A simulation-based approach for fleet design in a technician dispatch problem with stochastic demand

We address the problem of developing policies for selecting the proper number of technicians to hire (fleet size) for an on-call repair service operating over a yearly planning horizon in an environment where the number of requests to be serviced each day can vary significantly from day to day and on a seasonal basis. We propose a new approach based on the simulation of a large number of sampled weekly instances and the application of a previously developed optimization procedure for the daily dispatch of technicians. The sampled instances are derived from an extensive statistical analysis of demand data with respect to several key parameters. The results of the simulations are utilized to adjust performance curves in function of fleet size that are then used in an economic analysis of the trade-offs between service quality and cost. Efficient policies for fleet design are then deducted from this analysis.