A Procedure, Involving Simulation, For Selecting The Control Variables Of An (S, c, s) Joint Ordering Strategy^*

* This work was supported by a Canada Council Scholarship and by the Defence Research Board of Canada Grant No. 9740-16.

Abstract

This paper considers an inventory situation where the control procedure is dependent in nature, i.e. an item that would not be ordered at a particular point in time if it was being controlled independently, is allowed to be included in a replenishment triggered at that moment by the necessity of ordering another item. In particular, we develop a procedure for selecting the control variables (order-up-to-levels, can-order points, and must-order points) of an {S, c, s) system. The procedure, iterative in nature, is a combination of mathematical optimization and simulation. The cost savings over a usual independent inventory system are demonstrated.

Résumé

Cet article considère un problème de gestion de stock dans lequel la politique de commande est dépendante par nature. La commande d’un des articles du stock peut déclencher la commande d’un autre article dont le réapprovisionnement n’aurait pas été effectué dans le cas d’une politique de commande indépendante. Nous développons en particulier une méthode aboutissant au choix des variables de contrõ1e (seuil de réapprovisionnement, seuil possible de commande, seuil de commande) d’un modéle (S, c, s). La methode itérative utilisèe combine à la fois optimisation mathématique et simulation. L’article présente enfin les gains realisés par rapport a un modèle de gestion indépendante.

Notes

* This work was supported by a Canada Council Scholarship and by the Defence Research Board of Canada Grant No. 9740-16.

Additional information

Notes on contributors

Jean-Paul Schaack

JEAN-PAUL M. SCHAACK was born in 1947 in Metz, France and graduated in 1969 from the Ecole Nationale Sup6rieure des Mines de Paris where he specialized in OR and Applied Mathematics. In 1970 he received an MA SC in Management Sciences from the University of Waterloo. After returning to France, he worked with the French Institut de Recherche de la Sid6rurgie, where he was involved in optimization problems encountered in the control process of a blast furnace.

Edward Silver

EDWARD A. SILVER was born in Montreal, received a B ENG (Applied Mechanics) from McGill University in 1959 and an sc D in Operations Research from the Massachusetts Institute of Technology in 1963. His professional consulting experience includes four years as a member of the Operations Research Group of Arthur D. Little Inc. Other positions held in the past include Special Lecturer in Systems Engineering and Operations Research at MIT and Associate Professor of Business Administration at Boston University. In September 1969 Dr Silver was appointed to the faculty of the Department of Management Sciences at the University of Waterloo. His research and consulting activities centre on the practical applications of probabilistic models, with particular emphasis on bridging the gap between theoretical optimality and practical utility of solutions to complex problems.