Fuel Consumption of Offshore Support Vessels using Machine Learning Models

Fuel is the biggest expense for shipping companies, having a direct impact on the cost of transported products and consequently on the economy. At the same time, there is a worldwide concern about the level of pollutant emissions, which is directly related to the burning of fossil fuels. The reduction in the level of CO2, NOx and SOx emissions is part of the new requirements established by the Marine Environment Protection Committee (MEPC) of the IMO (International Maritime Organization). The decrease in these levels is directly related to the decrease in fuel consumption, and this translates into energy efficiency. According to the IMO, a series of measures can be adopted to reduce consumption on ships: Fuel-efficient operations; Optimized operation of the ship; Hull and thruster optimization; Optimization of machinery and equipment; Optimization of cargo handling; Conservation and conscientious use of energy. Within efficient fuel consumption operations, there are two points that can impact consumption efficiency: speed optimization and optimization of the power provided by the engines.

The class of offshore platform support vessels (PSV) is of great importance, due to its large number, as well as the great frequency of trips in the southeastern region of Brazil. This work aims to present the results of the application of artificial intelligence tools, such as machine learning, to a database containing information on consumption, speed, environmental conditions, distance, duration, etc