Component maintenance prediction system with behavior modeling

A method for managing a maintenance for a component in a vehicle, the method comprising:

• receiving sensor data for the vehicle;
• predicting target parameters using behavior machine learning models trained using first training data to predict the target parameters for a normal behavior of the component operating in a tolerance, wherein the target parameters characterize behavior of the component;
• determining prediction metrics from predicted values for the target parameters predicted by behavior machine learning models and actual values for the target parameters;
• and
• predicting whether the component will fall out of the tolerance after a time period using the prediction metrics and a maintenance machine learning model trained using second training data to predict whether the maintenance is needed for the component, wherein the second training data comprises historical prediction metrics determined for the target parameters.

A vehicle management system comprising:

• a computer system;
• and
• a maintenance manager in the computer system, wherein the maintenance manager is configured to:
• train behavior machine learning models using first training data to output predicted values for target parameters for a normal behavior of a component operating in a tolerance, wherein each behavior machine learning model in the behavior machine learning models predicts a target parameter in the target parameters for the component;
• determine historical prediction metrics from the predicted values for the target parameters predicted by behavior machine learning models in response to receiving historical sensor data and actual values for the target parameters for the component;
• train a maintenance machine learning model using second training data to predict whether a maintenance is needed for the component, wherein the second training data comprises the historical prediction metrics determined for the target parameters, wherein the maintenance machine learning model outputs a prediction as to whether the component will become out of the tolerance after a time period;
• and
• determine whether the maintenance is needed for the component using sensor data for the component, prediction metrics determined from the predicted values for the target parameters output by the behavior machine learning models and the actual values for the target parameters, and the maintenance machine learning model.

A vehicle management system comprising:

• a computer system;
• and
• a maintenance manager in the computer system, wherein the maintenance manager is configured to:
• receive sensor data for a vehicle;
• predict target parameters using behavior machine learning models trained using first training data to predict the target parameters for a normal behavior of a component operating in a tolerance, wherein the target parameters characterize behavior of the component;
• determine prediction metrics from predicted values for the target parameters predicted by the behavior machine learning models and actual values for the target parameters;
• and
• predict whether the component will fall out of the tolerance after a time period using the prediction metrics and a maintenance machine learning model trained using second training data to predict whether a maintenance is needed for the component, wherein the second training data comprises historical prediction metrics determined for the target parameters and the actual values for the target parameters.