A SMART FAULT DETECTION SYSTEM USING FUZZY LOGIC TECHNOLOGY

Abstract

In the present era, automobiles have become an integral aspect of individuals' daily
lives. Engine malfunctions can prime to noteworthy issues for regulars if not
distinguished early, punctually addressed, and truthfully repaired. Such letdowns
may pretense risks to existence and property, negatively impacting customer
pleasure and the status of vehicle firms. This novel approach leverages the flexible
properties of fuzzy logic, which are strategically applied to strengthen and improve
the operational effectiveness, safety, and dependability of automotive systems. To
ensure that the prototypical is skillful of handling a wide range of complex and
varied automotive data, ML.NET was used to train the dataset that was obtained
from the Kaggle Dataset repository. Impressively, this forward-thinking system
was developed utilizing an extensive array of state-of-the-art web machineries,
including Bootstrap 3.5, JavaScript, ASP.Net, CSS and JQuery, and SQL server,
attesting to its commitment to technological advancement and innovation.
Rigorous testing and meticulous evaluation have yielded promising outcomes,
showcasing the system's potential for widespread adoption while demonstrating
its prowess in averting accidents, curbing maintenance costs, and significantly
enhancing the overall driving experience. Achieving a commendable performance
accuracy of 73.14%, alongside a precision rate of 100% and an F1 Score peaking
at 76.62%, this visionary system stands at the forefront of transformative progress
in automotive fault detection, promising a paradigm shift in vehicular safety
protocols and maintenance standards