Spare Parts Quantity Problem Under Uncertainty – the Case of Entirely New Devices With Short Life Cycle

The paper presents a new scenario-based decision rule for the spare parts quantity problem (SPQP) under uncertainty with unknown objective probabilities. The goal of SPQP is to ensure the right number of extra parts at the right place at the right time. In the literature, SPQP is usually regarded as a stochastic problem since the demand for extra parts is treated as a random variable with a known distribution. The optimal stock quantity minimizes the expected loss resulting from buying a given number of parts before potential failures. The novel approach is designed for the purchase of non-repairable spare parts for entirely new seasonal devices, where the estimation of frequencies is complicated because there are no historical data about previous failures. Additionally, the decision maker’s knowledge is limited due to the nature of the problem. The new procedure is a three-criteria method. It is based on the Hurwicz and Bayes decision rules and supported with a forecasting stage enabling one to set the scenario with the greatest subjective chance of occurrence. The method takes into account the decision maker’s attitude towards risk and the asymmetry of losses connected with particular stock quantities. We assume that the future unit purchase cost of a service part bought after the breakdown is also uncertain and given as an interval parameter. The approach is designed for short life cycle machines.