Yield enhancement techniques using neural network pattern detection

Neural network pattern recognition techniques for the detection of wafer and die level electrical failure patterns have been discussed in literature for many years. As a result of the research, many companies have written software to utilize neural network algorithms to provide pattern detection with varying degrees of success. Much of the original software, however, was not sufficiently automated to provide the users with adequate benefits to justify the cost and effort. Recent improvements in computer performance, programming techniques and the integration of statistics have created opportunities to utilize neural network pattern recognition with substantially less effort and cost. Analysis systems are now commercially available. This paper will look at the NEDA system available from DYM Corporation of Bedford Ma