In this presentation, we will recount our journey collaborating with the maritime industry, discussing the challenges we encountered and elucidating how we transformed these challenges into gratifying experiences and impactful contributions. We will use our work on bunker procurement decisions with a global container shipping company as an example to illustrate the impact we made and the lessons we learned.

Bunker refueling decisions in international shipping are crucial operational choices. Each ship acts like a movable storage unit navigating through diverse markets, procuring bunker fuels from different ports to sustain its voyage. This involves grappling with challenges posed by varying bunker fuel prices over time and locations. To tackle this challenge, we propose data-driven structure-prescriptive (SP) approaches that combine the strengths of modern machine learning with the insights from traditional OR modeling and optimization. Instead of predicting future marine fuel prices, our approach directly learns the optimal refueling policy from data and adapts refueling decisions to the current market conditions, including fuel prices, crude oil price, NYSE index, etc.

Our focus lies in leveraging the well-established understanding that the optimal refueling decision adheres to a state-dependent base-stock refueling policy. This decision depends on factors such as the port of call, fuel tank capacity, market conditions, and is finite-valued, depending on the vessel’s schedule and voyage. We provide a practical framework to incorporate these structural properties into data-driven decision-making for bunker refueling operations. The proposed SP approaches successfully recovered the “true” optimal refueling policy in synthetic simulations. Moreover, our experiments unveiled that incorporating more structural properties into the learning process significantly improved the out-of-sample (OOS) performance. In the case study, we compared our proposed SP approach with the firm’s existing operation, resulting in a noteworthy reduction of fuel expenses, which amounts to approximately 2.52 million USD per year in savings for a fleet of six ships.

In addition, to facilitate our collaboration with industry, we propose an eXplainable multi-stage bunker procurement planning (X-BPP) framework for the maritime industry. In this presentation, we will showcase this framework, discuss its performance, and share the lessons we learn in implementing the system.