Our model is working but infeasible !
Answeredhello everyone,
This model that we develop is working but infeasbile. We can't finde the probleme where it is!
it is in the code or it is in the constraints or in the number that we used ?
thank you so much
this is the code in python
import sys
import gurobipy as gp
from gurobipy import GRB
# Parameters
N= 6 #tüm görevlerin kümesi
n=[0, 2, 2, 4, 4, 4] #i görevi için gerçekleşecek kasa besleme sayısı
R= 16 #maksimum rota sayısı
Q= 3 #trenin taşıyabileceği maksimum kasa sayısı
M = 2000000 #büyük bir sayı
u = [0, 48, 150, 40, 40, 40] #maksimum parça seviyesi
v = [0, 3, 5, 2, 2, 10] #minimum parça seviyesi
c = [0, 74, 36, 126, 126, 1] #parça çevrim süresi (sn)
e =[[0, 0, 0, 0],
[3552, 7104], #H2 in 1. referansı
[5400, 10800], #H2 in 2. referansı
[5040, 10080, 15120, 20160], #H4 in 1. referansı #c*u
[5040, 10080, 15120, 20160], #H4 in 2. referansı
[40, 80, 120, 160]] #H4 nin 3. referansı
#k. i görevi için kasanın en erken teslim zamanı
d =[[0, 0, 0, 0],
[12600, 25200], #H2 in 1. referansı
[12600, 25200], #H2 in 2. referansı
[6300, 12600, 18900, 25200], #H4 in 1. referansı
[6300, 12600, 18900, 25200], #H4 in 2. referansı
[6300, 12600, 18900, 25200]] #H4 nin 3. referansı
#k. i görevi için kasanın en geç teslim zamanı
p= [0, 3330, 5220, 4788, 4788, 30] #i görevinin süresi
w= [240, 15, 15, 15, 35, 35] #i görevinin kasa başına çalışma süresi
t= [[0, 45, 45, 45, 45, 55],
[45, 0, 0, 15, 15, 25],
[45, 0, 0, 15, 15, 25],
[45, 15, 15, 0, 0, 10],
[45, 15, 15, 0, 0, 10],
[55, 25, 15, 10, 10, 0]]
#trenin görev i konumundan görev j konumuna seyahat süresi
#Create a new model
model = gp.Model("csknz")
# Variables
x = model.addVars(N,4,R,N,4, vtype=GRB.BINARY, name="x")
y = model.addVars(N,4, name="y")
s = model.addVars(N,4, name="s")
for j in range(N):
for l in range(1, n[j]):
for r in range(R):
for i in range(N):
for k in range(1, n[i]):
x[j,l,r,i,k] = model.addVar(vtype=GRB.BINARY)
for i in range(N):
for k in range(1, n[i]):
s[i,k] = model.addVar()
for j in range(N):
for l in range(1, n[j]):
y[j,l] = model.addVar()
obj = gp.quicksum(t[i][j] * x[j,l,r,i,k]
for i in range(N)
for k in range(1, n[i])
for j in range(N)
for l in range(1, n[j])
for r in range(R))
model.setObjective(obj, GRB.MINIMIZE)
#addConstraints
# for i in range(N):
# if i != 0:
# model.addConstr(p[i] == (u[i]-v[i]) * c[i]) # cons.2
# for i in range(N):
# if i != 0:
# for k in range(1, n[i]):
# model.addConstr(e[i][k]+1 == e[i][k] + p[i]) # cons.3
# for i in range(N):
# if i != 0:
# for k in range(1, n[i]):
# model.addConstr(d[i][k] == e[i][k] + (v[i]- 0) * c[i]) # cons.4
# model.addConstr(x[3,1,0,0,1]==1)
# model.addConstr(x[4,1,0,3,1]==1)
for i in range(N):
if i != 0:
for k in range(1, n[i]):
model.addConstr(e[i][k] <= s[i,k]) # cons.5
model.addConstr(s[i,k] <= d[i][k])
model.addConstr(gp.quicksum(x[j,l,r,i,k] for j in range(N) if j != 0
for l in range(1,n[j]))==1) # cons.6
model.addConstr(gp.quicksum(x[j,l,r,i,k]for j in range(N) if j != 0
for l in range(1,n[j])
for r in range(R))<=1) # cons.7
model.addConstrs((gp.quicksum(x[i,k,r,i,k] for i in range(N)
for k in range(1,n[i])) == 0
for r in range(R))) # cons.8
for i in range(N):
if i != 0:
for k in range(1,n[i]):
model.addConstr((gp.quicksum(x[j,l,r,i,k] for j in range(N)
for l in range(1,n[j])
for r in range(R)) == 1)) # cons.9
for j in range(N):
if j != 0 :
for l in range(1,n[j]):
model.addConstr((gp.quicksum(x[j,l,r,i,k] for i in range(N)
for k in range(1,n[i])
for r in range(R)) == 1)) # cons.10
model.addConstrs((gp.quicksum(x[j,l,r,i,k]for i in range(N)
for k in range(1,n[i])
for j in range(N) if j != 0
for l in range(1,n[j]))<= Q for r in range(R))) #cons.11
for i in range(N):
for j in range(N):
for k in range(1,n[i]):
for l in range(1,n[j]):
model.addConstr(s[i,k]+(w[i]+t[i][j]*gp.quicksum(x[j,l,r,i,k]for r in range(R)))-
M*(1-gp.quicksum(x[j,l,r,i,k] for r in range(R)))+(y[j,l]-y[i,k])*(t[i][0]+w[0]+t[0][j]-t[i][j]) <= s[j,l])
# #cons.12
for i in range(N):
if i != 0 :
for l in range(1,n[j]):
model.addConstr((gp.quicksum(r*x[j,l,r,i,k] for i in range(N)
for k in range(1,n[i])
for r in range(R))==y[j,l])) #cons.13
for i in range(N):
for j in range(N):
for k in range(1,n[i]):
for l in range(1,n[j]):
model.addConstr(y[j,l] >= (y[i,k]- M*(1-gp.quicksum(x[j,l,r,i,k] for r in range(R))))) #cons.14
model.optimize()
status=model.status
-
Official comment
This post is more than two years old. Some information may not be up to date. For current information, please check the Gurobi Documentation or Knowledge Base. If you need more help, please create a new post in the community forum, or try Gurobot, our chatbot interface offering instant, expert-level support. -
Hi Hamza,
Please have a look at the article "How do I determine why my model is infeasible?".
The first step is to compute an IIS using the Model.computeIIS() method. The Gurobi Optimizer first needs to prove the model is infeasible before proceeding to finding a sub-system of constraints/bounds which are responsible for infeasibility. Given my experimentation of your model, it seems it takes time to prove infeasibility. How long did it take for the Gurobi to terminate with an infeasible status on your machine?
Instead of figuring out why the model is infeasible, another strategy is to figure out which changes should be done to the model to recover feasibility. To do this, you can use the Model.feasRelaxS() to find the smallest perturbation needed to recover feasibility. Please have a look at the article "How do I change variable and/or constraint bounds to make an infeasible model feasible using feasRelax?" to know how to interpret the solution given by the feasibility relaxation model.
Best regards,
Maliheh
1
Post is closed for comments.
Comments
2 comments