Procurement Policy Analysis in Intermodal Container Logistics

The focus of this thesis is to propose a framework for the development and analysis of procurement policies in intermodal container logistics, a field that involves the coordinated transportation of containers through multiple modes—such as vessels, trains, and trucks—and the management of temporary storage at key points in the supply chain. Procurement is essential to ensuring the smooth flow of container operations by securing access to suppliers capable of fulfilling demand for products and services across diverse logistics activities.

The procurement framework comprises vetted suppliers selected based on criteria such as cost, quality, and reliability. Contracts with these suppliers define mutual commitments, including fixed prices, lead times, and business volumes, while a spot market buffer provides flexibility to address short-term needs. Within this framework, procurement policies guide business volume allocation among suppliers, balancing objectives such as cost-efficiency, delivery performance, and contract compliance, while managing uncertainties in demand and market conditions.

Performance challenges arise from operational and market uncertainties. For vessel operations, where demand appears as onboard requisitions, we propose a discrete-event simulation model to capture the procurement process, including demand generation, supplier selection, and order issuance. The model is parameterized to support generalization across different contexts, with its practical utility illustrated through a numerical study that compares the cost-efficiency and compliance performance of two order allocation policies.

For container drayage operations, where demand is characterized by trucking move volumes, we develop a framework that jointly optimizes the operational volume allocation policy and capacity planning under uncertainty in container flows and spot market conditions. This framework integrates operations with planning to enhance cost-efficiency. We address the computational complexity of optimizing operational decisions using a simple approximation method and demonstrate the potential benefits of joint optimization through a small-scale example.

Finally, we optimize the volume allocation policy using a sophisticated approximation algorithm capable of handling larger problem instances. A numerical study is conducted to evaluate the algorithm’s performance and demonstrate its potential to solve problem sizes matching the scale of practical applications.