues of queue problems. Application of queueing theory can be a good solution to the problem of queue in daily life, so it plays an important role in many areas. Such as network traffic control, transportation, resource scheduling and so on, which the application of queueing theory to solve bank queuing performance is a hot topic.In this paper, using Monte Carlo method to simulate the queuing system in bank based on in-depth study of queuing theory, then optimize the queuing systems in bank including performance optimization and parameters optimization. Based on Monte Carlo method, comparing the average response time of from different queue model to the response time of the web server, we can find out the queue model which fit the web server well. This paper mainly includes the following three parts:(1)Simulate the queuing system in bank, put forward the Monte Carlo method to solve the queuing problems in the bank. Firstly, get the performance metrics of simplest queuing model by the computer simulation. The experimental results show that our method is feasible and high efficiency.(2)Optimize the performance of the queuing system in bank based on the result of advanced queuing systems. Use Monte Carlo method to obtain the performance metrics of complex queuing model by adjusting the distribution of the arrival time and service time (These models is difficult to solve or couldn’t solve by the method of theoretical analysis). Get the optimal performance queue model by analyzing these complex queuing models.(3)Optimize the parameters of the queuing systems in bank. As we know that the number of service windows in bank is an integer, so we use marginal analysis methods analysis of the optimal number of service windows in the bank, minimize the number of service windows in bank to reduce costs without reducing the efficiency of bank service. Based on the optimal number of service windows, we study the optimal service rate of the service windows, making the bank queue have the optimal performance.(4)We employ Monte Carlo method to simulate the queuing behaviors. We obtain the average response time by experiments. The parameters such as service time is estimated by maximizing the log-likelihood function of the measured average response time. Compare the average response time of from different queue model to the response time of the web server, we can find out the queue model which fit the web server well.Presently, queuing theory is widely used in queuing problem in bank and web server. The research results of this paper based on simulatiom and optimization of bank queueing system and web server have a certain theoretical innovation and application value, have a certain guide meaning for bank queuing system scheduling policy and play an important role in capacity planning and overload control for web server.