Go to Qiita Advent Calendar 2024 Top
LoginSignup
TaylorM
@TaylorM (kenkyu taylor)

Are you sure you want to delete the question?

Leaving a resolved question undeleted may help others!

Python エラー討伐してくれませんか...

Q&A

Closed

Pythonerror最適化RobustOptimization

解決したいこと

3カ所の需要地と9箇所の拠点候補で、輸送費や拠点開設費などの費用を最小化しながら、荷物を需要地へ送ることのできる拠点の位置を候補から決定するロバスト最適化問題を解くプログラムですが、拠点の荷物保管上限を減らすとエラーが出ます．

発生している問題・エラー

Model is infeasible or unbounded
Best objective -, best bound -, gap -
Traceback (most recent call last):
  File "C:\Users\OneDrive\hog.py", line 98, in <module>
    SP_obj = SP.ObjVal
             ^^^^^^^^^
  File "src\gurobipy\model.pxi", line 354, in gurobipy.Model.__getattr__
  File "src\gurobipy\model.pxi", line 1914, in gurobipy.Model.getAttr
  File "src\gurobipy\attrutil.pxi", line 100, in gurobipy.__getattr
AttributeError: Unable to retrieve attribute 'ObjVal'. Did you mean: 'ObjNVal'?

該当するソースコード

from gurobipy import *
import numpy as np
# Parameters
Ay = np.array([
              [100,0,0,0,0,0,0,0,0,-1,0,0,0,0,0,0,0,0],
              [0,100,0,0,0,0,0,0,0,0,-1,0,0,0,0,0,0,0],
              [0,0,100,0,0,0,0,0,0,0,0,-1,0,0,0,0,0,0],
              [0,0,0,100,0,0,0,0,0,0,0,0,-1,0,0,0,0,0],
              [0,0,0,0,100,0,0,0,0,0,0,0,0,-1,0,0,0,0],
              [0,0,0,0,0,100,0,0,0,0,0,0,0,0,-1,0,0,0],
              [0,0,0,0,0,0,100,0,0,0,0,0,0,0,0,-1,0,0],
              [0,0,0,0,0,0,0,100,0,0,0,0,0,0,0,0,-1,0],
              [0,0,0,0,0,0,0,0,100,0,0,0,0,0,0,0,0,-1],
              [0,0,0,0,0,0,0,0,0,1,1,1,1,1,1,1,1,1]])

by = np.array([0,0,0,0,0,0,0,0,0,200.])

G = np.array([
       [ 0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.],
       [ 0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.],
       [ 0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.],
       [-1., -1., -1., -1., -1., -1., -1., -1., -1.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.],
       [ 0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0., -1., -1., -1., -1., -1., -1., -1., -1., -1.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.],
       [ 0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  -1.,-1., -1., -1., -1., -1., -1., -1., -1.],
       [ 1.,  0.,  0.,  1.,  0.,  0.,  1.,  0.,  0.,  1.,  0.,  0.,  1.,  0.,  0.,  1.,  0.,  0.,  1.,  0.,  0.,  1.,  0.,  0.,  1.,  0.,  0.],
       [ 0.,  1.,  0.,  0.,  1.,  0.,  0.,  1.,  0.,  0.,  1.,  0.,  0.,  1.,  0.,  0.,  1.,  0.,  0.,  1.,  0.,  0.,  1.,  0.,  0.,  1.,  0.],
       [ 0.,  0.,  1.,  0.,  0.,  1.,  0.,  0.,  1.,  0.,  0.,  1.,  0.,  0.,  1.,  0.,  0.,  1.,  0.,  0.,  1.,  0.,  0.,  1.,  0.,  0.,  1.],
])

E = np.array([[0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0],
              [0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0],
              [0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0],
              [0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0],
              [0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0],
              [0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0],
              [0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0],
              [0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0],
              [0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1],
              
              ])

D = np.array([
       [  0.,   0.,   0.],
       [  0.,   0.,   0.],
       [  0.,   0.,   0.],
       [  0.,   0.,   0.],
       [  0.,   0.,   0.],
       [  0.,   0.,   0.],
       [ 40.,   0.,   0.],
       [  0.,  40.,   0.],
       [  0.,   0.,  40.]])
h = np.array([0.,   0.,   0.,  0.,   0.,   0., 85., 65., 50.])

MP = Model('MP')

# main problem
# 主問題
c = np.array([850, 650, 500, 850, 650, 500, 850, 650, 500])
a = np.array([50, 50, 50, 50, 50, 50, 50, 50, 50])
b = np.array([1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27])

bigM = 10**5
LB = -GRB.INFINITY
UB = GRB.INFINITY
epsilon = 1e-5
k = 1

y = MP.addMVar((9,), obj=c, vtype=GRB.BINARY)
z = MP.addMVar((9,), obj=a, vtype=GRB.CONTINUOUS)
d = MP.addMVar((9,), lb=0, name='d')
eta = MP.addMVar((1,), obj=1, vtype=GRB.CONTINUOUS)

# MP
MP.addConstr(Ay[:, :9]@y+Ay[:, 9:]@z >= by)
MP.optimize()
MP_obj = MP.ObjVal
LB = max(MP_obj, LB)


SP = Model('SP')
x = SP.addMVar((27,), vtype=GRB.CONTINUOUS, name='x')
pi = SP.addMVar(G.shape[0], vtype=GRB.CONTINUOUS, name='pi')
g = SP.addMVar((3,), ub=1, vtype=GRB.CONTINUOUS, name='g')
v = SP.addMVar((G.shape[0],), vtype=GRB.BINARY, name='v')
w = SP.addMVar((G.shape[1],), vtype=GRB.BINARY, name='w')

G1 = SP.addConstr(G@x >= h-D@g-E@np.concatenate([y.x, z.x]), name="G1")
SP.addConstr(G.T@pi <= b, name='pi')
SP.addConstr(pi <= bigM*v, name='v')
G2 = SP.addConstr(
    G@x-h+E@np.concatenate([y.x, z.x])+D@g <= bigM*(1-v), name='G2')
SP.addConstr(x <= bigM*w, name='w1')
SP.addConstr(b-G.T@pi <= bigM*(1-w), name='w2')
SP.addConstr(g[:2].sum() <= 1, name='g1')
SP.addConstr(g.sum() <= 1.5, name='g2')
SP.setObjective(b@x, GRB.MAXIMIZE)
SP.optimize()
SP_obj = SP.ObjVal
UB = min(UB, c@y.x+a@z.x+SP_obj)
MP.reset()
while abs(UB-LB) >= epsilon:
    if SP_obj < GRB.INFINITY:
        MP.reset()
        x_new = MP.addMVar((27,), vtype=GRB.CONTINUOUS)
        MP.addConstr(eta >= b.T@x_new)
        MP.addConstr(E[:, :9]@y+E[:, 9:]@z+G@x_new >= h-D@g.x)
        SP.reset()
        MP.optimize()
        MP_obj = MP.objval
        LB = max(LB, MP_obj)
    else:
        x_new = MP.addMVar((9,), vtype=GRB.CONTINUOUS)
        MP.addConstr(E[:, :9]@y+E[:, 9:]@z+G@x_new >= h-D@g.x)
    # update
    SP.remove(G1)
    SP.remove(G2)
    G1 = SP.addConstr(G@x >= h-D@g-E@np.concatenate([y.x, z.x]), name="G1")
    G2 = SP.addConstr(G@x-h+E@np.concatenate([y.x, z.x])+D@g <= bigM*(1-v), name='G2')
    SP.optimize()
    SP_obj = SP.ObjVal
    UB = min(UB, c@y.x+a@z.x+SP_obj)
    k += 1
    # go back to MP
    # check the solutions
    print("y:", y.x)
    print("z:", z.x)
    print("d:", d.x)
    print("輸送コスト:", eta.x)
    print("x:", x.x)
    print("g:", g.x)
    

    print("繰り返し回数："+str(k))
    print("上界：{}".format(UB))
    print("下界：{}".format(LB))

自分で試したこと

パラメータの次元数や計算の解析

予想されるミス箇所

Ayは拠点の上限を示しますが、200まで増やすとエラーが解消されるため、
おそらくここも問題
本来であれば上限を超えた需要はほかの拠点を開設して補うはず...

x_newが正しく生成されていない点

0

3Answer

pea1205
@pea1205

「モデルは実行不可能であるか、境界がありません」という内容のようです。
以下の内容に沿って実行してみました

SP.optimize()の前後にコードを差し込んでいます

SP.Params.DualReductions = 0
SP.Params.InfUnbdInfo=1
SP.optimize()
print('Status:' , SP.Status)
if SP.status == gp.GRB.INFEASIBLE:
    print("The model is infeasible. Computing IIS...")
    SP.computeIIS()
    print("IIS constraints:")
    for c in SP.getConstrs():
        if c.IISConstr:
            print('%s' % c.constrName)
else:
    print("The model is feasible.")

出力は以下のようになります

IIS computed: 6 constraints, 9 bounds
IIS runtime: 0.06 seconds (0.00 work units)
IIS constraints:
G1[3]
G1[4]
G1[5]
G1[6]
G1[7]
g1
1Like
nak435
@nak435

AttributeError: Unable to retrieve attribute 'ObjVal'.

モデルが実行不可能であるため起きたエラーのようです。
（解が無い？）

↓Gurobiのコミュニティで同じエラーを見つけましたが、それ以上のことは分かりません。

0Like
mochikuni
@mochikuni

該当するソースコードの109行目

   MP_obj = MP.objval

とありますが

   MP_obj = MP.ObjVal

の打ち間違いだったりしますか?

見当違いでしたらごめんなさい。

0Like

Your answer might help someone💌

LoginSign Up