A simple, robust leadtime-quoting policy

We examine leadtime-quoting policies for minimizing average lead time subject to customer service constraints on fill rate, tardiness, or relative tardiness in simple systems with exponential and normal processing times. By studying the resulting safety leadtimes implied by each policy we gain insight into why some policies perform more robustly with respect to different measures of customer service than do others. This analysis suggests that a simple constant safety leadtime policy should work reasonably well under most conditions. A series of simulation experiments of more complex production environments indicates that the constant safety leadtime policy does indeed exhibit robust performance. This, plus the fact that it is extremely simple to adapt to a wide range of production environments, makes it an attractive basis for real-world leadtime-quoting systems.