Modelo de estoque para peças de reposição sujeitas à demanda intermitente e lead time estocástico

Abstract

Inventory control of spare parts is very critical to most companies due to high inventory costs associated to these items and to the complexity of developing inventory models to control them. Most items are slow-moving with irregular and intermittent demand. This kind of demand pattern increases the complexity of forecasting demand. In this paper we have developed a methodology to classify and control spare parts subject to stochastic demand and lead time. It has used the Single Demand Approach SDA defined by Krever, Wunderink, Dekker & Schorr (2005) to calculate the mean and variance of lead time demand (LTD). The inventory models have used statistical distributions of Poisson, Laplace, Gamma and Normal combined with the optimization of the total cost in order to model the LTD and calculate reorder point, order quantity and safety stock. Using real data and simulation process, the inventory models were tested and their performance were compared and evaluated. The results showed a superior performance of Laplace model in terms of service level, fill rate and inventory cost. Compared with current system based on Normal distribution, the Laplace proposed model obtained significant savings and good results in service level and fill rate. The benefits for company in using this model subject to cost optimization process are described in the conclusion of study.