عنوان مقاله [English]
A call center is a centralized office used for the purpose of receiving and transmitting a large volume of requests by telephone. A call centre often operates through an extensive open workspace for call centre agents, with work
stations that include a computer for each agent, a telephone set/headset connected to a telecom switch, and one or more supervisor stations. Call centers can operate independently or be networked with additional centers,
called a multi call center.
System configuration optimization is one of the main problems in multi call centers. Improvement of system utilization and customer satisfaction are the most important objectives in a call center system.
In recent years, different approaches have been developed for solving this problem. Most of these approaches only stress the analysis of system performance. Furthermore, these approaches usually focus on either customer or
employer viewpoints. In this paper, a multi-objective optimization approach, based on response surface methodology and the design of experiments, is proposed. The proposed approach maximizes customer satisfaction and system
utilization simultaneously. Response surface methodology (RSM) is a collection of mathematical and statistical techniques that are useful for modeling and analysis in applications where responses of interest are influenced by several variables, and the objective is to optimize these responses simultaneously.In the proposed approach, decision variables are the number of operators in any center, number of links between centers and length of waiting queue.
Simultaneous objectives include maximization of operator utilization and link utilization; minimization of the percentage of balking calls, and customers waiting time in the system.
The proposed approach consists of three phases; design of experiments, simulation and optimization. First, a factorial design is developed to evaluate system performances. Then, an aggregated method is used to convert various
objectives into a single objective function. In this study, the aggregation measure is the probability of all four objective functions simultaneously meeting their desirable regions (P(Y$in$S)). In order to estimate this measure, we simulate the system. In the next step, the response surface of the aggregated measure, with respect to design variables, is obtained. Finally, by maximization of P(Y$in$S) , optimum values of factors are obtained.
The validity and performance of the proposed method were investigated on the data of a multi call center system consisting of three call centers. The results of the proposed method were compared with the solutions of common
methods in the literature. The result of the proposed method outperformed the results obtained by other approaches.