MODELING AND MANAGING THE COMPLEXITY OF LOGISTICS NETWORKS

Abstract

Modern supply chains and logistics networks are becoming increasingly complex. It is commonly believed that complexity impedes performance, and proactively managing complexity can lead to better performance and efficiency. However, complexity management has not been well-established and widely implemented in practice. This result is partly because little effort has been made to develop tools for complexity management. Motivated by this, the dissertation aims to develop practical complexity models. To begin with, this dissertation proposes a novel complexity measure, whose superiority over the existing ones is demonstrated by theoretical proofs and subsequently verified by numerical studies. This dissertation then proceeds to explore how to integrate complexity management into two logistics network design problems, namely, a robust facility location problem under facility disruption and a parcel locker network design problem for last-mile delivery. Through illustrative examples and case studies, this dissertation shows the effectiveness of the proposed approaches and provides implications.