Reliability Analytics performs reliability analysis of failure, maintenance, and usage data in industrial asset fleet. It is exracts full value from serial number data of reparable parts. The analytics is available as packaged AI/ML software application, Reliability Analytics Engine. It is also also available as a service, where customer provides data sets (database query results) and receives human readable analytical reports on periodic basis.
Reliability Analytics Engine is a packaged AI/ML software application that performs reliability analysis or failures of industrial asset fleet. It builds a sophisticated Reliability Digital Twin (RDT) model by analyzing data from maintenance operation (e.g., removals, repairs, and replacements of the parts) and usage data (e.g., usage time, usage missions, usage mileage, usage hours) to build reliability model across fleet. Reliability Analytics Engine provides predictive trending of fleet reliability by showing statistically anomalous time periods (months or quarters) where number of failures cannot be explained by part reliability, part population ages, and fleet usage intensity. Predictive trend monitoring supports executive decisions by reporting emerging fleet-wide reliability problems as they appear. Reliability Analytics Engine also provides root cause analysis of bad reliability by identifying “Bad Actors” that drag down the reliability of the fleet. Bad Actors include reparable parts that keep failing too often, in statically significant way. Bad Actors also include equipment hosts were parts keep failing too often (e.g., because of bad power supply). The Bad Actor lists allow industrial companies to focus resources on properly fixing Bad Actors or replacing them to improve the overall reliability of the fleet.