Abstract
The paper reports the results of several simulation experiments that compare alternative procedures for determining purchase quantities in MRP systems when multiple discounts are available. The procedures are the modified least unit cost (LUCm), the modified MeLaren's order moment (MOMm) and a ‘traditional’ procedure (DOQ) that is commonly used when evaluating quantity discounts. For this work, three purchased parts with two levels of discount attractiveness, two price break offerings, requirement lumpiness and requirement uncertainty were used to evaluate each lot-sizing procedures. A simulation model was used to test 810 observations. This research indicates that the LUCm and MOMm procedures outperform the traditional procedure when multiple discounts are available. In addition, for high discount attractiveness between the price breaks, the LCCm and MOMm procedures are more than five per cent better than the traditional procedure. When the discount attractiveness between the price breaks is low, there is very little difference in the performance of the alternative procedures. Moreover, the LUCm procedure requires 13 45 per cent fewer units of inventory than the MOMm procedure and 66.35 per cent fewer units than the traditional procedure. LUCm emerges as the procedure of choice for the multiple discount lot-sizing problem in material requirements planning (MRP) systems. The LUCm procedure is easily understood; it takes advantage of multiple discounts; it uses considerably less safety stock and places fewer orders.