摘要
寻求高效算法是最优混料试验设计所研究目标之一。然而现有的算法往往依赖于高级的数学规划求解器。本文基于D-最优等价性定理,提出了一种适用于混料试验的有效算法。同时,通过实例,验证了算法的有效性,且与已有的CO算法,MA算法,VDM算法以及理论结果相比,所提算法精度高,收敛性快。
Finding an efficient algorithm is an objective in optimal design for mixture experiments.However,the available algorithms commonly rely on advanced mathematical programming solvers.This paper proposes an efficient algorithm for mixture experiments based on the equivalence theorem of D-optimality.Moreover,compared with the available algorithms,such as the CO algorithm,MA algorithm,VDM algorithm,and theoretical results of D-optimality for mixture experiments,it is shown that the proposed algorithm has high precision and fast convergence by examples.
作者
许秀钿
李俊鹏
张新风
张崇岐
XU Xiu-tian;LI Jun-peng;ZHANG Xin-feng;ZHANG Chong-qi(School of Economics and Statistics,Guangzhou University,Guangzhou 510006,China;School of Science,Kaili University,Kaili 55601l,China)
出处
《数理统计与管理》
北大核心
2025年第4期650-660,共11页
Journal of Applied Statistics and Management
基金
国家自然科学基金(12071096)
凯里学院博士启动专项(BS202502028)
广州大学专业学位研究生课程案例库建设项目。
