Inventory Policy by Dynamic Order

This paper proposes a novel inventory control policy called the dynamic order policy. The proposed policy dynamically adjusts the reorder quantity based on two threshold levels (s∗, s, S, S∗), enhancing its flexibility compared to the traditional policy (s, S). The new policy allows placing a larger order when the inventory drops below a threshold of s∗ and places a standard order when the inventory falls between the thresholds of s∗ and s. A genetic algorithm is employed to optimize the inventory decision parameters due to its ability to solve complex, nonlinear discrete problems. The model is tested under various simulation scenarios by demand rate, lead time, and customer volume. The results demonstrate that the proposed policy reduces the total average cost by up to 6.7% compared to the traditional policy. This dynamic framework presents a promising alternative for managing uncertain inventory environments with backlogs and variable lead times.