A Comparative Study of Lot-Sizing Techniques under Fluctuating Demand

Abstract

This research aims to compare the efficiency of different lot-sizing techniques under fluctuating demand patterns. Demand was simulated using a normal distribution and categorized into two levels based on the variability coefficient (VC): low variability (VC < 0.25) and high variability (VC ≥ 0.25), across two planning horizons: 12 periods (monthly) and 52 periods (weekly). The study evaluates both the exact optimization approach using the Wagner–Whitin algorithm (WW) and various heuristic methods, including Lot-for-Lot (LFL), Economic Order Quantity (EOQ), Period Order Quantity (POQ), and Part-Period Balancing (PPB). Each scenario was tested across two levels of ordering cost and five levels of holding cost. The results indicate that the WW algorithm outperformed all other techniques across all simulated scenarios by achieving the lowest total inventory management cost. Among the heuristic methods, LFL and POQ emerged as viable alternatives; their efficiency improved significantly as holding costs increased because they align order quantities closely with actual demand, thereby reducing average inventory levels and carrying costs.