Abstract

In today's global economy, outsourcing has become increasingly popular in design and production. Manufacturers need to share massive information with their suppliers during an outsourcing activity. In the meantime, the manufacturers need to protect confidential information related to the product for the purpose of intellectual property (IP) protection. In such a context, secure collaboration has become an emerging research topic in global supply chain management. A number of methods were proposed to select secure and eligible suppliers among all potential suppliers involved in a supply chain system to satisfy security requirements and to minimize the cost at the same time. The selection can be performed by assessing suppliers' ability, risk assessment of information leakage, and cost analysis. However, depending on given security requirements, a valid selection of suppliers to meet such requirements may not always be possible to obtain. Moreover, such a selection process is usually very expensive with existing risk assessment algorithms. This thesis addresses both issues by proposing a method which is both secure and efficient for generating optimal selections of suppliers for a supply chain system. First, we introduce a multi-level trust model of suppliers to address the cases where existing approaches based on a flat trust model will fail to generate any valid supplier selection. To our best knowledge this is the first work on formally modeling the level of trust in suppliers. Second, we propose efficient heuristic algorithms for eliminating insecure selections of suppliers as early as possible in the process such that the need for expensive risk assessment on such selections is avoided. The effectiveness and efficiency of proposed approaches were analyzed and validated through a case study.