콜센터 인입 콜량 예측을 위한 시계열 모델 비교 분석

The critical decision-making task of the call center manager is to determine the appropriate number of agents who can respond to the call with minimal cost. To do this, it is necessary to predict the exact amount of incoming calls. However, there are not many studies on this at home and abroad. In actual call center, simple calculation method based on experience and intuition of the person in charge is still mainly used. In this study, we investigate various techniques and cases to predict call volume and develop and verify optimized models. We develop the call prediction model by using decomposition model, ARIMA model, regression model and artificial neural network based model, and fitting of optimized prediction model by using real call center data and verify its effectiveness.