Safety Stock versus Safety Time in MRP Controlled Production Systems

Abstract

The two management set parameters which determine the performance of a material requirements planning (MRP) system are the lead time and the safety stock. The appropriate values of these parameters are influenced by the accuracy of forecasts over the lead time, the variability of processing time and the degree of congestion, together with the costs of inventory and shortages. These influences are explored using stochastic models of a single stage manufacturing system for which work release is controlled using MRP. The major conclusion is that safety time is usually only preferable to safety stock when it is possible to make accurate forecasts of future required shipments over the lead time, otherwise safety stock is more robust in coping with changes in customer requirements in the lead time or with fluctuations in forecasts of lead time demand.production management, inventory control, input control, stochastic models

Re-Examining the Performance of MRP and Kanban Material Control Strategies for Multi-Product Flexible Manufacturing Systems

“…Regarding the first issue, Buzacott and Shanthikumar (1994) and Karaesmen et al (2002) analyze single product systems operating under MRP (push) policies under various scenarios. Both these studies assume that accurate estimates of lead times are available.…”

“…Both these studies assume that accurate estimates of lead times are available. With respect to determining future customer requirements, Buzacott and Shanthikumar (1994) consider two situations (a) future demands over the lead times are known exactly, and (b) only the mean demand rate is known, while Karaesmen et al (2002) only consider the former environment where future requirements over the lead times are known exactly. For the comparisons of multi-product systems in this paper, we also focus on the first environment and assume that accurate estimates of release lead times are available.…”

“…Regarding safety stock and safety lead times, Lambrecht et al, (1984), Buzacott et al (1992), and Buzacott and Shanthikumar (1994) report several key results for single product systems. First, they report that the service delay experienced by a customer order decreases by increasing the safety stock at any stage.…”

“…When safety lead times are used, the release lead time L i for each product is set higher than its expected product flow time F i by an amount known as the safety lead-time. Safety stock for a given product is the inventory target in the output buffer when there is no unmet demand and no lots delayed in the manufacturing process beyond the actual flow time (Buzacott and Shanthikumar, 1994 We are not aware of any literature to date that shows when safety lead time would be preferable to safety stock in multi-stage multi-product manufacturing systems. To gain insight into this question, we study the safety stock and safety lead time policies independently.…”

A Revised and Improved Version of the MRP Algorithm: Rev MRP

“…safety stock levels, planned lead time, time fences, planning horizons) [4,5,6,7,8]; • need for final demand forecast and related impacts on inventory costs and shortages [9];…”