Closed-loop diagnostic model maturation for complex systems

A method of maintaining an onboard reasoner for diagnosing failures on an aircraft that includes aircraft systems configured to report faults to the onboard reasoner, the method comprising:

• accessing diagnostic data received from an onboard computer of the aircraft that includes the onboard reasoner, the diagnostic data including a plurality of fault reports of failed tests reported by respective ones of the aircraft systems, and a plurality of diagnosed failure modes of at least some of the aircraft systems that caused the failed tests;
• building, using an off-board reasoner, an off-board diagnostic causal model that describes causal relationships between the failed tests and the plurality of diagnosed failure modes, the off-board diagnostic causal model built using a theoretic machine learning algorithm trained using historical diagnostic data;
• comparing the diagnostic data to the off-board diagnostic causal model;
• and based thereon,
• determining a new causal relationship in the causal relationships described by the off-board diagnostic causal model that is new relative to known causal relationships;
• and
• updating an onboard diagnostic causal model to further describe the new causal relationship, including producing an updated onboard diagnostic causal model, and uploading the updated onboard diagnostic causal model to the onboard computer using a loadable software airplane part upload tool set;
• wherein the comparing of the diagnostic data to the off-board diagnostic causal model further comprises comparing to actual failure data, diagnostic data, and maintenance action data collected from an onboard aircraft system, which reflect actual causal relationships.

A system for maintaining an onboard reasoner for diagnosing failures on an aircraft that includes aircraft systems configured to report faults to the onboard reasoner, the system comprising:

• an onboard computer including the onboard reasoner, the onboard computer configured to send diagnostic data including a plurality of fault reports of failed tests reported by respective ones of the aircraft systems, and a plurality of diagnosed failure modes of at least some of the aircraft systems that caused the failed tests;
• and
• an off-board computer including an off-board reasoner, the off-board computer configured to:
• access the diagnostic data received from the onboard computer;
• build an off-board diagnostic causal model that describes causal relationships between the failed tests and the plurality of diagnosed failure modes, the off-board diagnostic causal model built using a theoretic machine learning algorithm trained using historical diagnostic data, wherein the theoretic machine learning algorithm is a graph-theoretic algorithm for finding maximal cliques;
• compare the diagnostic data to the off-board diagnostic causal model;
• and based thereon,
• determine a new causal relationship in the causal relationships described by the off-board diagnostic causal model that is new relative to known causal relationships;
• and
• update an onboard diagnostic causal model to further describe the new causal relationship, including the off-board computer configured to produce an updated onboard diagnostic causal model, and upload the updated onboard diagnostic causal model to the onboard computer using a loadable software airplane part upload tool set.