关键词: 多产品, 库存系统, （R, Q）策略, 系统动力学, 启发式算法

Abstract: To optimize long-run average cost of the multi-item inventory system with more than one probabilistic resource constraint and with decaying products, several fundamental theorems of general multi-item inventory system with some resource constraints were derived firstly. And an efficient polynomial algorithm to find the optimized stock policy was put forward based on these theorems. A system dynamics simulation model was used to replace complicated calculation processes of the traditional algorithm. Multi-item inventory system with several probabilistic resource constraints was equivalent to certain multi-item inventory system with only one probabilistic resource constraint was demonstrated. Finally, an example was given to show that the stock system with probabilistic resource constraints outperformed the one with traditional linear resource constraints sharply in multi-resource environment and the new algorithm was feasible.

Key words: multi-item, inventory system, (R, Q) policy, system dynamics, heuristic method