Authors:
B. Dhanush, R. Vaishnavi, J. Adeline Sneha, Vinothini Kasinathan, Raja Rajeswari Ponnusamy, Kohila Malar Kalesamy

Addresses:
Department of Artificial Intelligence, SRM Institute of Science and Technology, Ramapuram, Chennai, Tamil Nadu, India. School of Computing, Asia Pacific University of Technology and Innovation, Kuala Lumpur, Wilayah Persekutuan, Malaysia.

Abstract:

Vehicle maintenance and fault diagnosis remain largely inaccessible to ordinary owners, particularly in developing regions where professional diagnostic tools are either unavailable or prohibitively expensive. This paper introduces the Intelligent Vehicle Diagnosis System (IVDS), a lightweight, browser-based platform that integrates artificial intelligence, geolocation services, and collaborative community features into a single cohesive application available to any smartphone or computer user without specialized hardware. The system is built upon a curated knowledge base of 159 fault records distributed across three vehicle categories: motorcycles, cars, and trucks, covering more than fifty problem types per category. Users interact with the platform through a dual-mode interface: a structured dropdown menu for precise, menu-driven queries, and a natural language chat module powered by Python's difflib. SequenceMatcher algorithm is configured with a similarity threshold of 0.7. A GPS-driven workshop discovery feature queries the Overpass OpenStreetMap API within a 10 km radius and renders results on an interactive Leaflet map. Community capabilities include encrypted group garages, peer-to-peer direct messaging, and a public discussion forum. System evaluation on the full 159-record dataset yields an overall fault-match rate of 92.5%, a natural language processing accuracy of 89.8%, and an average query response latency of 22 milliseconds. These results establish the IVDS as a capable, computationally affordable alternative to hardware-dependent diagnostic systems, with clear pathways for integrating real-time OBD-II data streams and deep learning symptom analysis.

Keywords: Vehicle Diagnosis; Fuzzy Matching; Expert System; Natural Language Processing; Predictive Maintenance; Community Platform; Conditional Random Fields; Support Vector Machines (SVMs).

Received on: 13/07/2025, Revised on: 08/09/2025, Accepted on: 29/11/2025, Published on: 10/08/2026

DOI: 10.69888/FTSCS.2026.000723

FMDB Transactions on Sustainable Computing Systems, 2026 Vol. 4 No. 3, Pages: 210-224

• Views : 17
• Downloads : 6

Download PDF