4 présentations

• 
  10h30 - 10h55

  Study of Call Type Dependence in a Multiskill Call Center

  • Amel Jaoua, prés., Université de Montréal
  • Pierre L'Ecuyer, Université de Montréal

  We are interested in modelling the nonlinear dependence between call types of an Hydro-Québec Call Center. For this purpose we use copulas. After characterizing this dependence structure, simulation experiments are carried out with ContactCenters to assess the impact of the dependence between call types on the call center output performances.

• 
  10h55 - 11h20

  Forecasting Intraday Arrivals at a Call Center

  • Rouba Ibrahim, prés., University of Montreal
  • Pierre L'Ecuyer, Université de Montréal

  We evaluate alternative time series methods for forecasting intraday (within-day) call volumes at a call center. Our methods take into account both interday (day-to-day) and intraday dependence structures and allow for dynamic intraday updating of call volume forecasts. We describe results from an empirical study analyzing real-life call center data.

• 
  11h20 - 11h45

  Call Center Routing Policy Based on Call Waiting and Agent Idle Times

  • Wyean Chan, prés., Université de Montréal
  • Pierre L'Ecuyer, Université de Montréal
  • Ger Koole, VU University Amsterdam

  The importance of routing policies in multiskill call centers is often overlooked. The policies implemented in routing hardware are usually too rigid like priority routing. We present a flexible routing policy based on call waiting and agent idle times that performs generally better than traditional routing policies.

• 
  11h45 - 12h10

  Distributional Robust Applications in Call Center Workforce Scheduling with Uncertain Arrival Rates

  • Jean-Philippe Vial, prés., ORDECSYS
  • Shuangqing Liao, Ecole Centrale, Paris
  • Christian van Delft, HEC, Paris

  Call center scheduling is critical to meet target service levels. In this paper, we consider uncertain arrival rates, that vary according to intra-day seasonality and global dayload factor. Both factors (seasonal and global) are estimated from past data and are subject to errors. We propose an approach combining stochastic programming and distributional robust optimization to minimize the total salary costs. The performance of the robust solution is simulated via Monte-Carlo techniques and compared to the pure stochastic programming and the pure min-max solutions.