System, method and computer product for performing automated predictive reliability

• US 7,107,491 B2
• Filed: 05/16/2001
• Issued: 09/12/2006
• Est. Priority Date: 05/16/2001
• Status: Active Grant

• Current Assignee: General Electric Company
• Original Assignee: General Electric Company
• Inventors: Nicolia, Raymond Louis Jr., Cardella, Aidan Thomas, Feher, Lucille Anna, Graichen, Catherine Mary, Osborn, Brock Estel
• Primary Examiner(s): Beausoliel, Robert
• Assistant Examiner(s): WILSON, YOLANDA L
• Application Number:
  US09/681,652
• Publication Number:
  US 20020174384A1
• Time in Patent Office: 1,945 Days
• Field of Search: 714/741, 714/37, 714/47, 714/48, 714/26, 714/25, 703/13
• US Class Current: 714/37
• CPC Class Codes: G06F 11/008 Reliability or availability analysis

Abstract

System, method and computer product for performing automated predictive reliability. A data acquisition component acquires service data for a complex system from a data repository. A statistical analysis component generates a statistical model for the service data. A simulation component predicts the reliability of the complex system according to the statistical model. An alert generation component generates alerts when predicted failures determined by the simulation component exceed predetermined alert criteria. A report generation component generates a summary of the analysis performed in the data acquisition, statistical analysis and alert generation components.

Claims

1. A system for performing automated predictive reliability on a complex system having a plurality of subsystems and a plurality of components within each subsystem, comprising:

a data acquisition component that acquires service data for the plurality of components of at least one of the plurality of subsystems and determines age information and failure information from the service data for each of the plurality of components;

a statistical analysis component that generates a statistical model according to the age information and failure information;

a simulation component that predicts future failures for the life cycle of the plurality of components according to the statistical model; and

an alert generation component that generates alerts for the predicted future failures.

2. The system according to claim 1, wherein the statistical model comprises a Weibull distribution model.

3. The system according to claim 1, further comprising a report generation component that compiles results produced from the simulation component.

4. A system performing automated predictive reliability on a complex system having a plurality of subsystems and a plurality of components within each subsystem, comprising:

at least one data repository containing a plurality of service data for the plurality of subsystems and components;

a predictive reliability system that predicts the reliability for the plurality of components of at least one of the plurality of subsystems according to the plurality of service data, the predictive reliability system comprising a data acquisition component that acquires the plurality of service data from the at least one data repository and determines age information and failure information from the service data for each of the plurality of components;

a statistical analysis component that generates a statistical model according to the age information and failure information; and

a simulation component that predicts future failures for the life cycle of the plurality of components according to the statistical model; and

an alert generation component that generates alerts for failures and incipient failures for the complex system from the predicted reliability; and

a first computing unit configured to serve the at least one data repository and the predictive reliability system.

5. The system according to claim 4, wherein the at least one data repository stores historical failure data for the complex system.

6. The system according to claim 4, wherein the at least one data repository stores analysis data for the complex system including data for subsystems and components that form the complex system.

7. The system according to claim 4, wherein the statistical model comprises a Weibull distribution model.

8. The system according to claim 4, wherein the predictive reliability system further comprises a report generation component that compiles results produced from the simulation component.

9. The system according to claim 4, further comprising a second computing unit configured to interact with the at least one data repository and the predictive reliability system served from the first computing unit over a network.

10. The system according to claim 4, further comprising an on-demand analyzer that enables a user to perform ad hoc analysis on the complex system.

11. The system according to claim 4, wherein the at least one data repository stores corrective actions for servicing any predicted problems.

12. The system according to claim 11, wherein the predictive reliability system provides a candidate set of corrective actions for servicing any predicted problems.

13. A method performing automated predictive reliability on a complex system having a plurality of subsystems and a plurality of components within each subsystem, comprising:

acquiring service data for the plurality of components of at least one of the plurality of subsystems;

determining age information and failure information from the service data for the plurality of components;

generating a statistical model according to the age information and failure information;

predicting future failures for the life cycle of the plurality of components according to the statistical model;

determining a plurality of statistics for the predicted future failures; and

issuing alerts for the predicted future failures.

14. The method according to claim 13, further comprising compiling the predicted future failures into a report.

15. The method according to claim 13, wherein the statistical model comprises a Weibull distribution model.

16. A computer-readable medium storing computer instructions for instructing a computer system to perform automated predictive reliability on a complex system having a plurality of subsystems and a plurality of components within each subsystem, the computer instructions comprising:

acquiring service data for the plurality of components of at least one of the plurality of subsystems;

determining age information and failure information from the service data for the plurality of components;

generating a statistical model according to the age information and failure information;

predicting future failures for the life cycle of the plurality of components according to the statistical model; and

issuing alerts for the predicted future failures.

17. The computer-readable medium according to claim 16, further comprising instructions for compiling the predicted future failures into a report.

18. The computer-readable medium according to claim 16, wherein the statistical model comprises a Weibull distribution model.