Supply disruption in multistage production systems

Abstract

This study aims to analyze multistage production systems under supply disruption. First part of the study examine an N stage serial system under supply disruption risk. It is assumed that the unreliable supplier is unique and known beforehand. It is assumed that the disrupted supplier can not place a new order during disruption however it can continue to deliver its on hand inventory. An approximation approach for an N stage serial system is also developed. A numerical analysis is presented and the reorder intervals that are integer multiples of the reorder intervals of their successors are obtained for the exact model of a two-stage serial system. In the approximate model, power-of-two policy is applied to determine the reorder intervals. Furthermore, the error due to applying the approximate model and the error due to using the power-of-two policy are computed. It is shown that the approximation provides good performance when the reorder interval of the unreliable supplier is large enough. Second part of the study deals with the assembly systems that are due to supply disruption. The examined assembly system is a simple system that consists of the stage that forms the nished product and its direct predecessors. It is assumed that the unreliable supplier is unique and can be one of the direct predecessors of the root stage. It is also assumed that the proportion of the reorder intervals of the direct predecessors of the root stage to each other and to the root stage must be integer. The model is analyzed by taking into consideration two cases since the average cost of the system changes according to the relation between reorder intervals. An approximation approach is developed by approximating the probability of disruption. A three-stage assembly system is examined in the numerical analysis part.