Data-Driven Hull-Fouling Cleaning Schedule Optimization to Reduce Carbon Footprint of Vessels

TL;DR

This work tackles the problem of reducing maritime CO$_2$ emissions by optimally timing hull cleaning to mitigate fouling-induced drag. It develops a data-driven predictive framework for voyage fuel consumption and embeds it in a dynamic programming formulation (CLEANING-2) that leverages a cumulative fouling state to achieve polynomial-time optimization. Using a large, real-world dataset from tramp vessels, the approach identifies significant fuel savings—up to $5\%$ over four years—when cleaning is scheduled optimally, with several vessels benefiting from multiple cleaning events. The results demonstrate the practical potential for data-driven, operation-level hull maintenance to reconcile environmental targets with economic viability, and the open-source tooling enables deployment by ship operators and researchers alike.

Abstract

In response to climate change, the International Maritime Organization has introduced regulatory frameworks to reduce greenhouse gas emissions from international shipping. Compliance with these regulations is increasingly expected from individual shipping companies, compelling vessel operators to lower the CO2 emissions of their fleets while maintaining economic viability. An important step towards achieving this is performing regular hull and propeller cleaning; however, this entails significant costs. As a result, assessing whether ship performance has declined sufficiently to warrant cleaning from an environmental and economic standpoint is a critical task to ensure both long-term viability and regulatory compliance. In this paper, we address this challenge by proposing a novel data-driven dynamic programming approach to optimize vessel cleaning schedules by balancing both environmental and economic considerations. In numerical experiments, we demonstrate the usefulness of our proposed methodology based on real-world sensor data from ten tramp trading vessels. The results confirm that over a four-year period, fuel consumption can be reduced by up to 5%, even when accounting for the costs of one or two additional cleaning events.

Figures (7)

• Figure 1: One of the ten sister ships included in the study, photographed immediately after .
• Figure 2: Three examples (left to right vessels $V_1$, $V_3$ and $V_7$) where the influence of the independent variable DSDDM can be observed against the FOC through box and whisker plots.
• Figure 3: Visualization of the relative angle between and wind/current direction.
• Figure 4: The forward feature selection process visualized through the training and test error of the model for vessels $V_1$, $V_2$ and $V_3$.
• Figure 5: Six example dependence plots using the model trained on the vessel $V_7$ data. On the horizontal axes are six different input variables and on the vertical axes are corresponding values, which can be interpreted as the contribution of an individual feature value to the difference between the instance-specific prediction and the mean prediction. The coloring allows to check for interactions to a second input variable.