Predictive modelling

Patent metadata

Field	Value
Publication number	US-10928817-B2
Application number	US-201715838122-A
Country	US
Kind code	B2
Filing date	Dec 11, 2017
Priority date	Dec 19, 2016
Publication date	Feb 23, 2021
Grant date	Feb 23, 2021

Abstract

Systems, methods, and non-transitory computer-readable media can be configured to perform receiving a notification of a maintenance event associated with a resource. The method includes retrieving historic maintenance data in relation to the resource with which the fault is associated, the maintenance information originating from a time period preceding the time of the maintenance event. The method includes identifying at least a portion of the retrieved historic maintenance data as being indicative of the maintenance event. The method also includes causing the portion of the retrieved historic maintenance data identified as being indicative of the maintenance event to be stored as a precursor signal of the maintenance event. The method also includes causing future maintenance data received from a plurality of resources related to the resource with which the maintenance event is associated to be monitored to predict a future occurrence of the maintenance event in the plurality of resources.

First claim

The invention claimed is:

1. A computer-implemented method of predicting maintenance events with respect to resources, the method being performed by one or more processors or special-purpose computing hardware, the method comprising:
diagnosing a fault in a first resource, comprising:
determining a parameter curve indicating how a first parameter measured by a first sensor changes with respect to a second parameter measured by a second sensor;
in response to detecting that the determined parameter curve deviates by more than a threshold amount from an average parameter curve indicating an expected relationship between the first parameter and the second parameter, determining that a fault of a type related to at least one of the first parameter or the second parameter is developing in the machine;
receiving a notification of a maintenance event associated with the diagnosing of a fault of the first resource of first interrelated resources of a first system of a first type, the first type being based at least partially on the first interrelated resources;
retrieving, in response to receiving the notification, historic maintenance data in relation to the first resource with which the fault is associated, the historic maintenance data originating from a time period preceding a time of the maintenance event;
causing a portion of the retrieved historic maintenance data identified as being indicative of the maintenance event to be stored as one or more precursor signals of the maintenance event; and
causing future maintenance data received from second resources, each of the second resources comprising a particular resource of respective second interrelated resources of a corresponding second system, each of the corresponding second systems being of a respective second type different from the first type, and each of the respective second types being based at least partially on the respective second interrelated resources of the corresponding second system, to be monitored to predict, based on the one or more precursor signals of the maintenance event associated with the fault of the first resource, a future occurrence of the maintenance event in one of the second resources, the predicting comprising comparing the future maintenance data in one of the second resources with the one or more precursor signals to determine reductions in probabilities that the maintenance event will occur in response to one or more particular maintenance tasks.

2. The method of claim 1 , further comprising:
providing a database storing historic maintenance data associated with the first interrelated resources related to the first resource with which the maintenance event is associated, and
comparing the retrieved maintenance data of the first resource with which the maintenance event is associated with the stored historic maintenance data of the first interrelated resources.

3. The method of claim 2 , wherein comparing the retrieved maintenance data with the stored historic maintenance data of the first interrelated resources comprises performing a dynamic time warping operation with data retrieved from one or more sensor logs.

4. The method of claim 3 , wherein the dynamic time warping operation comprises generating warped curves of a parameter indicating an internal function or internal status of at least one of the plurality of resources based on a warped time-frame established using curves of a second parameter indicating an external state of the at least one of the plurality of resources.

5. The method of claim 1 , wherein the notification contains an indication of a sub-system with which the maintenance event is associated.

6. The method of claim 5 , wherein monitoring future maintenance data comprises monitoring maintenance data from the sub-system with which the maintenance event is associated.

7. The method of claim 5 , wherein monitoring future maintenance data comprises monitoring maintenance data from a sub-system related to the sub-system with which the maintenance event is associated.

8. The method of claim 1 , wherein the maintenance data is obtained from at least one of: sensor logs, fault logs, or maintenance logs.

9. The method of claim 1 , wherein monitoring future maintenance data comprises calculating a probability that a maintenance event will occur in a future time period.

10. The method of claim 1 , wherein identifying at least a portion of the retrieved historic maintenance data as being indicative of the maintenance event comprises identifying a cluster of warning messages associated with the maintenance event.

11. The method of claim 1 , wherein the maintenance event comprises a decrease in efficiency or a collision.

12. The method of claim 1 , wherein the causing future maintenance data to be monitored to predict a future occurrence of the maintenance event in at least one of the plurality of second resources is further based on a number of times a measured parameter departs from a respective operational tolerance of the at least one of the second resources within a specific period of time.

13. The method of claim 1 , further comprising:
displaying, on an interface, the retrieved historic maintenance data comprising, for each fault, an object, an identifier, a date and a duration of the fault.

14. The method of claim 1 , wherein the detecting that the determined parameter curve deviates by more than the threshold amount from the average parameter curve further comprises determining a number and a duration of intervals during which the determined parameter curve deviates by more than the threshold amount from the average parameter curve, and the method further comprises:
recalculating the one or more precursor signals based on the future maintenance data of the one of the second resources in response to the future occurrence of the maintenance event in the one of the second resources.

15. A non-transitory computer-readable storage medium including instructions that, when executed by at least one processor of a computing system, cause the computing system to perform a method comprising:
diagnosing a fault in a first resource, comprising:
determining a parameter curve indicating how a first parameter measured by a first sensor changes with respect to a second parameter measured by a second sensor;
in response to detecting that the determined parameter curve deviates by more than a threshold amount from an average parameter curve indicating an expected relationship between the first parameter and the second parameter, determining that a fault of a type related to at least one of the first parameter or the second parameter is developing in the machine;
receiving a notification of a maintenance event associated with the diagnosing of a fault of the first resource of first interrelated resources of a first system of a first type, the first type being based at least partially on the first interrelated resources;
retrieving, in response to receiving the notification, historic maintenance data in relation to the first resource with which the fault is associated, the historic maintenance data originating from a time period preceding a time of the maintenance event;
causing a portion of the retrieved historic maintenance data identified as being indicative of the maintenance event to be stored as one or more precursor signals of the maintenance event; and
causing future maintenance data received from second resources, each of the second resources comprising a particular resource of respective second interrelated resources of a corresponding second system, each of the corresponding second systems being of a respective second

Assignees

• Palantir Technologies Inc

Inventors

• Spiro Ezra
• Menck Andre Frederico Cavalheiro
• Maag Peter
• Powell Thomas

Classifications

• G05B23/0237

  based on parallel systems, e.g. comparing signals produced at the same time by same type systems and detect faulty ones by noticing differences among their responses · CPC title

• G06F11/0766

  Error or fault reporting or storing · CPC title

• G06Q10/06

  Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling · CPC title

• G06F16/219

  Managing data history or versioning (querying versioned data G06F16/2474; querying temporal data G06F16/2477) · CPC title

• G06F11/3089

  Monitoring arrangements determined by the means or processing involved in sensing the monitored data, e.g. interfaces, connectors, sensors, probes, agents (software debugging using additional hardware using a specific debug interface G06F11/3656; performance evaluation by tracing or monitoring G06F11/3466) · CPC title