Method and system for identifying critical faults in machines

• US 6,324,659 B1
• Filed: 10/28/1999
• Issued: 11/27/2001
• Est. Priority Date: 10/28/1999
• Status: Expired due to Term

• Current Assignee: GE Global Sourcing LLC (Westinghouse Air Brake Technologies Corporation)
• Original Assignee: General Electric Company
• Inventors: Pierro, Michael J.
• Primary Examiner(s): LE, DIEU MINH T
• Application Number:
  US09/429,380
• Time in Patent Office: 761 Days
• Field of Search: 714/48, 714/47, 714/25, 714/38, 714/40
• US Class Current: 714/48
• CPC Class Codes: B61L 2205/04 Satellite based navigation systems, e.g. global positioning system [GPS]
• CPC Class Codes: B61L 27/57 for vehicles or trains, e.g. trackside supervision of train conditions
• CPC Class Codes: G06F 11/0715 in a system implementing multitasking multitasking per se G06F9/46
• CPC Class Codes: G06F 11/2257 using expert systems
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling

Abstract

A method and system for identifying a plurality of critical faults in machines is provided. The method allows for collecting from a predetermined plurality of the machines respective machine data indicative of each fault logged over a predetermined period of time. Respective identifying steps allow for identifying in the collected machine data respective faults most frequently occurring relative to one another and for identifying in the most frequently occurring faults, respective faults that, relative to one another, affect a higher number of machines. A classifying step allows for classifying the faults identified in the last-recited identifying step based on an expected level of machine degradation associated with the identified faults, and a storing step allows for storing any faults classified as likely to result in an imminent machine mission failure in a predetermined database of critical faults.

Claims

1. A method for identifying a plurality of critical faults in a fleet of machines, the method comprising:

a) collecting from a predetermined plurality of the fleet of machines respective fleet machine data indicative of each fault logged over a predetermined period of time;

b) determining the relative ranking of the faults logged in the collected fleet machine data by;

i) identifying in the collected fleet machine data respective faults most frequently occurring relative to one another;

ii) identifying in the most frequently occurring faults, respective faults that, relative to one another, affect a higher number of machines;

c) classifying the faults identified in step b) based on an expected level of machine degradation associated with the identified faults; and

d) storing any faults classified as likely to result in an inmminent machine mission failure in a predetermined database of critical faults.

2. The method of claim 1 further comprising a step of executing expert analysis prior to the storing step.

3. The method of claim 2 further comprising a step of inputting into the database of critical faults respective faults being classified as under investigation and/or of special interest.

4. The method of claim 2 further comprising a step of inputting into the database of critical faults respective faults being defined by the presence of predetermined trends in the respective machine data.

5. The method of claim 1 wherein the plurality of machines comprises a fleet of locomotives.

6. The method of claim 1 wherein the machine data comprises respective fault log data of respective subsystems of the machine.

7. The method of claim 6 wherein the machine data comprises respective operational parameter data of the machine.

8. The method of claim 7 wherein the operational parameter data comprises snapshot data.

9. The method of claim 7 wherein the operational parameter data comprises substantially continuous date.

10. The method of claim 1 further comprising a step of executing new iterations of steps a) through e) at predetermined intervals so as to maintain the database of critical faults substantially up to date.

11. The method of claim 1 further comprising a step of executing new iterations of steps a) through e) upon deployment of configuration changes and/or new models of the machine.

12. The method of claim 1 wherein the database of critical faults is used in a process for managing communication of electronic data between a diagnostic service center and a plurality of machines generally remote relative to each other, the electronic data comprising at least respective new machine data from selected machines, the process comprising:

storing in a database a list of respective cases to be processed;

assigning to each case a respective download priority;

determining each case to be populated next with new machine data based at least upon the assigned download priority; and

executing a download of the new machine data wherein said download of new machine data is triggered upon a call from a respective machine to the service center, the call identifying occurrence in the respective machine of one or more faults as stored in the critical fault database.

13. The method of claim 12 wherein the call to the service center is automated upon detection in the machine of one or more of the faults as stored in the critical fault database.

14. A system for identifying a plurality of critical taults in a fleet of machines, the system comprising:

a data-collecting module configured to collect from a predetermined plurality of the machines respective machine data indicative of each fault logged over a predetermined period of time;

a processor for determining the relative ranking of the faults logged in the collected fleet machine data by;

identifying in the collected machine data respective faults most frequently occurring relative to one another;

identifying in the most frequently occurring faults, respective faults that, relative to one another, affect a higher number of machines;

a classifier configured to classify the faults identified by the processor based on an expected level of machine degradation associated with the identified faults; and

a database coupled to the classifier- to store any faults classified as likely to result in an imminent machine mission failure, the stored faults comprising the plurality of critical faults.

15. The system of claim 14 further comprising a module for executing expert analysis prior to storing faults in the database of critical faults.

16. The system of claim 15 further comprising a module for inputting into the database of critical faults respective faults being classified as under investigation and/or special interest.

17. The system of claim 15 further comprising a module for inputting into the database of critical faults respective faults being defined by the presence of predetermined trends in the respective machine data.

18. The system of claim 14 wherein the plurality of machines comprises a fleet of locomotives.

19. The system of claim 14 wherein the machine data comprises respective fault log data of respective subsystems of the machine.

20. The system of claim 14 wherein the machine data comprises respective operational parameter data of the machine.

21. The system of claim 20 wherein the operational parameter data comprises snapshot data.

22. The system of claim 20 wherein the operational parameter data comprises substantially continuous data.

23. The system of claim 14 further comprising a module for executing new iterations upon new machine data at predetermined intervals so as to maintain the database of critical faults substantially up to date.

24. The system of claim 14 further comprising a module for executing new iterations upon new machine data available upon deployment of configuration changes and/or new models of the machine.

25. The system of claim 14 wherein the database of critical faults is used in a processor for managing communication of electronic data between a diagnostic service center and a plurality of machines generally remote relative to each other, the electronic data comprising at least respective new machine data from selected machines, the processor comprising:

a module configured to execute a download of the new machine data wherein said download of new machine data is triggered upon a call from a respective machine to the service center, the call identifying occurrence in the respective machine of one or more faults as stored in the critical fault database.

26. The system of claim 25 wherein the call to the service center is automated upon detection in the machine of one or more of the faults as stored in the critical fault database.

27. A method for identifying critical faults in a plurality of mobile assets, the method comprising:

a) collecting from a group of the plurality of mobile assets respective mobile asset data indicative of each fault logged over a predetermined period of time;

b) classifying respective faults in the collected mobile asset data based on the following criteria;

1) relative frequency of fault occurrence;

2) number of mobile assets affected in the group; and

3) expected level of mobile asset degradation;

wherein any of the three criteria comprises a first basis of classification, and a second classification is based on the results of the first classification so that any faults found to be critical include properties in at least two of the classifications; and

c) storing any faults found to be critical in a database of critical faults.

28. The method of claim 27 wherein all three criteria are separately considered in sequence and further wherein each classification is based on the results of any previous classification so that the faults found to be critical include properties in all three classifications.

29. A system for identifying critical faults in mobile assets, the system comprising:

memory configured to collect from a plurality of the mobile assets respective mobile asset data indicative of each fault logged over a predetermined period of time;

a first classifier configured to classify in the collected mobile asset data respective faults most frequently occurring relative to one another;

a second classifier configured to classify in the most frequently occurring faults from the first classifier, respective faults that, relative to one another, affect a higher number of assets;

a third classifier configured to classify the faults from the second classifier based on an expected level of mobile asset degradation; and

a database coupled to the third classifier to store any faults classified as likely to result in an imminent mobile asset mission failure, the stored faults comprising the plurality of critical faults.