Article ID: | iaor19992980 |
Country: | Netherlands |
Volume: | 109 |
Issue: | 3 |
Start Page Number: | 550 |
End Page Number: | 558 |
Publication Date: | Sep 1998 |
Journal: | European Journal of Operational Research |
Authors: | Ryan Sarah M. |
Keywords: | time series & forecasting methods |
This paper describes a simulation study of the effect of forecast revisions and hedges against demand uncertainty in a rolling horizon heuristic for capacity expansion. The model is based on data collected in the utilities division of a large chemical manufacturing plant. A seasonal integrated moving average model for the demand is used to generate forecasts, while capacity additions are determined by applying a simple timing rule to various hedges around the forecast. The simulation results indicate that hedging forecasts by their prediction limits rather than a fixed buffer significantly reduces undercapacity at the expense of a small increase in capacity cost. The prediction limit hedge is more robust to delays in reforecasting.