Optimal Research On The Online Customer Service System Based On Queuing Theory

With the development of the human society, the services play an increasingly important role in socio-economic status, and replace manufacturing as the largest sector of the economy. Therefore, to a large extent, how to improve the service quality and the service efficiency related to corporate survival. An increasing number of people choose the Internet to obtain the necessary information. So the customer service approach entered into the network area. Nowadays, customers have been the competitive focus in markets.This paper is based on the data mining, and the queuing theory with detail item. Using these theories can realize and analysis online customer service system. Therefore, using mathematic model to describe the customer service system can result in the best parameter design and undoubtedly has economic value and practical significance. The following two aspects included:Firstly, the paper users LRFM model to analyze customer’s behavior. The LRFM model was modified from traditional RFM model by introducing L as the relationship length between customers and company as well as loyalty of customers’behavior. Then the paper uses two-stage analysis method which combines K-means cluster method and decision tree method to identify customers by using personal data. Then, the scheme contains four customer clusters:VIP customers, important customers, general customers and temporary customers. It uses decision tree method to analyze the cluster results and the attribute differences among them. The paper also brings out some propositional companies’ marketing suggestions and strategies.Secondly, in this paper, the queuing theory is used for the system, and the M/M/c/∞queuing model is established, educe the stationary distributing and the primary target parameter of the online customer service system. The paper also makes use of the software to compose calculation procedure, has gained the corresponding target parameter. The wishes model is used for system optimization. It obtained optimal service seats number under the rate of losing Pn<0.02and advanced the efficiency of thread. This scheme provides a theoretical basis for the system upgrade and maintenance.