Data-Driven Machine Learning Approach for Vehicle Fuel Economy Prediction and Performance Monitoring Using Real-World OBD Data

Year : 2026 | Volume : 13 | Issue : 02 | Page : 47 66
By

Arumugam Palanichamy,

Suresh Alex Selvaraj,

Rajavel Rangasamy,

Karthikeyan Subramanian,

  1. Research Scholar, Department of Marine Engineering, AMET University, Chennai, Tamil Nadu, India
  2. Professor, Department of Marine Engineering, AMET University, Chennai, Tamil Nadu, India
  3. Principal, Sri Balaji Chockalingam Engineering College, A C S Nagar, Irumbedu, Arni, Tamil Nadu, India
  4. Sr.Manager, Fuel Cell Product Development, Ashok Leyland Technical Centre, Chennai, Tamil Nadu, India

Abstract

Modern passenger vehicles generate large volumes of operational data through On-Board Diagnostics (OBD) systems, enabling continuous observation of vehicle performance under real-world driving conditions. However, much of the existing research mainly analyses previously recorded data and does not provide predictive mechanisms for monitoring vehicle performance under dynamically varying operating conditions. This study presents an AI and machine learning–based method for predicting and monitoring real-world vehicle performance and fuel economy using OBD data collected from a gasoline-powered passenger car during normal driving operation. Key vehicle parameters including vehicle speed, engine speed, throttle position, engine load, and fuel consumption were acquired through the OBD interface and processed using a structured data analytics pipeline. Machine learning techniques including K-Means clustering, K-Nearest Neighbour (KNN) classification, and Random Forest regression were employed to identify representative operating modes, classify driving behaviour patterns, and predict expected fuel economy under varying driving conditions. The developed predictive method enables continuous comparison between predicted and observed vehicle performance, facilitating identification of operational deviations across different driving environments. Analysis of real-world driving data demonstrates the ability of the proposed approach to characterise vehicle operational patterns and support intelligent monitoring of vehicle performance and fuel economy. The study contributes a data-driven predictive monitoring methodology for real-world vehicle analytics, providing a foundation for AI-enabled vehicle performance assessment using OBD big data.

Keywords: Machine learning, on-board diagnostics (OBD), fuel economy prediction, vehicle operational data analytics, vehicle performance monitoring, big data analysis.

[This article belongs to Trends in Machine design ]

How to cite this article: Arumugam Palanichamy, Suresh Alex Selvaraj, Rajavel Rangasamy, Karthikeyan Subramanian. Data-Driven Machine Learning Approach for Vehicle Fuel Economy Prediction and Performance Monitoring Using Real-World OBD Data. Trends in Machine design. 2026; 13(02):47-66.
How to cite this URL: Arumugam Palanichamy, Suresh Alex Selvaraj, Rajavel Rangasamy, Karthikeyan Subramanian. Data-Driven Machine Learning Approach for Vehicle Fuel Economy Prediction and Performance Monitoring Using Real-World OBD Data. Trends in Machine design. 2026; 13(02):47-66. Available from: https://journals.stmjournals.com/tmd/article=2026/view=253858

References

  1. Rykała M, Nowakowski T, Szpytko J. Modeling vehicle fuel consumption using a low-cost OBD-II interface. Energies. 2023;16(21):7266. https://doi.org/10.3390/en16217266
  2. Michailidis ET, Panagiotopoulou A, Papadakis A. A review of OBD-II-based machine learning applications for sustainable, efficient, secure and safe vehicle driving. 2025;25(13):4057. https://doi.org/10.3390/s25134057
  3. Yoo S, Park J, Kim D. Machine learning based vehicle fuel efficiency prediction using big data analytics. Scientific Reports. 2025;15:96999. https://doi.org/10.1038/s41598-025-96999-0
  4. Abediasl H, Al-Samarraie H, Ghodsi A. Real-time vehicular fuel consumption estimation using on-board diagnostics data. Proc Inst Mech Eng Part D J Automobile 2024;238(5):845-857. https://doi.org/10.1177/09544070231185609
  5. Molina-Campoverde JJ, García A, Varga I. Driving pattern analysis and gear shift classification using OBD-II data. Sensors. 2025;25(13):4043. https://doi.org/10.3390/s25134043
  6. Keskin R, Arslan Bayesian LSTM based fuel consumption prediction using vehicle operational data. Sensors. 2025;25(22):7031. https://doi.org/10.3390/s25227031
  7. Ping P, Qin W, Xu Y, Miyajima C, Takeda Impact of driver behavior on fuel consumption: Classification and prediction using machine learning. IEEE Access. 2019;7:78515-78532. https://doi.org/10.1109/ACCESS.2019.292229 4
  8. Lattanzi E, Castellano N, Regini Machine learning techniques to identify unsafe driving behaviour using vehicle sensor data. Expert Systems with Applications. 2021;177:114952. https://doi.org/10.1016/j.eswa.2021.114952
  9. Yao Y, Zhao X, Liu Y. Vehicle fuel consumption prediction based on driving behavior data using machine learning. Mathematical Problems in Engineering. 2020;2020:9263605. https://doi.org/10.1155/2020/9263605
  10. Singh SK, Mishra R, Gupta P. Machine learning based classification of OBD-II data for driving behavior Bulletin of Electrical Engineering and Informatics. 2025;14(1):550-560. https://doi.org/10.11591/eei.v14i1.9398
  11. Fugiglando U, Massaro E, Santi Driving behavior analysis through CAN-Bus data in uncontrolled environments. IEEE Trans Intelligent Transportation   Systems. 2018;19(3):737-746. https://doi.org/10.1109/TITS.2017.2696351
  12. Malik M, Tiwari S. Driving pattern analysis using OBD-II data and machine learning Materials Today Proceedings. 2023;72:3215-3221. https://doi.org/10.1016/j.matpr.2022.09.340
  13. Cao Z, Wang J, Liu Y. Identification of high-emission vehicles using on-board diagnostics data analytics. Environmental 2025;349:122084. https://doi.org/10.1016/j.envpol.2024.122084
  14. Khan MAA, Ali MH, Haque AF. A machine learning approach for driver identification based on CAN-BUS sensor data. IEEE 2022;10:80341-80351. https://doi.org/10.1109/ACCESS.2022.319558 3
  15. Dong W, Li J, Yao R. Characterizing driving styles with deep learning. IEEE Trans Intelligent Transportation Systems. 2019;20(7):2510-2520. https://doi.org/10.1109/TITS.2018.2866052
  16. Singh S, Kumar Driving pattern analysis using real-time ECU data collected through OBD-II. Int J Electrical and Computer Engineering Systems. 2022;13(2):189-198. https://doi.org/10.32985/ijeces.13.2.5
  17. Chen Z, Wang Y. Data-driven vehicle energy consumption prediction using machine learning models. Applied Energy. 2019;235:701-712. https://doi.org/10.1016/j.apenergy.2018.10.094
  18. Wang H, Zhang J. Real-time vehicle monitoring using OBD-II and IoT based analytics. IEEE Access. 2020;8:203456- https://doi.org/10.1109/ACCESS.2020.303509 7
  19. Zhang Y, Wang L. Vehicle fuel consumption prediction using random forest regression Energy. 2019;181:888-897. https://doi.org/10.1016/j.energy.2019.05.090
  20. Li X, Liu Z. Machine learning based vehicle performance evaluation using big data analytics. Transportation Research Part C. 2020;115:102621. https://doi.org/10.1016/j.trc.2020.102621
  21. Park S, Kim H. Data driven vehicle diagnostics using machine learning IEEE Trans Intelligent Transportation Systems. 2022;23(8):11321-11331. https://doi.org/10.1109/TITS.2021.3106024
  22. Huang J, Li Vehicle anomaly detection using OBD-II sensor data and machine learning. IEEE Access. 2021;9:75620-75632. https://doi.org/10.1109/ACCESS.2021.308164 2
  23. Chen Y, Zhang H. Intelligent vehicle monitoring using OBD data and cloud analytics. Future Generation Computer 2021;115:52-60. https://doi.org/10.1016/j.future.2020.08.019
  24. Wang Z, Chen H. Big data analytics for vehicle performance monitoring. Transportation Research Part 2018;95:451-463. https://doi.org/10.1016/j.trc.2018.07.012
  25. Liu Y, Sun Machine learning based fuel efficiency estimation using vehicle operational data.Applied Energy 2021;298:117223. https://doi.org/10.1016/j.apenergy.2021.11722 3
  26. Barbado A, Navarro C. Interpretable machine learning models for vehicle fuel consumption prediction. Engineering Applications of Artificial Intelligence. 2022;113:104908. https://doi.org/10.1016/j.engappai.2022.10490 8
  27. Shirole V, Kulkarni P. Data driven driver behavior analysis using OBD-II sensor data. Journal of Intelligent Transportation Systems. 2025;29(3):245-259. https://doi.org/10.1080/15472450.2024.23087 12
  28. Singh R, Patel D. Driving behavior classification using OBD-II speed and acceleration signals. Advances in Science Technology and Engineering Systems 2021;6(1):512-519. https://doi.org/10.25046/aj060160
  29. Haghshenas SS, Chen Fuel consumption and CO₂ emissions prediction using vehicle telematics data. Cleaner Engineering and Technology. 2025;19:100742. https://doi.org/10.1016/j.clet.2024.100742
  30. Selvam HP, Jayaprakash B, Li Physics- informed machine learning for predicting engine emissions using OBD data. IEEE Access. 2025;13:22345-22358. https://doi.org/10.1109/ACCESS.2025.335112 0
Regular Issue Subscription Original Research
Volume 13
Issue 02
Received 20/05/2026
Accepted 29/05/2026
Published 15/06/2026
Publication Time 26 Days

Login

My IP

PlumX Metrics

Support