Diagnostic information models: a practical approach to verticality, fault accountability, and diagnostic maturation of hierarchical systems

Modular testing and assembly is common for most high technology equipment such as military avionics systems. Systems that have several levels of testing and assembly are susceptible to the diagnostic condition of ReTest OK (RTOK) or Can Not Duplicate (CND). This paper presents a modeling approach to reduce the number of RTOKs and CNDs by focusing on fault accountability and test tolerances. Built-in test (BIT) is the primary means of detecting and isolating one or more failed black boxes on the aircraft. In the future, aircraft BIT may have the capability to isolate to one or more circuit cards of a specific black box. Hierarchical testing, which requires strict attention to test tolerances and fault accountability, is necessary to ensure the faults that are detectable by the higher levels of testing are also detectable by the lower levels of testing. A Diagnostic Information Model (DIM) can capture test tolerances and detectable faults for all levels of testing. The DIM can also capture the functionality of a system along with the faults that affect system performance. Processing DIM data with an automated tool can assist the engineer with performing a failure mode and effects analysis, verifying test tolerances, and creating a fault accountability matrix. The structure of the DIM supports information that directly pertains to testability, reliability, and test development, including all the data necessary to produce a test requirements document (TRD). Topics of the paper include fault accountability, vertical and horizontal accountability, cones of tolerance, as well as the elements and applications of the DIM.