Next generation TPS architecture

The authors describe the symptom-model-based (SMB) approach, which correlates the failure symptom with the ambiguity group using historical data and diagnostic knowledge of the specific line replaceable units (LRUs). The SMB approach incorporates three key techniques for developing a next-generation TPS (test program set) architecture. The first technique is model-based diagnosis, which involves isolating the cause of failure based on the defined structure and functions of the components. Several different techniques and levels of detail for modeling an LRU are considered. The second technique is empirical diagnosis, which involves computing the most probable cause of failure using historical data and results from failure modes and effects analysis (FMEA). The third technique is rule-based diagnosis, which uses the knowledge of experts to isolate failures in an expedient manner. The implementation of each of these techniques is evaluated based on the capability to replace the fault with the correct component, the time to isolate the fault and the complexity of the associated TPS structure