New developments on automation of sneak analysis

The paper describes the approach used to develop a sneak analysis procedure and a computerised tool (SNAP) to support its performance. The main SNAP capabilities are outlined and the approach used to automatically propose (tentative) answers to the clues is presented. For two trial cases, the results obtained with SNAP are compared with those obtained through manual analysis. A reduction in the manpower required of about one order of magnitude resulted from the use of SNAP. It is stressed that the optimal way to perform sneak analysis involves a coordinated use of the analyst's skills and computerised support. The paper also discusses how the sneak analysis procedure can be applied to systems containing computer software. The results of sneak path analysis (SPA) application on a trial case are presented. Advantages of SPA for hardware/software systems include the capability to apply the same procedure on both sides of the interface and the ability to deal with dynamic problems.<>