US 11,232,368 B2 · System for predicting equipment failure events and optimizing manufacturing operations

Abstract

A system receives sensor data from sensing parameters of a piece of factory equipment. The system includes a first model to generate predicted degradation states of the piece of factory equipment by being trained to generate a stochastic degradation model for classification of the predicted degradation states of a particular asset. The system includes a second model to which the predicted degradation states are provided. The second model trained to generate a covariate indicative of a failure condition of the piece of factory equipment. The system may supply the covariate to the first model to generate predicted degradation states compensated with the covariate. From the predicted degradation states compensated with the covariate a policy of a maintenance action may be generated with the system to optimize life expectancy of the piece of factory equipment. The system may adjust operation of the piece of factory equipment based on the maintenance action.

Claims20

Dependent claims

A system for predicting equipment failure events and optimizing manufacturing operations, the system comprising:

an interface for receiving sensor data respectively associated with one or more assets;
a processor in communication with the interface;
and
non-transitory computer readable media in communication with the processor that stores instruction code, which when executed by the processor, causes the processor to:
input the sensor data into a first model trained to generate a stochastic degradation model for classification of a current degradation state of a particular asset;
input the current degradation state of the particular asset into a second model trained to infer a root cause for a failure of the asset based on the stochastic degradation model, wherein the second model provides a prediction of a future root cause for failure of the particular asset in a form of a covariant, and the first model provides a compensated predicted degradation state based on the covariant;
input the sensor data into a reinforcement learning engine configured to generate a maintenance schedule based at least in part on the stochastic degradation model of the particular asset;
and
communicate the maintenance schedule to an entity associated with the particular asset or the particular asset.

6 more elements

A method comprising:

receiving a plurality of sensor data from a plurality of sensing parameters of a piece of factory equipment;
generating predicted degradation states of the piece of factory equipment, with a first model trained to generate a stochastic degradation model for classification of the predicted degradation states of a particular asset;
fitting the predicted degradation states to a second model trained to generate a covariate indicative of a failure condition of the piece of factory equipment;
supplying the covariate to the first model to generate predicted degradation states compensated with the covariate;
generating a policy of a maintenance action based on the predicted degradation states to optimize life expectancy of the piece of factory equipment;
generating a maintenance/operation schedule based on the policy;
and adjusting operation of the piece of factory equipment based on the maintenance/operation schedule.

4 more elements

A non-transitory computer readable medium configured to store instructions executable by a processor, the computer readable medium comprising:

instructions executable with the processor to receive a plurality of sensor data from a plurality of sensing parameters of a piece of factory equipment;
instructions executable with the processor to generate predicted degradation states of the piece of factory equipment, with a first model trained to generate a stochastic degradation model for classification of the predicted degradation states of a particular asset;
instructions executable with the processor to fit the predicted degradation states to a second model trained to generate a covariate indicative of a failure condition of the piece of factory equipment;
instructions executable with the processor to supply the covariate to the first model to generate predicted degradation states compensated with the covariate;
instructions executable with the processor to generate a policy of a maintenance action based on the predicted degradation states to optimize life expectancy of the piece of factory equipment;
instructions executable with the processor to generate a maintenance/operation schedule based on the policy;
and
instructions executable with the processor to adjust operation of the piece of factory equipment based on the maintenance/operation schedule.

5 more elements

Dependencies are inferred from the claim text.

System for predicting equipment failure events and optimizing manufacturing operations

Life of this patent

System for predicting equipment failure events and optimizing manufacturing operations

Life of this patent

Abstract

Claims20