Water navigation is crucial for the movement of people and goods in many locations, including the Amazon region. It is essential for the flow of inputs and outputs, and for certain Amazon cities, boat access is the only option. Fuel consumption accounts for over 25% of a vessel’s total operational costs. Shipping companies are therefore seeking procedures and technologies to reduce energy consumption. This research aimed to develop a fuel consumption prediction model for vessels operating in the Amazon region. Machine learning techniques such as Decision Tree, Random Forest, Extra Tree, Gradient Boosting, Extreme Gradient Boosting, and CatBoost can be used for this purpose. The input variables were based on the main design characteristics of the vessels, such as length and draft. Through metrics like mean, median, and coefficient of determination (R²), six different algorithms were assessed. CatBoost was identified as the model with the best performance and suitability for the data. Indeed, it achieved an R² value higher than 91% in predicting and optimizing fuel consumption for vessels operating in the Amazon and similar regions.

La Houille Blanche - Revue internationale de l eau, est une publication de la Société Hydrotechnique de France

Waterway transport is of great importance for the population of the Amazon Region. It is responsible for moving, according to Agência Nacional de Transportes Aquaviários (Antaq, 2018), approximately 9.8 million passengers and 3.4 million tons of cargo distributed by state, interstate longitudinal transport interstate and crossing. However, despite the great importance, there are several blockages related to the conditions of the roads and vessels in certain routes. The study in question proposes the use of a mathematical tool to analyze vessels from the relative efficiency involving parameters of input and output for two analysis scenarios: Energy Efficiency Analysis and Modality Analysis, being analyzed variables such as power, consumption, total time of the trip, total distance of the line of action, passenger capacity and ticket price. For that, the data of 652 vessels used in the Amazon region registered in the Brazilian states (Pará, Amazonas, Amapá and Rondônia) were verified, and from these data, a sample space of 148 vessels from the same utilization level was selected. With these data, a model was developed using the Data Envelopment Analysis (DEA) methodology, with the aid of the DEA-Solver software. After the use of the solver, the relative efficiency indexes of each vessel were obtained, allowing to rank and measure the efficient and inefficient Decision Units (DMU'S). Lastly, analysis of the general characteristics of efficient vessels were made in order to stratify the parameters that may have transformed them into efficient DMU`s, making it possible to trace a profile of the type of vessel employed in each type of line, thus helping in decision making of new projects.