A Multi-Level Systems Approach to the Design of Sustainable Integrated Supply Chains

Jin Can

Abstract

Global competition has imposed great pressure on companies. To design an efficient and resilient supply chain is crucial for maintaining companies’ business edge. Problems like selecting business partners, establishing connection between entities and so forth, often involve lots of careful considerations. This thesis proposes a four-level hierarchical systems approach for the supply chain network configuration and risk management at the planning-management stage. Decision making is required at each level. More precisely, the first level evaluates performances of potential candidates and selects the best ones to join the system. The second level integrates the unconnected partners across different stages by the use of the optimization methods. Most of the time, the objective is to maximize profits or minimize costs subject to a set of constraints. The third level uses Bayesian Network to identify the overall risks and vulnerability within system and the dependency between them. Influence Diagram is employed to analyze the impacts of different policies. The fourth level validates and evaluates the proposed supply chain. The networks with inadequacies will feedback to the previous levels for reconfigurations. The four levels interactively work together and move towards an optimal solution according to the required objectives of the problem. The overall decision process is the integration of the four levels. This thesis presents the methodologies used in each level, and illustrates the hierarchical process by case studies.