Fusion of first-principles and statistical analyses in complex electronics systems

This paper present a hybrid deterministic and stochastic formulation for the quantitative statistical analysis of IC and electronic systems in large complicated enclosures. The work seamlessly integrates the universal statistical properties of the systems and the site-specific features within a comprehensive statistical analysis framework. A physics-based surrogate modeling capability is established, which generates the statistical response in nearly real time while retaining the underlying first-principles analysis. The work is validated through representative experiments.