Orthogonal LS-PLS approach to ship fuel-speed curves for supporting decisions based on operational data

The shipping industry relies on ship fuel-speed curves to describe the fuel consumption (and CO2 emissions levels) per hour as a function only of the vessel’s speed over ground, based on dedicated test data. However, they are affected by additional factors in real cases. In this article, a novel method is developed elaborating the orthogonal least-squares partial least-squares (LS-PLS) approach to enhance fuel-speed curves accuracy when information is available on additional factors from multi-sensor systems. Through real data examples, the approach is shown capable of detecting anomalies in CO2 emission levels and testing the effectiveness of ship energy efficiency initiatives.