Model Selection Algorithm in Computer Experiments for Quality Improvements  

Oleh:

Lee, Young-Saeng ; Park, Jeong-Soo

Jenis: Article from Proceeding
Dalam koleksi: Asian Network for Quality (ANQ) Congress 2011, Ho Chi Minh City, Vietnam, 27-30 September 2011, page 1-6.
Topik: Gaussian Processes; Kriging; Simulation; Numerical Optimization; Maximum Likelihood Estimation; Spatial Linear Model;
Fulltext: KSQM15.YoungSaeng, LEE_Fullpaper.pdf (257.98KB)

Isi artikel

The model in our approach is to assume that the computer responses are a realization of a random function (Gaussian processes) superimposed on a spatial regression model as considered by Sacks, Welch, Mitchell and Wynn (1989), Statistical Science. Algorithms to build a good prediction model are proposed. It is a post-work of the algorithm of Welch et.al (1992), Technometrics. It selects some â’ s while the pre-selected è’s are fixed. Using tree examples, we illustrated the superiority of our proposed model over the other models. An application to quality improvement of integrated circuit design via computer experiments is provided.

Opini Anda

Klik untuk menuliskan opini Anda tentang koleksi ini!

Kembali