Call centre staffing is important as the workforce accounts for 60-70% of the total operating cost of a call centre. The staffing procedure involves two distinct but interrelated research areas: a) forecasting the call arrival rates and b) modelling the call centre as queueing system to decide on staffing levels, using the forecast arrival rates. However, while most of the forecasting methods developed assume time-varying arrival rates, the classic Erlang-C based queueing model for staffing assumes stationarity and does not consider the dynamic transient behaviour of the call centre queueing system. In addition, the forecasting models themselves are imperfect and thus have forecast errors, which also affects the performance of the queueing models. This thesis specially designed an experimental framework that mirrors the key elements required by the call centre staffing activity. The framework enables us to conduct a wide range of experiments for analysing the individual, combined and interactive effects of forecasting methods and queueing models for staffing on call centre system performance. We introduce a geometric discrete time modelling (Geo-DTM) approach and use it with an iterative-staffing algorithm (ISA) to determine staffing levels. Empirical tests show that under perfect knowledge of arrival rates, there are many benefits of using the Geo- DTM+ISA method compared to steady-state staffing methods. With simulated call arrivals data, we evaluate the effects of forecasting errors on call centre performance using various forecasting models and the Geo-DTM+ISA for staffing. The results show that even with a good quality dynamic queueing model (Geo- DTM+ISA), better forecasting accuracy does not necessarily translate into better service levels. The system performance exhibited depends on a combination of factors. We also study the combined effects on call centre performance in the likely practical case where both forecasting and queueing models are suboptimal. Our results show that under a quality driven service regime, stationary models perform similarly to Geo-DTM+ISA. However, under an efficiency driven service regime, the stationary based staffing methods perform much worse than Geo-DTM+ISA, although both are affected by forecasting errors. Insights from the empirical results are used to provide guidance for call centre workforce management.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:747980 |
| Date | January 2014 |
| Creators | Chen, Xi |
| Publisher | Lancaster University |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
Page generated in 0.0159 seconds