Artificial Neural Network approach for Incoming Call Forecasting in Call Center
Nowadays call center faces problem to figure effective workforce management particularly in agent scheduling. Agent staffing procedure should base on incoming call process modelling, which required balance between maintaining high service level and low operating costs. Forecasting incoming call higher than actual may cause overstaffing and leading to ineffective costs. Otherwise lower number may result high abandonment rates and long waiting time which eventually may lead to customers and revenue loss. In the past, incoming call forecast have been develop with various mathematical models to gain effective workforce management. In this paper an artificial neural network has been proposed to approach time series prediction of incoming call in call center. In this paper a new approach based on artificial neural network has been proposed to predict incoming call in order to support workforce management especially on agent staffing distribution. A feed forward neural network has been formulated and comparation performance under levenberg-marquardt and resilient backpropagation training function is analyzed.