Published Paper Details:

Failed engines in commercial vehicles result in loss of resources, time and money for the fleet operators and safety concerns for vehicle operators. Since the first observations in the 1900s, low engine oil pressure has been the most frequent cause of engine failure [1]. With time, indicators and the sensors have been added to check oil levels and communicate abnormal engine oil pressure to the vehicle operators via the engine controls units. With the advent of advance sensors in the 2100s, this data began to be shared on central servers for tracking and maintaining vehicles remotely. Using data analytic and model-based predictions have attempted to make necessary interventions to reduce downtime by conducting preventive maintenance. In this research, the implementation begins with rule-based labeling time series-data from lab tests and field observations. Next, to apply a model-based approach, parameters like engine oil pressure, engine speed, vehicle speed, and engine coolant temperature were identified as parameters indicating engine status. We attempt to use Different Machine learning models to classify the engine failed vehicle due to engine oil pressure.