A Research On Spare Parts Inventory Management Optimization

Spare parts inventory management is important for the equipment intensive enterprises to keep continuous operation and make efficient use of inventory budget. Optimization technologies could turn the abundant inventory data in MIS into useful information to help enterprises winning the competition. In this paper, the spare parts inventory optimization problem is investigated from three aspects, thay are single depot problem, multi depot transshipment and inventory-routing problem.For single-depot multiple-item spare parts inventory structure optimization problem, we set the stockout times or stockout piece number as objective, respectively, and use cycle inventory, safety inventory and annual ordering number exchange curve to surport the decision making. At the same time, the unit costs of ordering, stock holding and backorder are estimated.Next, the joint optimization of age replacement and spare parts provisioning policy is discussed. The total expected cost per time unit, as the performance measure of the policy, is evaluated analytically by examining the cost of replacement, breakage, ordering, holding and shortage. A computation procedure is proposed and some numerical results are presented.For base stock policy of single-echelon multi-depot inventory transshipment system, an exact model is set up by using the multi-demension birth-death prosess. When backorder was happened system wide, three different policies, such as ermegency supply, backorder transshipment and non backorder transshipment, are examined. The algorithm for optimizing depot inventory level is also given.Then we analyze the bach ordering policy of single-echelon multi-depot inventory transshipment system. Two sourcing rules for transshipment are considered, such as ramdom source rule and minimum transshipment cost rule. Models for same depots and different depots are developed.The bach ordering policy of two-echelon multi-depot inventory transshipment system is studied next. Three models are developed for these situations: (1) warehouse using batch ordering policy while depots using base stock policy; (2) warehouse and depots all using batch ordering policy but the depots being same; (3) warehouse and depots all using batch ordering policy, the depots being different. The models are examined by simulation.For Inventory-routing problem, a MIP model is developed for the time-varying demand IRP. According to the properties of the problem, hybrid GA approach and branch-and-cut algorithm adding with TSP heuristics are applied. Computational experiments demonstrating the effectiveness of our approaches are presented.