DEVELOPMENT AND APPLICATION OF DECISION RULES FOR BUFFERING AGAINST DEMAND UNCERTAINTY IN A TIME-PHASED REORDER POINT SYSTEM
Abstract
This research investigates the use of service level decision rules in a single stage time phased reorder point system. Time phased reorder point uses the forward-looking logic of Material Requirements Planning and offsets the order quantities for lead time. The research uses multiple regression analysis for response surface mapping to develop models for buffering against demand quantity uncertainty. Several methods of buffering are used; including the use of safety stock, an inflated order cycle, and a combination of the two. The analysis relates two performance criteria, the total cost and the service level of the system to several factors that are operating characteristics of the firm and demand characteristics. Several tests were performed to test the benefit of using the complex decision rules. Among these was a test against a simple linear model. The decision rule performed better than a simple linear model, but the decision rule is more complex computationally and more difficult to interpret. This study is the first to develop procedures for determining the amount of buffering to use in this type of inventory system, and the major conclusion is that the use or safety stock is preferable to the use of other procedures.
Degree
Ph.D.
Subject Area
Management
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.