This thesis presents a study of the effects of disruption on the production and inventory replenishment decisions of the overall supply chain system. The study focused on analyzing the optimal inventory and recovery policies for each node, so that the supply chain system is optimized as a whole. The problems considered are instances of the class of inventory management problems under disruptions with a finite horizon. A decision support system tool for managing supply chain disruptions was presented. Specifically, an efficient supply chain optimization framework was developed, in which the problems were formulated as mathematical models and heuristic algorithms were developed to solve these underlying optimization problems. The framework was applied to three different models: (1) a single stage production-inventory system following the Economic Production Quantity system (2) a two stage serial supply chain system and (3) a three stage supply chain system with multiple suppliers. Two different types of disruptions were explored, including (1) supply disruption and (2) transportation disruption. The results of the experimental analysis showed that the optimal recovery schedule is highly dependent on the relationship between the backorder cost and the lost sales cost parameters. When lost sales cost is low, it is found that the recovery duration is shorter. However, as the lost sales cost increases, it is more cost effective to have backorders, thus the recovery duration will become longer. In addition, the efficient heuristics developed in all three cases performed well and provided quality solutions in reasonable time. In the final work of this thesis, a simulation model was designed to investigate the effects of various pre-defined disruption scenarios over time. Results from the study showed that supply disruption at the supplier with higher inventory holding cost causes higher recovery costs. As for transportation disruption, the solution is more sensitive to the lot size as opposed to the cost parameters, when lost sales cost is large. A statistical analysis to find the correlation between the core dependent variables and total system cost was performed, which showed that backorder quantity and recovery duration have strong positive correlations with total cost.
展开▼
