Improving the manufacturability of electronic designs

The application of statistical circuit design techniques can prevent problems of low yield in manufacture of circuits. Statistical circuit design techniques analyze the yields of circuit designs whose underlying components exhibit random fluctuations. These techniques can help produce more robust designs by calling attention to areas where statistical variations are likely to combine in such a way as to cause circuit failure. Nonparametric boundary analysis (NBA), a technique introduced in Hewlett Packard EEsof's IC-CAP 5.0, permits yield analysis when the random fluctuations result from an arbitrary stochastic process, in addition to well-studied processes such as the Gaussian.