Authors: Muhammad Fawad Nasim 1, Muhammad Anwar 2, Almuhannad S. Alorfi 3, Hamza Awad Ibrahim 4, Ali Ahmed 5, *, Arfan Jaffar 1, 6, Sheeraz Akram 1, 6, 7, Abubakar Siddique 8, Hafiz Muhammad Zeeshan 9
Affiliations:
1Faculty of Computer Science and Information Technology, The Superior University, Lahore, Pakistan
2Department of Information Sciences, Division of Science and Technology, University of Education, Lahore, Pakistan
3Department of Information Systems, Faculty of Computing and Information Technology, King Abdulaziz University, Rabigh, Saudi Arabia
4Department of Computer College of Engineering and Computers, Al-Gunfudah Umm-Al Qura University, Makkah, Saudi Arabia
5Faculty of Computing and Information Technology, King Abdulaziz University, Rabigh, Saudi Arabia
6Intelligent Data Visual Computing Research (IDVCR), Lahore, Pakistan
7Information Systems Department, College of Computer and Information Sciences, Imam Mohammad Ibn Saud Islamic University, Riyadh, Saudi Arabia
8Wellington Institute of Technology, Te Pukenga, Wellington, New Zealand
9Department of Computer Science, National College of Business Administration and Economics, Lahore, Pakistan
Recently, there have been efforts to create automated systems for diagnosing engine problems using sound detection. However, most of these methods lack robustness and interpretability, functioning as "black boxes" that make it difficult to understand their decision-making processes. The Learning Classifier System (LCS), a machine learning approach that operates using a set of rules, has demonstrated potential for providing robust, interpretable, and generalizable solutions across different domains. This work aims to develop a new LCS-based system for automatically detecting engine problems, with a focus on making its decision-making process understandable, contributing to explainable artificial intelligence. The system's performance is evaluated using features from the time domain, frequency domain, and time-frequency domain. Its robustness is tested with noisy sound data gathered under various normal and abnormal conditions. Experimental results show that this new approach outperforms conventional state-of-the-art methods by 2.6%−6.0%, achieving a maximum performance accuracy of 98.6%.
Audio classification, Acoustic features classification, Engine sound classification, Learning classifier system, Rule-based systems
https://doi.org/10.21833/ijaas.2025.08.001
Nasim, M. F., Anwar, M., Alorfi, A. S., Ibrahim, H. A., Ahmed, A., Jaffar, A., Akram, S., Siddique, A., & Zeeshan, H. M. (2025). Cognitively inspired sound-based automobile problem detection: A step toward explainable AI (XAI). International Journal of Advanced and Applied Sciences, 12(8), 1–15. https://doi.org/10.21833/ijaas.2025.08.001