Fuel consumption reduction in tuna purse seiners using oceanographic data and genetic algorithms

Abstract

Between 30% and 75% of the total operational costs of tuna vessels are fuel-related [1]. Both these costs, with their effect on fish prices and food security [2], and the impact on climate change of the greenhouse gases emitted during the vessels’ activity [3], make it necessary to find ways to increase fuel efficiency. The European Union’s Horizon 2020 SusTunTech project was born with the following goals in mind: to reduce the greenhouse gases emitted by the fishing vessels between 20% and 25%, to diminish the time they spend at sea and their fuel costs, and to increment their revenues and improve the economic and environmental sustainability of the tuna fishing industry, one of the most important in that sector. One way to reach these objectives is by providing skippers with an optimal fishing route planner which includes fuel oil consumption prediction. These routes are obtained combining different methodologies, such as machine learning, big data, artificial intelligence and –as outlined in this article– genetic algorithms, together with datasets collected from two tuna fishing vessels (by using sensors installed on them) and oceanographic data from Copernicus and EMODnet.