作者: Zhen He , Peng-Fei Zhu , Sung-Hyun Park
DOI: 10.1016/J.EJOR.2012.03.009
关键词: Robust optimization 、 Confidence interval 、 Desirability function 、 Genetic algorithm 、 Mathematical optimization 、 Multi response 、 Robustness (computer science) 、 Computer science
摘要: Abstract A robust desirability function approach to simultaneously optimizing multiple responses is proposed. The considers the uncertainty associated with fitted response surface model. uniqueness of proposed method that it takes account all values in confidence interval rather than a single predicted value for each and then defines robustness measure traditional using worst case strategy. hybrid genetic algorithm developed find optima. presented compared its conventional counterpart through an illustrated example from literature.