Abstract
This paper presents a methodology for estimating flowtimes and setting due-dates in complex production systems. This is accomplished by modeling flowtime estimation as a forecasting problem, and using the empirical distribution of forecast errors to set job due-dates in production settings with multiple workcenters, multiple servers, feedback queues, and machine breakdowns. Several due-date performance objectives are considered, including cost minimization, attainment of service level targets, and minimization of mean absolute lateness and mean squared lateness. Simulation experiments demonstrate the effectiveness of the method in comparison with both theoretical and empirical methods previously introduced in the literature.