Title: SARIMA models for forecasting call volume in emergency services

Authors: E.V. Gijo; N. Balakrishna

Addresses: SQC and OR Unit, Indian Statistical Institute, 8th Mile, Mysore Road, Bangalore 560059, India ' Department of Statistics, Cochin University of Science and Technology, Cochin – 682 022, Kerala, India

Abstract: Management of manpower and other resources is one of the most important tasks to be carried out in any business organisation. Various approaches are followed for the planning activities in industry and business. This article discusses the manpower planning and allocation problem in a call centre, working round the clock for handling emergency telephone calls related to ambulance services. The organisation under study provides ambulance services to take care of the emergency needs of the people in one of the Indian states. As a first step in the manpower planning process, we develop time series models for forecasting the call volume. Time series models were used to explain the evolution of hourly and daily call volume data. As the series was showing seasonality, seasonal ARIMA models were fitted for these data. The adequacy tests confirm that the selected model performs well for the data. The short-term forecasts computed using the model seen to be very close to the corresponding observed values. These forecast values will be helpful in manpower and resource planning.

Keywords: emergency services; time series models; modelling; forecasting; seasonal ARIMA model; ADF test; ACF plots; PACF plots; stationary series; residual analysis; Box plot; SARIMA; call volumes; emergency calls; manpower planning; manpower allocation; call centres; ambulance services; India; resource planning.

DOI: 10.1504/IJBEX.2016.079252

International Journal of Business Excellence, 2016 Vol.10 No.4, pp.545 - 561

Received: 23 Dec 2014
Accepted: 27 Jan 2015

Published online: 15 Sep 2016 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article