Modeling Vehicle Fuel Consumption Using a Low-Cost OBD-II Interface

Magdalena Rykała (), Małgorzata Grzelak, Łukasz Rykała, Daniela Voicu and Ramona-Monica Stoica
Additional contact information
Magdalena Rykała: Faculty of Security, Logistics and Management, Military University of Technology, 00-908 Warsaw, Poland
Małgorzata Grzelak: Faculty of Security, Logistics and Management, Military University of Technology, 00-908 Warsaw, Poland
Łukasz Rykała: Faculty of Mechanical Engineering, Military University of Technology, 00-908 Warsaw, Poland
Daniela Voicu: Faculty of Aircraft and Military Vehicles, Military Technical Academy “Ferdinand I”, 050141 Bucharest, Romania
Ramona-Monica Stoica: Faculty of Aircraft and Military Vehicles, Military Technical Academy “Ferdinand I”, 050141 Bucharest, Romania

Energies, 2023, vol. 16, issue 21, 1-23

Abstract: As a result of ever-growing energy demands, motor vehicles are among the largest contributors to overall energy consumption. This has led researchers to focus on fuel consumption, which has important implications for the environment, the economy, and geopolitical stability. This article presents a comprehensive analysis of various fuel consumption modeling methods, with the aim of identifying parameters that significantly influence fuel consumption. The scientific novelty of this article lies in its use of low-cost technology, i.e., an OBD-II interface paired with a mobile phone, combined with modern mathematical modeling methods to create an accurate model of the fuel consumption of a vehicle. A vehicle test drive was performed, during which variations in selected parameters were recorded. Based on the obtained data, a model of the vehicle’s fuel consumption was built using three forecasting methods: a multivariate regression model, decision trees, and neural networks. The results show that the multivariate regression model obtained the lowest MSE, MAR, and MRSE coefficients, indicating that this was the best forecasting method among those tested. Sufficient forecast error results were obtained using neural networks, with increases of approximately 73%, 10%, and 131% in MSE, MAE, and MRAE, respectively, compared to regression results. The worst results were obtained with the decision tree model, with increases of approximately 163%, 21%, and 92% in MSE, MAE, and MRAE compared to the regression results.

Keywords: fuel consumption; OBD-II; mobile phone; GPS; signal processing; regression; decision trees; neural networks (search for similar items in EconPapers)
JEL-codes: Q Q0 Q4 Q40 Q41 Q42 Q43 Q47 Q48 Q49 (search for similar items in EconPapers)
Date: 2023
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.mdpi.com/1996-1073/16/21/7266/pdf (application/pdf)
https://www.mdpi.com/1996-1073/16/21/7266/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jeners:v:16:y:2023:i:21:p:7266-:d:1267793

Access Statistics for this article

Energies is currently edited by Ms. Agatha Cao

More articles in Energies from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().