Skip to Main content Skip to Navigation

Modèles stochastiques pour l'aide à la décision dans les centres d'appels

Abstract : In the recent years, the call center industry has been rapidly expanding. The companies are more and more frequently selecting this solution which offers them a privileged relation with their customers. This provides these companies a convivial way to better serve their clients and acquire new ones. This work is motivated by problems of the company Bouygues Telecom.
In the first part of this work, we study the phenomenon of retrials. We start, initially, by proving its importance. Thereafter, we show that the fact of ignoring this phenomenon generates errors when staffing the call center, resulting in unsatisfied target service levels or an overstaffed system with huge costs. This analysis is performed in the stationary regime. In the study of the multi-period system, we determine the evolution of the real time demand, based on observed arrivals.
In the second part, we study a multi-class system with non-preemptive priority and start by proposing estimators of customer waiting times. We analyse different routing policies based on these estimators. We compare the system with several queues with the single queue one. We finish this part by comparing two probabilistic priority policies with the well-known strict priority rule. The performance of each system is measured by the number of service representatives needed to achieve the target service level
Document type :
Complete list of metadatas
Contributor : Anne Prevot <>
Submitted on : Friday, April 17, 2009 - 11:46:11 AM
Last modification on : Wednesday, July 15, 2020 - 10:36:10 AM
Long-term archiving on: : Thursday, June 10, 2010 - 8:50:26 PM


  • HAL Id : tel-00376312, version 1



Mohamed Salah Aguir. Modèles stochastiques pour l'aide à la décision dans les centres d'appels. Sciences de l'ingénieur [physics]. Ecole Centrale Paris, 2004. Français. ⟨tel-00376312⟩



Record views


Files downloads