目录
2.排样问题(nesting problem)又称为下料问题(cutting and stock problems) 或填充问题(packing problem),其目标是在材料切割过程中寻找一个较高的材料利用率。排样问题属于经典的np-hard问题,其时间复杂度随着问题规模的增加迅速上升,难以在合理时间内精确求解大规模实例。
一篇专利
雏鸟PRO17幼儿隐藏入口|波多野结衣的电影|不用付费就可以看亏亏网站 - 上海友图科技有限公司
GitHub - mses-bly/2D-Bin-Packing: Library to solve 2D bin packing problems with irregular pieces.
GitHub - eourm20/2d_bin_packing: exact method
接下来分别详细解读一篇RL-based learning to cut的文章和一篇GNN-based end-to-end learning to solve MILP的文章。
现在研究强化学习+组合优化的paper不少了(几十篇+),但方法似乎就这么几种,对此您怎么看? - 知乎
以下参考:
深度强化学习求解组合优化问题(路径、调度问题);DRL for OR/COR - 知乎
1、Solve routing problems with a residual edge-graph attention neural network ; 文章链接:https://www.sciencedirect.com/science/article/pii/S092523122200978X?; 开源代码地址:GitHub -?Lei-Kun/DRL-and-graph-neural-network-for-routing-problems
2、A Multi-action Deep Reinforcement Learning Framework for Flexible Job-shop Scheduling Problem ; 文章链接:https://www.sciencedirect.com/science/article/pii/S0957417422010624;?开源代码地址:https://github.com/Lei-Kun/End-to-end-DRL-for-FJSP?;?https://github.com/Lei-Kun/Dispatching-rules-for-FJSP