GMCM-2022-B

Our solution to National Post-Graduate Mathematical Contest in Modeling in 2022 of Problem B

Methods

Q1：3H切割方式或者3E切割方式 + Item按面积、长边或者短边排序 + 填充按面积、长边或者短边排序。

Q2：Item总数量和总面积约束的DBSCAN算法进行组批，组批完采用Q1的方法进行排样。

Usage

Just run the main.py to obtain the results, or modify the config settings in main.py and run.

python main.py

Results

Question 1:

表 1 A组数据排样结果表

数据	消耗板材数	板材利用率
dataA1	88	0.949
dataA2	86	0.963
dataA3	87	0.962
dataA4	86	0.952

A1:

less_bins: 88

best_efficiency: 0.949084892868486

best_sets: best_longside-rotated-sorting-SplitSameAsPreviousAxis-DESCSS-0

A2:

less_bins: 86

best_efficiency: 0.9634703138378655

best_sets: best_longside-rotated-sorting-SplitSameAsPreviousAxis-DESCSS-0

A3：

less_bins: 87

best_efficiency: 0.9622196866495535

best_sets: best_area-rotated-sorting-SplitSameAsPreviousAxis-DESCSS-0

A4:

less_bins: 86

best_efficiency: 0.9515846616938433

best_sets: best_shortside-rotated-sorting-SplitSameAsPreviousAxis-DESCSS-0

Question 2:

表 2 组批及排样结果表

数据	批次数	消耗板材数	板材利用率
dataB1	57	3777	0.788
dataB2	44	2506	0.769
dataB3	49	2502	0.772
dataB4	43	2549	0.796
dataB5	74	4030	0.767

B1:

less_bins: 3777

best_efficiency: 0.7883577515382038

best_sets: best_long_side-rotated-sorting-SplitSameAsPreviousAxis-DESCSS-0

B2:

less_bins: 2506

best_efficiency: 0.7687222884909595

best_sets: best_long_side-rotated-sorting-SplitSameAsPreviousAxis-DESCSS-0

B3:

less_bins: 2502

best_efficiency: 0.772718752436911

best_sets: best_long_side-rotated-sorting-SplitSameAsPreviousAxis-DESCSS-0

B4:

less_bins: 2549

best_efficiency: 0.7962636535144704

best_sets: best_area-rotated-sorting-SplitSameAsPreviousAxis-DESCSS-0

B5:

less_bins: 4030

best_efficiency: 0.7665058966345528

best_sets: best_long_side-rotated-sorting-SplitSameAsPreviousAxis-DESCSS-0

Visualization

Examples of visualization results.

@Programming：Minhao Zou

@Modeling and Writing：Yutong Wang、Xiaotong Zhong