Sheppard, J.W.; Kaufman, M.A., "A Bayesian approach to diagnosis and prognosis using built-in test,"

Sheppard, J.W.; Kaufman, M.A., "A Bayesian approach to diagnosis and prognosis using built-in test," Instrumentation and Measurement, IEEE Transactions on , vol.54, no.3pp. 1003- 1018, June 2005

Abstract: Accounting for the effects of test uncertainty is a significant problem in test and diagnosis, especially within the context of built-in test. Of interest here, how does one assess the level of uncertainty and then utilize that assessment to improve diagnostics? One approach, based on measurement science, is to treat the probability of a false indication [e.g., built-in-test (BIT) false alarm or missed detection] as the measure of uncertainty. Given the ability to determine such probabilities, a Bayesian approach to diagnosis, and by extension, prognosis suggests itself. In the following, we present a mathematical derivation for false indication and apply it to the specification of Bayesian diagnosis. We draw from measurement science, reliability theory, signal detection theory, and Bayesian decision theory to provide an end-to-end probabilistic treatment of the fault diagnosis and prognosis problem.URL:

Model Based Methods (MBD) vs. Data Driven Methods (DDM)

To analyze any dynamic system there could be two approaches; first is Model Based Methods (MBD) which requires knowledge of the physics of the system and second is Data Driven Methods (DDM) which is driven by the input, output and testdata to derive the system. Naturally, each method has its own advantages and disadvantages. MBD provides insight into the behavior of the system by using differenial equations, state space equations etc. On the negative side, MBD is time consuming and it cannot model system which contains nonlinear parameters or system is nonlinear. Contrary, DDM a fast and easy to develop methods which requires input output data to train the algorithms and learn the system parameters. DDM requires multiple data sets to cover a wide range of system operations.

Challenge lies on integrating MBD and DDM. MATLAB and Simulink offers a good platform to test both approaches. It will be valuable if MBD and DDM can be performed under one umbrella.

Tangram

I came to know about an interesting project Tangram. The project is related to develop model based methods to test and integrate complex high-tech systems. Key idea is to involve model based methods right from the beginning and carry it through out the product development phase. See Tangram publications page for more details.