摘要
列生成算法是一种用于解决线性规划问题的高效算法,其核心思想是通过逐步添加列来构建松弛问题的最优解。然而,随着问题规模的增加,算法的计算复杂度也会急剧增加,导致算法的效率下降。为了优化这一问题,经过对算法改进策略的研究,分析其优缺点,并给出基于机器学习的列生成算法改进模型。该策略的应用可以大大减少算法的计算时间和空间消耗,提高算法的求解效率和精度。
The column generation algorithm is an efficient algorithm for solving linear programming problems,whose core idea is to construct the optimal solution of the relaxed problem by gradually adding columns.However,as the size of the problem increases,the computational complexity of the algorithm also increases sharply,leading to a decrease in its efficiency.To optimize this problem,after researching the improvement strategies of the algorithm,analyzing its advantages and disadvantages,a column generation algorithm improvement model based on machine learning is proposed.The application of this strategy can greatly reduce the computation time and space consumption of the algorithm,improve its solving efficiency and accuracy.
作者
罗凤娥
张鑫
赵强
杨思瀚
Luo Feng’e;Zhang Xin;Zhao Qiang;Yang Sihan(School of Air Traffic Management,Civil Aviation Flight University of China,Guanghan 618307,China)
出处
《现代计算机》
2023年第11期56-59,共4页
Modern Computer
关键词
列生成
改进策略
图神经网络
column generation
improvement strategies
graph neural network
