Research On Call Center Human Resource Management System Based On Genetic Algorithm

The reasonable and effective human resource management, as an important part of the call center operation and management system, can help to improve the efficiency of call center operations and try to achieve the maximum benefits. It is difficult to allocate resources under the certain cost and avoid waste or redundance of human resource and get balance between the cost and the service level. Based on the actual data of a company call center, this thesis develops a call center human resource management system, which include the main system functions of the number of arriving calls forecasting and the shift scheduling and seating arrangement and real-time status tracking and performance evaluation and statistics, with the focus on the number of agents needed to forecaste algorithm and the shift scheduling algorithm.The number of arriving calls and the average value of AHT are studied, which are the main input parameters of ErlangC queuing model. The second and third Exponential Smoothing Algorithm are used to forcast the number of arriving calls. The result shows the MAPE of the second Exponential Smoothing Algorithm is smaller than the third Exponential Smoothing Algorithm. Based on exponential distribution, the average value of AHT is estimated and the result of the data analysis shows the estimate value can follow the trend of AHT. Then, the number of agents needed in every hour in the example is calculated based on ErlangC queuing model by a certain service level.Further more, the business characteristics of the shifts creating phase of call center staff scheduling is analysed thoroughly. The staff scheduling is modeled and the objective function and constraints functions are designed and the basic theory of the genetic algorithm is applied to the study of call center shift scheduling. According to the genetic algorithm, the fitness function and the genetic operation functions are designed. The simulation is done by the Matlab genetic algorithm toolbox. The result shows that the genetic algorithm is an effective way to solve the taff scheduling problem.