Final phase inventory management of spare parts under nonhomogeneous poisson demand rate

Abstract

In product lifecycle, there are three phases, initial phase, normal phase and final phase. Final phase begins when the product is out of production, and ends when the last contract expires. It is generally the longest period in the lifecycle. Although the product is not manufactured any more, spare parts of the product need to be supplied to the market. Firms need to provide these parts at the retailer level until the end of the phase due to legal responsibilities. Because of lack of historical data and unavailability of forecasting, retailers need a systematic policy to decide replenishment quantity and time to prevent excessive holding, backordering, unit and setup costs. In our problem, we assume that demand of the spare part is a non-homogeneous Poisson process where the rate parameter is a non-increasing function of time. We consider all costs and lead time are fixed and known. Due to characteristics of the final phase, the planning horizon is taken as finite and known. In this study, we developed two alternative heuristics for retailer’s problem to minimize total cost during the final phase. First heuristic is a continuous-review policy based on estimation of future replenishments by solving series of deterministic demand subproblems. Second heuristic is a periodic-review policy with variable period lengths, which solves myopic problems, by selecting subsequent time points to check inventory position. We also developed a simulation model to evaluate performances of the heuristics. This study provides an efficient way to decide on replenishment quantity and time. Limited numerical results show that heuristics provide near-optimal results for homogeneous cases studied in the literature. Moreover, this is one of the initial studies that considers final phase with non-homogeneous demand rate. In that sense, it makes a contribution to the literature of final phase problems and provides a systematic way of replenishment decisions for the retailers.