Methods and systems for predicting a fault

• US 9,081,656 B2
• Filed: 12/20/2011
• Issued: 07/14/2015
• Est. Priority Date: 12/20/2011
• Status: Active Grant

• Current Assignee: NCR Voyix Corporation
• Original Assignee: NCR Corporation (NCR Voyix Corporation)
• Inventors: Yingling, Harold R. Jr., Evans, Cynthia Hill, Bass, Larry D.
• Primary Examiner(s): BUTLER, SARAI E
• Application Number:
  US13/330,847
• Publication Number:
  US 20130159787A1
• Time in Patent Office: 1,302 Days
• Field of Search: 714/47.3, 714/25, 714/48
• US Class Current: 1/1
• CPC Class Codes: G06F 11/008 Reliability or availability analysis

Abstract

Disclosed is a fault prediction system and method that uses non-fatal event data received from a terminal to make predictions concerning future fatal faults for the terminal. A complex fault pattern associated with a fault is applied to the non-fatal event data to predict the fault. A corrective action is provided for each predicted fault and historical data is used to predict a time to the predicted fault to govern the type of service response to create to prevent the fault.

Claims

1. A method for predicting a fault in a self-service or assisted service terminal comprising:

receiving event data associated with a terminal;

selecting event data relevant to a fault pattern associated with a predictable fault;

comparing the selected event data to the fault pattern associated with the predictable fault, wherein the fault pattern is based at least in part on historical event data and the fault pattern includes information that defines two event patterns, a first event pattern identifying that within a predetermined period of time a certain number of non-fatal events of a certain type have been detected, and a second event pattern determines that the first event pattern has been repeated a predetermined number of times; and

predicting a fault for the terminal in an instance in which the selected event data matches the fault pattern.

2. The method of claim 1, further comprising:

updating the historical event data with the received event data; and

modifying the fault pattern based at least in part on the updated historical event data.

3. The method of claim 1, wherein the fault pattern defines a frequency condition that is met when a specified event occurs a first specified number of times within a specified time duration, and wherein predicting the fault further comprises determining that the event data indicates that the specified event has occurred at least the first specified number of times within the specified time duration.

4. The method of claim 3, wherein the fault pattern defines a recurrence condition that is met when the frequency condition is repeated a second specified number of times, and wherein predicting the fault further comprises determining that the event data indicates that the frequency condition has been repeated the second specified number of times.

5. The method of claim 1, wherein comparing the event data to a fault pattern further comprises comparing the event data to each of a plurality of fault patterns, and wherein predicting a fault further comprises predicting a fault for each fault pattern that the event data matches.

6. The method of claim 1, further comprising:

generating a signal identifying the predicted fault.

7. The method of claim 6, wherein the signal comprises an indication of a predicted time until the predicted fault occurs.

8. The method of claim 1, wherein historical event data comprises event data generated prior to the received event data.

9. The method of claim 1, further comprising:

generating a request to schedule a service call related to the predicted fault.

10. The method of claim 1, further comprising:

generating a request to perform an action to prevent the predicted fault.

11. The method of claim 1, further comprising:

receiving tally data indicating a number of times a particular event associated with a component of the terminal has occurred during the life of the component;

determining that the number exceeds an expected number of times the particular event should occur prior to failure of the component; and

adjusting the fault pattern based at least in part on the determination.

12. The method of claim 1, wherein the terminal performs at least one of the receiving, comparing, and predicting steps.

13. A method for predicting a fault in a self-service or assisted service terminal comprising:

receiving event data associated with a terminal;

identifying a condition that predicts a fault based at least in part on the event data; and

creating a fault pattern representing the condition for predicting the fault and creating the fault pattern with information that defines two event patterns, a first event pattern identifying that within a predetermined period of time a certain number of non-fatal events of a certain type have been detected, and a second event pattern determines that the first event pattern has been repeated a predetermined number of times.

14. The method of claim 13, further comprising:

receiving additional event data associated with the terminal;

updating the event data with the additional event data;

testing the fault pattern using the updated event data; and

modifying the fault pattern based at least in part on the testing.

15. The method of claim 13, wherein receiving event data associated with a terminal further comprises receiving event data associated with at least one additional terminal.

16. The method of claim 15, further comprising:

receiving additional event data associated with the at least one additional terminal;

updating the event data with the additional event data;

testing the fault pattern using the updated event data; and

modifying the fault pattern based at least in part on the testing.

17. The method of claim 13, wherein the terminal performs at least one of the receiving, identifying, and creating steps.

18. The method of claim 13, further comprising:

receiving additional event data associated with the terminal;

comparing the additional event data to the fault pattern; and

predicting a fault for the terminal in an instance in which the additional event data matches the fault pattern.

19. An apparatus for predicting a fault in a self-service or assisted service terminal comprising:

at least one processor; and

at least one memory comprising computer program code, the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus at least to;

receive event data associated with a terminal;

compare the event data to a fault pattern associated with a predictable fault, wherein the fault pattern is based at least in part on historical event data and the fault pattern includes information that defines two event patterns, a first event pattern identifying that within a predetermined period of time a certain number of non-fatal events of a certain type have been detected, and a second event pattern determines that the first event pattern has been repeated a predetermined number of times; and

predict a fault for a hardware component associated with the terminal in an instance in which the event data matches the fault pattern.

20. The apparatus of claim 19, wherein the at least one memory and the computer program code are further configured to, with the at least one processor, cause the apparatus to:

event data;

update the historical event data with the received event data; and

modify the fault pattern based at least in part on the updated historical event data.

21. The apparatus of claim 19, wherein the fault pattern defines a frequency condition that is met when a specified event occurs a first specified number of times within a specified time duration, and wherein predicting the fault further comprises determining that the event data indicates that the specified event has occurred at least the first specified number of times within the specified time duration.

22. The apparatus of claim 21, wherein the fault pattern defines a recurrence condition that is met when the frequency condition is repeated a second specified number of times, and wherein predicting the fault further comprises determining that the event data indicates that the frequency condition has been repeated the second specified number of times.

23. The apparatus of claim 19, wherein in order to compare the event data to a fault pattern, the at least one memory and the computer program code are further configured to, with the at least one processor, cause the apparatus to compare the event data to each of a plurality of fault patterns;

and wherein in order to predict a fault, the at least one memory and the computer program code are further configured to, with the at least one processor, cause the apparatus to predict a fault for each fault pattern that the event data matches.

24. The apparatus of claim 19, wherein the at least one memory and the computer program code are further configured to, with the at least one processor, cause the apparatus to:

generate a signal identifying the predicted fault.

25. The apparatus of claim 24, wherein the signal comprises an indication of a predicted time until the predicted fault occurs.

26. The apparatus of claim 19, wherein historical event data comprises any event data generated prior to the received event data.

27. The apparatus of claim 19, wherein the at least one memory and the computer program code are further configured to, with the at least one processor, cause the apparatus to:

generate a request to schedule a service call related to the predicted fault.

28. The apparatus of claim 19, wherein the at least one memory and the computer program code are further configured to, with the at least one processor, cause the apparatus to:

generate a request to perform an action to prevent the predicted fault.

29. The apparatus of claim 19, wherein the at least one memory and the computer program code are further configured to, with the at least one processor, cause the apparatus to:

receive tally data indicating a number of times a particular event associated with a component of the terminal has occurred during the life of the component;

determine that the number exceeds an expected number of times the particular event should occur prior to failure of the component; and

adjust the fault pattern based at least in part on the determination.

30. The apparatus of claim 19, wherein the apparatus is the terminal.