SciELO - Scientific Electronic Library Online

 
vol.16 issue3An enumeration algorithm based on partial ordering to solve the minimization of tool switches problemMethodology for the construction and choice of credit risk prediction models author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand

Journal

Article

Indicators

Related links

Share


Gestão & Produção

Print version ISSN 0104-530XOn-line version ISSN 1806-9649

Abstract

BOUZADA, Marco Aurélio Carino  and  SALIBY, Eduardo. Forecasting a call center demand using a Multiple Regression model. Gest. Prod. [online]. 2009, vol.16, n.3, pp.382-397. ISSN 1806-9649.  http://dx.doi.org/10.1590/S0104-530X2009000300006.

This work describes - with the aid of a case study -a demand forecast problem for a specific product reported to the call center of a large Brazilian company in an industry called Contax, and the way it was approached with the use of Multiple Regression using dummy variables. After highlighting and justifying the studied matter relevance, the article presents a small literature review regarding demand forecast methods and their use in the call center industry. The case is described presenting the studied company and the way it deals with the Forecasting Demand for a telephone all center regarding telephone services products. Therefore, a Multiple Regression with dummy variables model was developed to work as the basis of the proposed demand forecast process. This model uses available data capable of influencing the demand such as the week day, occurrence of holidays, and the date of critical events such as the date on which the bill is sent and the date of payment collect. The model presented an improvement of Demand Forecasting Accuracy of 0.3% in the studied period when compared to the previously tool in use

Keywords : Call center; Demand Forecast; Multiple Regression.

        · abstract in Portuguese     · text in Portuguese     · Portuguese ( pdf )

 

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License