Dynamic rules for due-date assignment

Abstract

This paper presents two new dynamic due-date assignment rules which utilize shop congestion information. The new rules estimate job flowtime based on a sampling of recently completed jobs. These rules are compared with other established flowtime estimate models on the criterion of due-date performance via computer simulation. To evaluate the robustness of the rules, an experimental design with three different queue sequencing heuristics and two different shop balance levels was used. The results of this investigation clearly indicate that flowtimes from recently completed jobs provide very useful information for establishing effective due-dates in a job shop environment. In addition it is shown how the use of particular sequencing rules greatly increases the precision of flowtime estimates.