Systems and methods for using a corrective action as diagnostic evidence

• US 8,732,524 B2
• Filed: 08/03/2011
• Issued: 05/20/2014
• Est. Priority Date: 08/03/2011
• Status: Active Grant

• Current Assignee: Honeywell International Inc.
• Original Assignee: Honeywell International Inc.
• Inventors: Barton, Bradley John, Kolbet, David Michael, Shao, Qingqiu Ginger, Magnuson, Randy R.
• Primary Examiner(s): SCHELL, JOSEPH O
• Application Number:
  US13/197,132
• Publication Number:
  US 20140082417A1
• Time in Patent Office: 1,021 Days
• Field of Search: 714/25, 714/26, 714/2, 714/47.1, 714/47.3
• US Class Current: 714/26
• CPC Class Codes: G05B 2219/24019 Computer assisted maintenance
• CPC Class Codes: G05B 2219/32226 Computer assisted repair, maintenance of system components
• CPC Class Codes: G05B 23/0283 Predictive maintenance, e.g. involving the monitoring of a system and, based on the monitoring results, taking decisions on the maintenance schedule of the monitored system; Estimating remaining useful life [RUL] preventive maintenance, i.e. planning maintenance according to the available resources without monitoring the system G06Q10/06
• CPC Class Codes: G06F 11/079 Root cause analysis, i.e. error or fault diagnosis in a hardware test environment G06F11/22; in a software test environment G06F11/36
• CPC Class Codes: G06F 11/0793 Remedial or corrective actions recovery from an exception in an instruction pipeline G06F9/3861; by retry G06F11/1402; for recovering from a failure of a protocol instance or entity H04L69/40
• CPC Class Codes: G06F 11/2257 using expert systems
• CPC Class Codes: Y02P 90/02 Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]

Abstract

Methods, systems and computing devices are provided for using a completed corrective action as evidence of a fault. The methods, systems and computing devices receive equipment status evidence and determine an equipment fault based on the equipment status evidence. The methods, systems and computing devices also create and rank a list of potential failure modes based at least in part on the determined equipment fault, recommend a corrective action to correct the equipment fault based at least in part on the ranking of the potential failure modes and receiving additional equipment status evidence indicating that the recommended corrective action failed to correct the equipment fault. The methods, systems and computing devices then associate a detection probability and a false negative rate with the failed corrective action to create additional status evidence, and re-rank the list of potential failure modes for subsequent performance based on the additional status evidence.

Claims

1. A method using a completed corrective action as evidence of a fault by a diagnostic reasoner, comprising:

receiving equipment status evidence;

determining an equipment fault based on the equipment status evidence;

creating and ranking a list of potential failure modes based at least in part on the determined equipment fault;

recommending a corrective action to correct the equipment fault based at least in part on the ranking of the potential failure modes;

receiving additional equipment status evidence indicating that the recommended corrective action failed to correct the equipment fault;

associating a detection probability and a false negative rate with the failed corrective action to create additional status evidence wherein the false negative rate is equal to a ratio, wherein the numerator of the ratio is equal to the number of times the corrective action was performed on a valid equipment fault minus the number of times the corrective action corrected the valid equipment fault, and the denominator is the number of times the corrective action was performed on a valid equipment fault; and

re-ranking the list of potential failure modes for recommendation of a subsequent corrective action to be performed based on the additional status evidence.

2. The method of claim 1, wherein recommending a corrective action is selecting the corrective action associated with the failure mode with the highest relative probability of failure.

3. The method of claim 1, wherein the detection probability and false negative rates are retrieved from a remote persistent data store.

4. The method of claim 1, wherein the recommended corrective action and a successful corrective action are stored in a remote persistent data store.

5. The method of claim 1, wherein associating the detection probability and the false negative rate comprises storing the indication of the failed corrective action, the detection probability of the corrective action and the false negative rate in a common data record.

6. A system configured to use a completed corrective action as evidence of a fault, comprising:

an equipment status determination means;

a computing device configured to;

determine an equipment fault based on equipment status evidence received from the equipment status determination means;

create and rank a list of potential failure modes based at least in part on the determined equipment fault;

receive an indication that a first corrective action for the equipment fault has failed;

associate a detection probability and a false negative rate with the failed corrective action thereby creating additional equipment status evidence, wherein the false negative rate is equal to a ratio, wherein the numerator of the ratio is equal to the number of times the corrective action was performed on a valid equipment fault minus the number of times the corrective action corrected the valid equipment fault, and the denominator is the number of times the corrective action was performed on a valid equipment fault; and

re-rank and determine a second corrective action based at least in part on the additional equipment status evidence.

7. The system of claim 6, wherein an equipment status determination means is a sensor.

8. The system of claim 6, wherein an equipment status determination means is a manual input of an operator observation.

9. The system of claim 6, wherein an indication that the first corrective action has failed is a manual input of an operator observation.

10. The system of claim 6, wherein an indication that the first corrective action has failed is Built-in-Test (BIT) error indication.

11. The system of claim 6, wherein an indication that the first corrective action has failed is lack of a sensor input.

12. The system of claim 6, wherein an indication that the first corrective action has failed is a sensor input.

13. A computing device configured to:

receive equipment status evidence from an equipment status determination means;

determine an equipment fault based on the equipment status evidence;

create and rank a list of potential failure modes based at least in part on the determined equipment fault;

recommend a corrective action to correct the equipment fault based at least in part on the ranking of the potential failure modes from the list;

receive an indication that the recommended corrective action failed to correct the equipment fault;

associate a detection probability and a false negative rate with the failed corrective action to create additional equipment status evidence, wherein the false negative rate is equal to a ratio, wherein the numerator of the ratio is equal to the number of times the corrective action was performed on a valid equipment fault minus the number of times the corrective action corrected the valid equipment fault, and the denominator is the number of times the corrective action was performed on a valid equipment fault; and

re-rank the list of potential failure modes based on the additional equipment status evidence.

14. The computing device of claim 13, wherein recommending a corrective action is selecting the corrective action associated with the failure mode with the highest relative probability of failure.

15. The computing device of claim 13, wherein recommending a corrective action comprises selecting a corrective action associated with a subset of the potential failure modes with a highest combined relative probability of failure that share the same corrective action.

16. The computing device of claim 13, wherein an equipment status determination means is a manual input of an operator observation.

17. The computing device of claim 13, wherein an indication that the recommended corrective action has failed is a manual input of an operator observation.

18. The computing device of claim 13, wherein an indication that the recommended corrective action has failed is Built-in-Test (BIT) error indication.

19. The computing device of claim 13, wherein an indication that the recommended corrective action has failed is lack of a sensor input.

20. The computing device of claim 13, wherein an indication that the recommended corrective action has failed is a sensor input.

21. A computer based method using a history of validated faults and completed corrective actions to compute a detection probability and false negative rate for a corrective action associated with an equipment fault model, comprising:

over a period of time isolating, collecting and analyzing data associated with a corrective action, a corrective action success rate, a modeling error and an actual defect rate in regard to an associated validated fault in a complex system;

determining an actual defect rate of a component of the complex system from the data;

determining a modeling error of the equipment default model from the data;

determining a corrective action success rate from the data;

determining a corrective action count from the data;

summing the actual defect rate and the modeling error thereby determining that detection ration for the corrective action; and

subtracting the corrective action count from the corrective action success rate to determine a difference and dividing the difference by the corrective action count.