A0699
Title: An efficient global optimization method for sequential order-of-addition experiments
Authors: Jianbin Chen - Beijing Institute of Technology (China) [presenting]
Abstract: The order-of-addition (OofA) experiment has received a great deal of attention in the recent literature. The primary goal of the OofA experiment is to identify the optimal order in a sequence of m components. Existing methods require pre-specifying a model and cannot flexibly adjust the model according to the progress of the experiment. Moreover, these methods are not applicable to deal with large m (e.g., m 7). With this in mind, this article proposes an efficient global optimization algorithm based on the sequential experiment and variable selection, utilizing information gained during the iterative process. Moreover, an efficient construction method for the optimal initial design, which achieves space-filling and balance properties for OofA components, is proposed. Theoretical supports are given to illustrate the effectiveness of the proposed method. The proposed method is able to obtain the optimal order for large m efficiently. Numerical experiments are used to demonstrate the effectiveness of the proposed method.
