Virtual prototyping advanced by statistic and stochastic methodologies

Abstract

The paper reports three examples of best industrial practice showing the substantial benefits gained in terms of time-to-market reduction when virtual prototyping is enhanced by statistical and stochastic methodologies. These examples from a microelectronics setting of high volume component and module manufacturing deal with different fields: i) ball grid array (BGA) design optimization based on sophisticated design-of-experiments (DoE) and response surface (RS) schemes, ii) material modeling based on stochastic parameter identification and optimization, and iii) process pre-qualification by involving a stochastic robustness analysis.