Abstract: A method and system for cross-verification of alarms in real-time comprising identifying primary variables and secondary variables causing the event, labelling the primary variables and the secondary variables by primary engine and secondary engine based on Artificial Intelligence based predictive model building, predicting the labels by one or more inference engine based on previous history and data patterns, triggering secondary engine for cross-verification of alarms whenever there is a prediction from the primary engine, identifying the correlation between the labels from the primary engine and the secondary engine by validation engine, identifying alarm type based on correlation, recommending predictive maintenance and displaying on dashboard the cross-verification status of the alarms. The method reduces misclassification of alarm types based on predictive or preventive maintenance, reduces the maintenance costs of assets, and helps in prioritizing the critical alarms based on the alert type.

Abstract: A method and system for predictive maintenance of a machinery asset is provided. Data is collected from the sensors in the machine, and normalized for dealing with anomalies without removing them but passing onto the data transformation stage. Then the data is transformed and categorized based on multiple comparisons and calculations using the number of machine parts, and the operating values of the machine provided by the manufacturer. The categorized data is then used to predict the maintenance and recommended action for the machine part.