On using higher-order moments for stochastic inventory systems

The lead time demand in a stochastic inventory system is often simplified as a normal distribution in practice, whereas the higher-order moments are seldom applied in inventory management due to the difficulty of accessing necessary statistics tables for practitioners. In this paper, we have introduced the Bowman-Shenton formulae for calculating safety factors in modelling a compound distribution. This approximation makes it easy to implement the higher-order moment approach. Our numerical examples show that the system performance measure, in terms of service level, has been significantly improved when we use a Pearson curve, especially at a high service level.