URGENT!!! presolve, constraint six_one['c1',1]: all variables eliminated, but lower bound = 1 > 0 and also Bad suffix .iis for _con

[N ABDULLAHI]

good day, all
I have try to fixed the problem I earlier post in this group, but
still I am having this particular one
"presolve, constraint six_one['c1',1]:
all variables eliminated, but lower bound = 1 > 0"

so when I try to expand the particular constraint I got this:

subject to six_one['ci','1']:=0;

it means everything on the left of ":=" is 0, which is equal 1.

I have been trying to figure out where the problem lies but, without success.

Please can someone help me out.
the complete model and the data is attached to this mail

Thanks

[Robert Fourer]

The message

presolve, constraint six_one['c1',1]:
all variables eliminated, but lower bound = 1 > 0

is basically telling you that constraint six_one['c1',1] is "infeasible": any solution to the other constraints cannot also satisfy this one. When I expand this infeasible constraint, I see

ampl: expand six_one['c1',1];
subject to six_one['c1',1]:
amt_res_cust['r1','c1',1] = 1;

So this constraint states that the variable amt_res_cust['r1','c1',1] must equal 1. However I also see that the lower and upper bounds on this variable are zero:

ampl: display amt_res_cust['r1','c1',1].lb,amt_res_cust['r1','c1',1].ub;
amt_res_cust['r1','c1',1].lb = 0
amt_res_cust['r1','c1',1].ub = 0

This tells me that presolve has found, by an analysis of other variable bounds and of constraints, that amt_res_cust['r1','c1',1] can only take the value 0 -- which implies that the constraint six_one['c1',1] saying that amt_res_cust['r1','c1',1] = 1 cannot be satisfied.

You will need to figure out why amt_res_cust['r1','c1',1] can only have the value 0, by looking at the other constraints that affect this variable. If you want to use a solver to find an irreducible infeasible subset (IIS), then type "solve;" a second type and the problem will be sent to the solver even though presolve has already found it to be infeasible.

Bob Fourer
am...@googlegroups.com

[Nasiru Abdullahi]

Thanks Bob,  let me check and I would be right back if I have some issues again

--
You received this message because you are subscribed to the Google Groups "AMPL Modeling Language" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ampl+unsubscribe@googlegroups.com.
To post to this group, send email to am...@googlegroups.com.
Visit this group at https://groups.google.com/group/ampl.
For more options, visit https://groups.google.com/d/optout.