The automated sneak analysis software described in this paper works well on classic sneak examples. It correctly identifies sneak paths without falsely reporting on nonexistent sneaks. It has been tried out on much larger real-world examples, and reproduces the kind of behavior that it had demonstrated on the classic sneaks. Future development of the tool will enable the exploration of internal states of complex components such as ECUs. In addition of provision to complete a sneak analysis in the presence of failures will further enhance the tool. The use of simulation and identification of significant system operation has enabled this level of performance. The system is able to use qualitative simulation early in the design process, to identify potential sneaks as early as possible. It can use the industry-standard simulator Saber later in the design process, to give a more detailed analysis. In each case, component models are not specific to sneak circuit analysis, but can be used to verify the operation of the design, to generate a failure modes and effects analysis (FMEA) report, and to provide information useful for building diagnostic systems.
[1]
Chris J. Price,et al.
Function-directed electrical design analysis
,
1998,
Artif. Intell. Eng..
[2]
Neal Snooke.
Simulating Electrical Devices with Complex Behaviour
,
1999,
AI Commun..
[3]
Chris Price,et al.
Identifying design glitches through automated design analysis
,
1999,
Annual Reliability and Maintainability. Symposium. 1999 Proceedings (Cat. No.99CH36283).
[4]
J. B. Bowles,et al.
Combining sneak circuit analysis and failure modes and effects analysis
,
1993,
Annual Reliability and Maintainability Symposium 1993 Proceedings.