Efficient Development of Diagnostics: Information Extraction from High Fidelity Analog System Simulation

Though analog simulation software has been available for nearly three decades, its full potential for the design of electrical, electronic and mechatronic systems is far from being realized. We describe research performed at Ford Motor Company directed at extracting more value from our analog simulation investment. Our approach uses models developed to analyze the nominal operation of systems to explore the operation of the systems when faults are present. Since we characterize complex systems, the amount of information generated by this analysis is very large and difficult to interpret. We discuss how the information can be efficiently presented to a design engineer using the techniques of data filtering, information extraction, and functional labels. These techniques streamline the Failure Mode and Effects Analysis (FMEA) process, allowing it to be performed earlier in the design process and updated more readily, resulting in timely and efficient development of diagnostic and test strategies.