TY - GEN
T1 - Selecting an Optimal Subset with Regression Metamodels
AU - Gao, Fei
AU - Li, Yanwen
AU - Gao, Siyang
AU - Xiao, Hui
PY - 2019/8
Y1 - 2019/8
N2 - In this paper, we consider the ranking and selection problem of selecting the optimal subset from a finite set of alternative designs. Given the total simulation budget constraint, we aim to maximize the probability of correctly selecting the top-m designs. In order to further improve the selection efficiency, we incorporate the information from across the domain into quadratic regression equation. Under some common assumptions in most regression based approaches, we propose an approximately optimal rule that determines the design locations need to be simulated and the number of simulation replications allocated to the selected designs. Numerical experiments demonstrate that our approach dramatically improves the selection efficiency on some typical selection examples compared to the existing approaches.
AB - In this paper, we consider the ranking and selection problem of selecting the optimal subset from a finite set of alternative designs. Given the total simulation budget constraint, we aim to maximize the probability of correctly selecting the top-m designs. In order to further improve the selection efficiency, we incorporate the information from across the domain into quadratic regression equation. Under some common assumptions in most regression based approaches, we propose an approximately optimal rule that determines the design locations need to be simulated and the number of simulation replications allocated to the selected designs. Numerical experiments demonstrate that our approach dramatically improves the selection efficiency on some typical selection examples compared to the existing approaches.
UR - http://www.scopus.com/inward/record.url?scp=85077788839&partnerID=8YFLogxK
UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-85077788839&origin=recordpage
U2 - 10.1109/CCTA.2019.8920514
DO - 10.1109/CCTA.2019.8920514
M3 - RGC 32 - Refereed conference paper (with host publication)
SN - 978-1-7281-2768-2
T3 - CCTA - IEEE Conference on Control Technology and Applications
SP - 851
EP - 855
BT - IEEE CCTA 2019
PB - Institute of Electrical and Electronics Engineers
T2 - 3rd IEEE Conference on Control Technology and Applications, IEEE CCTA 2019
Y2 - 19 August 2019 through 21 August 2019
ER -