USE OF THE KNN ALGORITHM TO PREDICT FUEL CONSUMPTION IN CARGO TRANSPORT VEHICLES USED IN MINING

• DOI: 10.22533/at.ed.317362316021

• Palavras-chave: KNN; Dispatch; Mining; Regression.

• Keywords: KNN; Dispatch; Mining; Regression.

• Abstract:

  The paper describes the use of the K-Nearest Neighbor (KNN) algorithm, applied in programming language, with the objective of predicting fuel consumption in cargo transport vehicles used in mining. For this development, the data available in a specific dispatch system for this type of application was used. Mining is an important sector of the Brazilian economy, as it operates by supplying raw materials to various areas of the industry. Among the operating costs of companies in the sector, an important portion goes to fuel consumption, this happens due to the constant need to move material, often over long distances. Tools that allow a prediction of fuel consumption value can contribute both to better planning and to the optimization of the vehicle allocation process. This fact becomes even more significant when most companies already have systems that provide the necessary information for this implementation, but they do not use them.