HE Shu-guang1,*, LI Li2, QI Er-shi1
School of Management, Tianjin University, Tianjin 300072, China School of Electronics and Information Engineering, Tianjin Professional College, Tianjin 300402, China email@example.com, firstname.lastname@example.org, Esqi@tju.edu.cn
Based on thestudy of the processes of telecommunication call centers, the service quality metrics of the call centers are put forward. And the mode of the continuous service quality improvement of the call centers based on data warehouse and data mining is studied. Then the process of the IVR (Interactive Voice Response) is analyzed and a mode for the efficiency improvement of IVR is put forward based on theexchange of the orders of the service items in the IVR. Then a service quality metrics of the agents, the ratio of recall in one hour, is put forward. This metrics can be used in the performance analysis of the agents. Furthermore, the model of the performance analysis and control of the ASA (Average Speed of Answer) based on data mining and SPC (Statistical Process Control) is put forward. At last, amethod for forecasting the call arriving in is put forward based the time series analysis using dynamic data mining. The result certified that the efficiency and service quality of the telecommunication call center can be improved obviously using the method in this paper. Keywords: Data mining; Telecommunication call center; Continuous quality improvement; Statistical process control 1INTRODUCTION In recent years, the industry of telecommunication of China improves with a very high speed. Up to September 2006, the total number of users of the mobile telephones has exceeded 400 million. At the same time, the competition in the market is becoming more and more severe. And the customers are becoming more and more mature and rational in selecting the service providers of their mobiletelephones. Because the products of different service providers almost have no difference, service quality is the most important method in absorbing the new customers and preserving the old customers. In this kind of market environment, the call center is becoming more and more important as a communication channel between the service providers and their customers. All the service providers of mobiletelephones are paying more attention in improving the service quality of their call centers. J R Artalejo(2007) has studied the maximum queuing length problem in call centers based a two-dimensional markov chain model, and this is an effective method in solving the optimization problem in unstable environment. Luo Y M(2006) studied the probability distribution of service duration with a case, and itis certified that a lognormal distribution model can be used to describe the service duration. Zhu Y W(2006) has studied the abandon percentage model and it is a useful supplement of the queuing model of call centers. Ward W (1999) has put forward a dynamic staffing model in telephone call center to immediately answer all calls, this model can be used in the optimization of high service level callcenters. Alexandre(2007) has set up the markov chain models of a telephone call center with call blending and these models are used in a realistic call center. Ken K(2007) has put forward a queuing model of call center based QBD (quasi-birth-death) model, and this model can be used to improve the efficiency of call centers. Sung M B(2005) has studied the problem of obtaining the customer’srequirements with data mining in call centers. Different from these research results, based on the analysis of the process of telecommunication, a continuous service quality improvement model is put forward. With the analysis of historical database using OLAP (On-Line Analytical Processing) and data mining tools, the service quality metrics are analyzed for improving the efficiency of the call centers....