Forecasting future customer call volumes: A case study

Citation:
Njeru K, NA K, IJMwaniki. "Forecasting future customer call volumes: A case study." International Journal on Future Revolution in Computer Science & Communication Engineering. 2018;4 (6):12-16.

Abstract:

Forecasting future volumes of customer calls in call centers has proved to be a tedious and challenging task. This study, using time series analysis proposes two adequate ARIMA (p, d, q) models that are suitable to forecast two volumes of customer calls, IVR Hits Volumes and Offered Call volumes. 1472 times series data points from date 01/01/2014 to 11/01/2018 were obtained from a call center based in Kenya on the two variables of interest (IVR Hits Volumes and Offered Call volumes). The appropriate orders of the two models are picked based on the examination of the results of the ACF and PACF plots. The AIC criterion is used to select the best model for the data. The best ARIMA model for log IVR Hits volumes is ARIMA (5, 1, 3) with and the best ARIMA model for log Offered Call Volumes is ARIMA (6, 1, 3) with. The two models are recommended to model and forecast the daily arrival volumes of customer call data. The obtained forecast will be used in providing insights for appropriate workforce management

Notes:

Keywords
Call Center, Forecasting, Time series, Workforce Management

Website

UoN Websites Search