Reduced Cost Attributes

Taylor Leonard

• August 11, 2019 14:19

Hello, 

I have a restricted master problem that runs just fine.  But I'm not able to access the reduced cost or the shadow price attributes for the problem.  All of the variables are continuous.  I tried accessing the reduced cost by individual variable too (ex. r[0,1] ) but that does not work either.  Any suggestions?  Also, is there a method yet to access the reduced cost for integer variables?  Thank you in advance!

m = Model("Master Problem")
m.Params.OutputFlag=1
m.read("m_param.prm")

m.update()
r = {}
s = {}
z = {}
r = m.addVars(airports,uc,lb=0,ub=1,vtype=GRB.CONTINUOUS,name='r')
s = m.addVars(secmea,airports,uc,lb=0,vtype=GRB.CONTINUOUS,name='s')
z = m.addVars(secmea,airports,uc,lb=0,vtype=GRB.CONTINUOUS,name='z')

objn1 = quicksum(r[k,j]*Rx[int(uniquecombon[j])] for k in range(airports) for j in range(uc))
objn2 = -1*quicksum(
(1-SA[SAP,1])*fr[0]*r[k,j]*qnew1[int(uniquecombon[j])] + (1-SA[SAP,2])*fr[1]*r[k,j]*qnew2[int(uniquecombon[j])]
for k in range(airports) for j in range(uc))
objn3 = quicksum(SL[k,int(uniquecombon[j])]*z[d,k,j] for d in range(secmea) for k in range(airports) for j in range(uc))

m.setObjectiveN(objn1,index=0,priority=0,weight=1)
m.setObjectiveN(objn2,index=1,priority=0,weight=1)
m.setObjectiveN(objn3,index=2,priority=0,weight=1)

c0 = {}
c1 = {}
c2 = {}
c3 = {}
c4 = {}
c5 = {}
for d in range(secmea):
c0[d] = m.addConstr(quicksum(z[d,k,j] for k in range(airports) for j in range(0,uc)) >= E[d],"Existing_Avail_"+str(d))
c1[d] = m.addConstr(quicksum(z[d,k,j] for k in range(airports) for j in range(0,uc)) <= U[d],"Resource_Avail_"+str(d))
for k in range(airports):
c2[k] = m.addConstr(quicksum(r[k,j] for j in range(uc)) == 1, 'Airport_'+str(k))
for d in range(secmea):
c3[d,k] = m.addConstr(quicksum(z[d,k,j] for j in range(uc)) <= quicksum(K[d,k] for j in range(uc)),"linearity1_"+str(d)+str(k))
c4[d,k] = m.addConstr(quicksum(s[d,k,j] - (1-r[k,j])*K[d,k] for j in range(uc)) <= quicksum(z[d,k,j] for j in range(uc)),"linearity2_"+str(d)+str(k))
c5[d,k] = m.addConstr(quicksum(z[d,k,j] for j in range(uc)) <= quicksum(s[d,k,j] for j in range(uc)),"linearity3_"+str(d)+str(k))

m.update
m.optimize()
shadow_price = m.getAttr(GRB.Attr.Pi)
print("Master Complete")
mval = m.ObjVal
r.RC
s.RC
z.RC

 

Thanks, 

Taylor

0

• 

  Thomas Opfer

  • August 12, 2019 07:31

  

  

  Concerning your question about integer variables: There are no (trivial) duals / reduced costs in MIPs. What you might want to do is to fix the integer solution and solve the remaining LP.

  Concerning your other (continous) question: The code seems incomplete, so I cannot test it. What do you mean by "does not work"? Do you get an exception? Do you get zeros (which might be perfectly ok)? For a very simple LP I just tested, the code

  print m.getAttr(GRB.Attr.RC)
  print m.getAttr(GRB.Attr.Pi)

  works.

  0
• 

  Taylor Leonard

  • August 12, 2019 12:59

  

  

  

  Hi Thomas, thanks for the reply.  The error I get when optimizing a standard linear program with continuous variables is 

  AttributeError: 'tupledict' object has no attribute 'RC'

  I understand that zeros are fine, but I did not expect to receive no attribute because I know that the solution is not the optimal one.

  Thank you,

  Taylor

   

  0
• 

  Thomas Opfer

  • August 12, 2019 14:08

  

  

  Ah, I see.

  You are calling:

  r.RC
  s.RC
  z.RC

  I guess, this should be something like:

  r.getAttr(GRB.Attr.RC)
  s.getAttr(GRB.Attr.RC)
  z.getAttr(GRB.Attr.RC)

  0

Please sign in to leave a comment.