Repairable item inventory model optimization with uncertainty theory

Traditional spare parts optimization models based on the probability theory have greatly improved the performance of support system. However, those models have suffered limitations in factual situations due to the lack of adequate statistical data. Uncertainty theory is utilized in this paper to deal with this problem. We introduce an uncertain variable to denote uncertain demands, and describe a single operating base supply system briefly. Then two uncertain spare parts optimization models are proposed for the repairable-item inventory system, including expected model and the chance constraint programming model. We utilize the genetic algorithm for mathematical model solution. Finally, a numerical example will be provided for the illustration of the effectiveness of the uncertain models and the algorithm.