Brake Maintenance Diagnostic with Fuzzy-Bayesian Expert System

Brakes, one of a vehicle’s most crucial safety systems, are necessary to ensure safety. They are our primary protection mechanism while driving a car on the road. A brake failure can end up causing an accident and putting lives at risk, which is why it is essential to check all its elements periodically. The car must go to service in case of brake issues: unusual noises, abnormal movements or sensations, inability to stop quickly, and warning lights. Sometimes, drivers do not associate them with brake failure and wait to take the car to check service. However, if they could determine that an issue relates to brake problems, they could immediately seek assistance. State of the art shows that expert systems and domain expertise revolutionize maintenance, reshaping diagnostics, decision-making, and predictive strategies by blending advanced AI techniques, data analysis, and real-time monitoring. On the one hand, fuzzy logic is a branch of artificial intelligence and mathematics used to model and manage uncertainty and imprecision in data and expert systems. On the other hand, Bayesian reasoning allows determining beliefs about a hypothesis based on facts. In this work, we propose developing a Fuzzy-Bayesian expert system for assisting the drivers in the maintenance of car brake systems encompassing goal setting, knowledge acquisition, interface design, and testing. Our proposal, programmed in Python, uses UPAFuzzySystems to describe fuzzy rules and Twilio to allow SMS integration in a user interface, empowering users to make informed brake system decisions from their mobile and obtain information about the status of their vehicle’s brake system.