Title: Predictive model using machine learning to determine fuel consumption in CAT-777F mining equipment

Authors: Marco Cotrina; Jairo Marquina; Jose Mamani; Solio Arango; Joe Gonzalez; Johnny Ccatamayo; Eduardo Noriega

Addresses: Department of Mining Engineering, Universidad Nacional de Trujillo, Trujillo, 13001, Peru ' Department of Mining Engineering, Universidad Nacional de Trujillo, Trujillo, 13001, Peru ' Department of Chemical Engineering Universidad Nacional del Altiplano de Puno, Puno, 21001, Peru ' Department of Mining Engineering, Universidad Nacional de Trujillo, Trujillo, 13001, Peru ' Department of Industrial Engineering, Universidad Nacional de Trujillo, Trujillo, 13001, Peru ' Department of Mining Engineering, Universidad Nacional San Cristobal de Huamanga, Ayacucho, 05000, Peru ' Department of Mining Engineering, Universidad Nacional de Trujillo, Trujillo, 13001, Peru

Abstract: The main objective of the research was to determine fuel consumption in CAT-777F equipment considering parameters such as speed, resistance and payload of the equipment, using a machine learning neural network algorithm. The methodology used was exploratory level, quantitative approach, applied type and non-experimental cross-sectional design. The results obtained when performing the EDA showed a high relationship of 84% between the 'payload' of the equipment and the fuel consumption, indicating that keeping the payload of the truck constant, the fuel consumption will be lower. When validating the model, the result of the ANN metric indicates that the coefficient of determination (R²) obtained a value greater than or equal to 68%. Likewise, using the random forest regressor algorithm to determine the percentage of incidence in each of the parameters, it was obtained that the payload has a percentage of incidence of 87%. Finally, it is concluded that the use of neural networks can predict fuel consumption in CAT-777F equipment.

Keywords: fuel consumption; neural networks; payload; resistance; truck speed.

DOI: 10.1504/IJMME.2024.140073

International Journal of Mining and Mineral Engineering, 2024 Vol.15 No.2, pp.147 - 160

Received: 21 Dec 2023
Accepted: 10 May 2024
Published online: 19 Jul 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article