update jsp problem: add gurobi solver for pulp

dothinking · Oct 2, 2022 · 9e6f92b · 9e6f92b
1 parent d00b92e
commit 9e6f92b
Show file tree

Hide file tree

Showing 2 changed files with 39 additions and 18 deletions.
diff --git a/docs/2021-08-08-作业车间调度问题求解框架.md b/docs/2021-08-08-作业车间调度问题求解框架.md
@@ -50,9 +50,9 @@ https://github.com/dothinking/jsp_framework
 
 - [OR-Tools 约束求解器](2021-08-22-作业车间调度问题求解框架：OR-Tools约束求解器.md)
 
-- [PuLP求解器](2021-08-29-作业车间调度问题求解框架：PuLP求解器.md)
+- [PuLP求解器](2021-08-29-作业车间调度问题求解框架：PuLP求解框架.md)
 
-- [基于规则指派](2021-08-28-作业车间调度问题求解框架：基于规则指派求解器.md)
+- [基于规则指派](2021-08-28-作业车间调度问题求解框架：规则指派算法.md)
 
 
 ## 参考文献

diff --git a/docs/2021-08-29-作业车间调度问题求解框架：PuLP求解器.md → docs/2021-08-29-作业车间调度问题求解框架：PuLP求解框架.md b/docs/2021-08-29-作业车间调度问题求解框架：PuLP求解器.md → docs/2021-08-29-作业车间调度问题求解框架：PuLP求解框架.md
@@ -3,12 +3,12 @@ categories: [optimization, mathematics]
 tags: [job shop schedule]
 ---
 
-# 作业车间调度问题求解框架：PuLP 求解器
+# 作业车间调度问题求解框架：PuLP 求解框架
 
 ---
 
 
-本文根据作业车间调度问题的数学描述，利用 Python 整数规划包 PuLP 进行建模和求解，并集成到 `jsp_framework` 框架中。
+本文根据作业车间调度问题的数学描述，利用 Python 整数规划框架 PuLP 进行建模，并分别使用默认的`CBC`求解器及商业求解器`Gurobi`求解。
 
 ## PuLP
 
@@ -165,7 +165,7 @@ class PuLPSolver(JSSolver):
 
 !!! warning "注意"
 
-    当前版本采用了默认的 CBC （COIN-OR Branch-and-Cut）求解器。
+    当前版本支持 CBC（COIN-OR Branch-and-Cut）、SCIP （开源） 和 Gurobi（商业）求解器。其中 CBC 为 Pulp 默认求解器，开箱即用；其余二者需要自行下载安装和授权。
 
 
 ### PuLP 建模
@@ -273,7 +273,8 @@ names = ['ft06', 'la01', 'ft10', 'swv01', 'la38', \
 problems = [JSProblem(benchmark=name) for name in names]
 
 # solver
-solvers = [PuLPSolver(max_time=300)]
+solvers = [PuLPSolver(name=name, solver_name=name, max_time=300) \
+    for name in ('cbc', 'gurobi')]
 
 # solve 
 benchmark = BenchMark(problems=problems, solvers=solvers, num_threads=5)
@@ -286,19 +287,39 @@ benchmark.run(show_info=True)
 +----+---------+--------+---------------+--------------+----------+---------+-------+
 | ID | Problem | Solver | job x machine |   Optimum    | Solution | Error % |  Time |
 +----+---------+--------+---------------+--------------+----------+---------+-------+
-| 1  |   ft06  |  pulp  |     6 x 6     |      55      |   55.0   |   0.0   |  3.5  |
-| 2  |   la01  |  pulp  |     10 x 5    |     666      |  666.0   |   0.0   | 209.0 |
-| 3  |   ft10  |  pulp  |    10 x 10    |     930      |  1034.0  |   11.2  | 301.2 |
-| 4  |  swv01  |  pulp  |    20 x 10    |     1407     |  2941.0  |  109.0  | 301.2 |
-| 5  |   la38  |  pulp  |    15 x 15    |     1196     |  1769.0  |   47.9  | 301.2 |
-| 6  |   ta31  |  pulp  |    30 x 15    |     1764     | 13294.0  |  653.6  | 301.1 |
-| 7  |  swv12  |  pulp  |    50 x 10    | (2972, 3003) |  9527.0  |  218.9  | 309.9 |
-| 8  |   ta42  |  pulp  |    30 x 20    | (1867, 1956) | 23202.0  |  1113.8 | 507.1 |
-| 9  |   ta54  |  pulp  |    50 x 15    |     2839     | 26229.0  |  823.9  | 608.0 |
-| 10 |   ta70  |  pulp  |    50 x 20    |     2995     | 36851.0  |  1130.4 | 625.6 |
+| 1  |   ft06  |  cbc   |     6 x 6     |      55      |   55.0   |   0.0   |  3.5  |
+| 2  |   la01  |  cbc   |     10 x 5    |     666      |  666.0   |   0.0   | 209.0 |
+| 3  |   ft10  |  cbc   |    10 x 10    |     930      |  1034.0  |   11.2  | 301.2 |
+| 4  |  swv01  |  cbc   |    20 x 10    |     1407     |  2941.0  |  109.0  | 301.2 |
+| 5  |   la38  |  cbc   |    15 x 15    |     1196     |  1769.0  |   47.9  | 301.2 |
+| 6  |   ta31  |  cbc   |    30 x 15    |     1764     | 13294.0  |  653.6  | 301.1 |
+| 7  |  swv12  |  cbc   |    50 x 10    | (2972, 3003) |  9527.0  |  218.9  | 309.9 |
+| 8  |   ta42  |  cbc   |    30 x 20    | (1867, 1956) | 23202.0  |  1113.8 | 507.1 |
+| 9  |   ta54  |  cbc   |    50 x 15    |     2839     | 26229.0  |  823.9  | 608.0 |
+| 10 |   ta70  |  cbc   |    50 x 20    |     2995     | 36851.0  |  1130.4 | 625.6 |
 +----+---------+--------+---------------+--------------+----------+---------+-------+
 ```
 
-计算效果并不理想，只有前两个问题即工序总数 50 以内得到了最优解，其余案例误差较大。由此表明，**PuLP 默认的 CBC 求解器在作业车间调度问题上的性能不如 Google OR-Tools 的约束求解器 CpSolver**。
+```
++----+---------+--------+---------------+--------------+----------+---------+-------+
+| ID | Problem | Solver | job x machine |   Optimum    | Solution | Error % |  Time |
++----+---------+--------+---------------+--------------+----------+---------+-------+
+| 1  |   ft06  | gurobi |     6 x 6     |      55      |   55.0   |   0.0   |  0.7  |
+| 2  |   la01  | gurobi |     10 x 5    |     666      |  666.0   |   0.0   |  5.2  |
+| 3  |   ft10  | gurobi |    10 x 10    |     930      |  930.0   |   0.0   |  60.2 |
+| 4  |  swv01  | gurobi |    20 x 10    |     1407     |  1757.0  |   24.9  | 301.4 |
+| 5  |   la38  | gurobi |    15 x 15    |     1196     |  1196.0  |   0.0   | 300.9 |
+| 6  |   ta31  | gurobi |    30 x 15    |     1764     |  2142.0  |   21.4  | 304.5 |
+| 7  |  swv12  | gurobi |    50 x 10    | (2972, 3003) | unsolved |   n.a.  | 308.9 |
+| 8  |   ta42  | gurobi |    30 x 20    | (1867, 1956) |  2457.0  |   28.5  | 362.8 |
+| 9  |   ta54  | gurobi |    50 x 15    |     2839     | unsolved |   n.a.  | 607.6 |
+| 10 |   ta70  | gurobi |    50 x 20    |     2995     |  4513.0  |   50.7  | 608.2 |
++----+---------+--------+---------------+--------------+----------+---------+-------+
+```
+
+
+`cbc`求解器的计算效果并不理想，只有前两个问题即工序总数 50 以内得到了最优解，其余案例误差较大。`Gurobi`的结果略有改进，前5题结果尚可，后5题或者无法求解或者误差较大，整体不如前文的OR-Tools 约束求解器。
+
+由此表明，对于作业车间调度问题，**混合整数规划求解器**（如CBC、Gurobi）的求解性能不如**约束求解器**（如Google OR-Tools 的CpSolver）。
+
 
-PuLP 支持其他求解器例如商业求解器 CPLEX 和 GUROBI，有机会可以对比一下结果。