ENHANCING EQUIPMENT FAILURE PREDICTION USING AI TECHNIQUE IN INDUSTRIAL PREDICTIVE MAINTENANCE

In terms of repair costs and production downtime, gearbox and machinery failures are
expensive and frequently lead to whole plant failures. By using Artificial Neural Networks (ANN) to forecast machinery breakdowns, this study sought to reduce these losses by facilitating the application of predictive maintenance techniques. This strategy offers major operational gains by increasing machine dependability and attaining large cost savings. An ANN model that had undergone extensive training, validation, and testing procedures was used to gather and analyse sensor data. The outcomes showed a remarkable 97.3% prediction accuracy. The Mean Time Between Failures (MTBF) increased by 3,000 hours and the Mean Time to Repair (MTTR) decreased by 6.5 hours, two significant increases in key performance parameters. Predictive maintenance also resulted in 95 million Naira in cost savings. This study’s use of AI approaches greatly increased the accuracy of failure prediction, enabling better predictive maintenance. This led to increased machine dependability and significant cost savings, highlighting the importance of
incorporating cutting-edge AI techniques into industrial maintenance procedures