Diagnostic system and method for enabling multistage decision optimization for aircraft preflight dispatch

• US 7,209,814 B2
• Filed: 03/12/2004
• Issued: 04/24/2007
• Est. Priority Date: 12/04/2002
• Status: Expired due to Fees

• Current Assignee: The Boeing Co.
• Original Assignee: The Boeing Co.
• Inventors: Kipersztok, Oscar, Dildy, Glenn A.
• Primary Examiner(s): To; Tuan C
• Application Number:
  US10/799,989
• Publication Number:
  US 20040199307A1
• Time in Patent Office: 1,138 Days
• Field of Search: 701/29, 701/34, 701/100, 701/30, 701/3, 701/209, 701/14, 340/945, 340/963, 340/439, 702/182, 702/183, 702/185, 244/1.R, 714/47, 714/30, 714/4, 714/25, 382/149
• US Class Current: 701/29
• CPC Class Codes: B64F 5/60 Testing or inspecting aircraft components or systems

Abstract

A diagnostic system and method for enabling multistage decision optimization in aircraft preflight dispatch. The diagnostic system includes an interface for receiving one or more inputs relating to one or more observed symptoms indicative of a failed component in an aircraft. The diagnostic system extends an entropy-based value of information (VOI) diagnostic model by adding an explicit value function into the VOI diagnostic model to accommodate various variables associated with the aircraft preflight dispatch problem. The construction of the entropy-based VOI diagnostic model and thus the extended VOI diagnostic model are both based upon at least one of systemic information relating to aircraft components and input-output relationships of the aircraft components, experience-based information relating to direct relationships between aircraft component failures and observed symptoms, and factual information relating to aircraft component reliability.

Claims

1. A method enabling multistage decision making in preflight dispatch of a mobile platform, the method comprising:

extending an entropy-based value of information (VOI) diagnostic model to preflight dispatch to accommodate one or more variables associated with preflight dispatch of a mobile platform;

receiving at least one input relating to one or more observed symptoms indicative of one or more failed components of the mobile platform;

automatically determining a maintenance action for the mobile platform in accordance with the extended VOI diagnostic model and the one or more observed symptoms;

wherein the automatically determining includes;

correlating the at least one input relating to the one or more observed symptoms with one or more suspect components each capable of causing the one or more observed symptoms upon failure in accordance with the extended VOI diagnostic model;

prioritizing the one or more suspect components based upon a relative likelihood that the respective suspect components caused the one or more observed symptoms;

identifying and prioritizing one or more tests that can each be performed to refine the prioritization of the one or more suspect components; and

using the prioritized listing of tests to identify a test to perform or determine that testing should stop.

2. The method of claim 1, wherein the construction of the entropy-based VOI diagnostic model and the extended VOI diagnostic model are based upon a Bayesian network.

3. The method of claim 1, wherein the automatically determining includes automatically determining an optimal maintenance action for the mobile platform in accordance with the extended VOI diagnostic model and the one or more observed symptoms.

4. The method of claim 1, wherein the automatically determining includes automatically prioritizing one or more tests and remedial actions for the mobile platform over multiple stages in accordance with the extended VOI diagnostic model and the one or more observed symptoms.

5. The method of claim 1, wherein the maintenance action includes at least one of:

performing a test to refine a prioritization of the plurality of suspect components;

repairing a suspect component;

replacing a suspect component;

deferring maintenance on a suspect component;

delaying a flight;

canceling a flight; and

canceling a flight and replacing the mobile platform.

6. The method of claim 1, further comprising:

receiving at least one input relating to an outcome of a test; and

re-determining the maintenance action in light of the outcome of the test.

7. The method of claim 1, further comprising identifying the one or more variables associated with preflight dispatch.

8. The method of claim 1, wherein the one or more variables accommodated by the extended VOI diagnostic model include at least one of:

a decision parameter;

a utility function;

a constraint;

a cost function;

a cost limit;

a time deadline;

an airworthiness guideline;

maintenance crew expertise;

labor availability;

a future destination of the mobile platform;

repair equipment availability; and

component availability.

9. The method of claim 1, wherein the one or more variables accommodated by the extended VOI diagnostic model include:

a test decision; and

a repair decision interleaved with the test decision.

10. The method of claim 9, wherein:

the test decision includes at least one of;

performing a test to refine a prioritization of the plurality of suspect components; and

stopping testing and addressing the repair decision;

the repair decision includes at least one of;

repairing a suspect component;

replacing a suspect component;

deferring maintenance on a suspect component;

delaying a flight;

canceling a flight; and

canceling a flight and replacing the mobile platform.

11. The method of claim 1, wherein the entropy-based VOI diagnostic model is constructed based upon at least one of systemic information relating to components of the mobile platform and input-output relationships of the components, experience-based information relating to direct relationships between component failures and observed symptoms, and factual information relating to component reliability.

12. A method enabling multistage decision making in preflight dispatch of a mobile platform, the method comprising:

extending an entropy-based value of information (VOI) diagnostic model to preflight dispatch to accommodate one or more variables associated with preflight dispatch of a mobile platform;

receiving at least one input relating to one or more observed symptoms indicative of one or more failed components of the mobile platform;

automatically determining a maintenance action for the mobile platform in accordance with the extended VOI diagnostic model and the one or more observed symptoms;

wherein the one or more variables accommodated by the extended VOI diagnostic model includes a test decision and a repair decision interleaved with the test decision;

wherein the test decision includes at least one of;

performing a test to refine a prioritization of the plurality of suspect components; and

stopping testing and addressing the repair decision;

wherein the repair decision includes at least one of;

repairing a suspect component;

replacing a suspect component;

deferring maintenance on a suspect component;

delaying a flight;

canceling a flight; and

canceling a flight and replacing the mobile platform;

wherein the test decision is stopping testing and addressing the repair decision when no test having a positive value of information (VOI) or an estimated time and cost of completion within cost and time constraints prescribed for dispatch of the mobile platform can be identified.

13. A method enabling multistage decision making in preflight dispatch of a mobile platform, the method comprising:

extending an entropy-based value of information (VOI) diagnostic model to preflight dispatch to accommodate one or more variables associated with preflight dispatch of a mobile platform;

receiving at least one input relating to one or more observed symptoms indicative of one or more failed components of the mobile platform;

automatically determining a maintenance action for the mobile platform in accordance with the extended VOI diagnostic model and the one or more observed symptoms;

wherein the one or more variables accommodated by the extended VOI diagnostic model includes a test decision and a repair decision interleaved with the test decision;

wherein the test decision includes at least one of;

performing a test to refine a prioritization of the plurality of suspect components; and

stopping testing and addressing the repair decision;

wherein the repair decision includes at least one of;

repairing a suspect component;

replacing a suspect component;

deferring maintenance on a suspect component;

delaying a flight;

canceling a flight; and

canceling a flight and replacing the mobile platform;

wherein the test to perform includes a positive net VOI and an estimated time and cost of completion within cost and time constraints associated with the dispatch of the mobile platform.

14. A method enabling multistage decision making in a maintenance troubleshooting operation, the method comprising:

extending an entropy-based value of information (VOI) diagnostic model constructed based upon at least one of systemic information relating to components of a mobile platform and input-output relationships of the components, experience-based information relating to direct relationships between component failures and observed symptoms, and factual information relating to component reliability, the extended VOI diagnostic model accommodating one or more variables associated with the maintenance troubleshooting operation;

receiving at least one input relating to one or more observed symptoms indicative of one or more failed components of the mobile platform; and

automatically determining a maintenance action for the mobile platform in accordance with the extended VOI diagnostic model and the one or more observed symptoms.

15. The method of claim 14, wherein the extending includes extending the entropy-based value of information (VOI) diagnostic model to preflight dispatch to accommodate one or more variables associated with preflight dispatch of the mobile platform.

16. The method of claim 14, wherein the construction of the entropy-based VOI diagnostic model and the extended VOI diagnostic model are based upon a Bayesian network.

17. The method of claim 14, wherein the automatically determining includes automatically determining an optimal maintenance action for the mobile platform in accordance with the extended VOI diagnostic model and the one or more observed symptoms.

18. The method of claim 14, wherein the automatically determining includes automatically prioritizing one or more tests and remedial actions for the mobile platform over multiple stages in accordance with the extended VOI diagnostic model and the one or more observed symptoms.

19. The method of claim 14, wherein the maintenance action includes at least one of:

performing a test to refine a prioritization of the plurality of suspect components;

repairing a suspect component;

replacing a suspect component;

deferring maintenance on a suspect component;

delaying a flight;

canceling a flight; and

canceling a flight and replacing the mobile platform.

20. The method of claim 14, wherein the automatically determining includes:

correlating the at least one input relating to the one or more observed symptoms with one or more suspect components each capable of causing the one or more observed symptoms upon failure in accordance with the extended VOI diagnostic model;

prioritizing the one or more suspect components based upon a relative likelihood that the respective suspect components caused the one or more observed symptoms;

identifying and prioritizing one or more tests that can each be performed to refine the prioritization of the one or more suspect components; and

using the prioritized listing of tests to identify a test to perform or determine that testing should stop.

21. The method of claim 14, further comprising:

receiving at least one input relating to an outcome of a test; and

re-determining the maintenance action in light of the outcome of the test.

22. The method of claim 14, further comprising:

receiving at least one input relating to a remedial action undertaken with respect to at least one suspect component; and

re-determining the maintenance action in light of the remedial action.

23. The method of claim 14, further comprising identifying the one or more variables associated with the maintenance troubleshooting operation.

24. The method of claim 14, wherein the one or more variables accommodated by the extended VOI diagnostic model includes at least one of:

a decision parameter;

a utility function;

a constraint;

a cost function;

a cost limit;

a time deadline;

an airworthiness guideline;

maintenance crew expertise;

labor availability;

a future destination of the mobile platform;

repair equipment availability; and

component availability.

25. The method of claim 14, wherein the one or more variables accommodated by the extended VOI diagnostic model includes:

a test decision; and

a repair decision interleaved with the test decision.

26. The method of claim 25:

wherein the test decision includes at least one of;

performing a test to refine a prioritization of the plurality of suspect components; and

stopping testing and addressing the repair decision;

wherein the repair decision includes at least one of;

repairing a suspect component;

replacing a suspect component;

deferring maintenance on a suspect component;

delaying a flight;

canceling a flight; and

canceling a flight and replacing the mobile platform.