2026/04/13 by Hussein Shakir, Nerda Zura Zaibidi · 1 voice
Business, Management and Accounting · Decision Sciences · #Forecasting Techniques and Applications #Supply Chain Resilience and Risk Management #Supply Chain and Inventory Management
paper · pdf · doi:10.17654/0972087126072
openalex publication_date 2026/04/13 · openalex created_date 2026/04/14 · openalex updated_date 2026/07/07
Safety stock is a fundamental component of inventory management, intended to protect against uncertainty in demand and lead time. Conventional safety stock formulations rely heavily on symmetric measures of variability, most notably the standard deviation of demand. While statistically valid, such measures implicitly assume that deviations above and below the mean contribute equally to service risk¾an assumption that may be economically misleading when demand variability is asymmetric. This study investigates the performance of standard deviation-based safety stock under asymmetric demand conditions using a simulation-based analysis of a continuous-review (R, Q) inventory system. Demand is generated under three controlled scenarios: (i) variability dominated by downside outliers, (ii) variability driven by upside outliers, and (iii) mixed demand variability combined with stochastic lead time. Four inventory policies are evaluated: no safety stock, standard deviation-based safety stock, an average-maximum rule, and a quantile-based reorder point. Performance is assessed using replicated simulations (50 independent runs per scenario and policy) and reported in terms of service-level metrics (cycle service level, fill rate, stockout frequency and volume) and economic outcomes (average on-hand inventory and total cost). The results show that when demand variability is dominated by downside outliers, standard deviation-based safety stock systematically increases inventory levels and total cost without improving service performance. In such settings, the absence of safety stock achieves comparable service levels at significantly lower cost. When demand includes genuine upper-tail risk or when lead time is stochastic, safety stock becomes necessary to maintain service reliability; however, quantile-based reorder points consistently achieve similar service levels with lower inventory investment than symmetric dispersion-based policies. Overall, the findings demonstrate that the limitation lies not in the safety stock concept itself, but in the use of symmetric uncertainty estimators under asymmetric demand conditions. From a managerial perspective, the study highlights the importance of aligning safety stock estimation methods with the structure of demand risk, thereby avoiding unnecessary inventory accumulation while preserving service quality. Received: January 12, 2026Revised: February 2, 2026Accepted: February 10, 2026