SYSTEM AND METHOD FOR FAILURE ASSOCIATION ANALYSIS

• US 20120173301A1
• Filed: 01/04/2011
• Published: 07/05/2012
• Est. Priority Date: 01/04/2011
• Status: Abandoned Application

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Feris, Rogerio S., Hampapur, Arun, Jiang, Zhong Bo, Mahatma, Shilpa N., Sun, Wei, Xie, Lexing, Dong, Wei Shan
• Application Number:
  US12/984,019
• Publication Number:
  US 20120173301A1
• Time in Patent Office: Days
• US Class Current: 705/7.28
• CPC Class Codes: G06Q 10/0635 Risk analysis of enterprise or organisation activities

Abstract

A system and method for mining the failure association rules of geographically dispersed physical assets is provided. One approach of the present invention has steps of joining input data sources, extracting spatio-temporal (ST) information, quantilizing ST continuous value in automated manner, or based on pre-built knowledge, applying association rule mining algorithm to find associations between attributes and failure and outputting identified ST failure association rules.

Claims

1. A method for mining the failure association rules of geographically dispersed physical assets comprising:

modeling asset failure pattern considering spatial and temporal correlation;

extracting, automatically, spatio-temporal attributes; and

quantilizing, automatically, a continuous spatio-temporal value.

2. The method in claim 1 further comprising taking into account additional environment variables for failure association mining.

3. The method in claim 2 further comprising considering spatial-temporal correlation in relation to association rule mining.

4. The method in claim 3 further comprising basing the failure association rule mining on historic data.

5. The method in claim 4 further comprising recording the historic failure data.

6. A computer-readable medium storing computer instructions, which, when executed, enables a computer system having at least one processor and memory to implement a method for mining the failure association rules of geographically dispersed physical assets, the method comprising:

modeling asset failure pattern considering spatial and temporal correlation;

extracting, automatically, spatio-temporal attributes; and

quantilizing, automatically, a continuous spatio-temporal value.

7. The computer-readable medium in claim 6 wherein the method further comprises taking into account additional environment variables for failure association mining.

8. The computer-readable medium in claim 7 wherein the method further comprises considering spatial-temporal correlation in relation to association rule mining.

9. The computer-readable medium in claim 8 wherein the method further comprises basing the failure association rule mining on historic data.

10. The computer-readable medium in claim 9 wherein the method further comprises recording the historic failure data at regular or irregular intervals.

11. A method for deploying a system for mining the failure association rules of geographically dispersed physical assets comprising:

modeling asset failure pattern considering spatial and temporal correlation;

extracting, automatically, spatio-temporal attributes; and

quantilizing, automatically, a continuous spatio-temporal value.

12. The method in claim 11 further comprising taking into account additional environment variables for failure association mining.

13. The method in claim 12 further comprising considering spatial-temporal correlation in relation to association rule mining.

14. The method in claim 13 further comprising basing the failure association rule mining on historic data.

15. The method in claim 14 further comprising recording the historic failure data at regular or irregular intervals.

16. A system having at least one processing unit and memory for mining the failure association rules of geographically dispersed physical assets comprising:

a failure association rules mining processing unit; and

input data memory.

17. The system as defined in claim 16 further comprising a spatio-temporal (ST) attributes extractor.

18. The system as defined in claim 17 further comprising a data joining manager for receiving information from asset internal attributes database, external data database and failure history database and joining that data.

19. The system as defined in claim 18 further comprising a continuous ST value quantilizer for receiving ST data and pre-built knowledge.

20. The system as defined in claim 19 further comprising a hierarchical association rule miner for receiving joined data and quantilized ST data and producing output rules.