[Help-glpk] glp_intopt runs forever

[Dušan Plavák]

Hi guys,

I would like to ask you for a help with one lp problem. We are using version 4.61 with the following code:

glp_write_lp(lp, 0, "problem.lp");

glp_iocp parm;

glp_init_iocp(&parm);

parm.msg_lev = GLP_MSG_OFF;

parm.presolve = GLP_ON;

glp_intopt(lp, &parm);

and glp_intopt() call runs "forever". I was reading docs and found note there:

Note that the GLPK branch-and-cut solver is not perfect, so it is unable to solve hard or very large scale MIP instances for a reasonable time.

The lp problem on which it freezes is following:

\* Problem: amounts *\

Minimize

 obj: + 0.03 x_6 + 0.0375 x_7 + 0.1 x_8 + 0.0333333333333333 x_9

Subject To

 r_1: + 9.281400000008 x_5 + 2.548 x_4 + 0.17905 x_3 + 2.9625 x_2

 + 1.0491 x_1 - ~r_1 = -80000

 r_2: + 9.2814 x_5 + 0.744 x_4 + 0.02325 x_3 + 0.34875 x_2 + 0.0651 x_1

 - ~r_2 = -32000

 r_3: + 4e-12 x_5 + 0.369 x_4 + 0.10291 x_3 + 1.99875 x_2 + 0.0082 x_1

 - ~r_3 = -12000

 r_4: + 4e-12 x_5 + 1.435 x_4 + 0.05289 x_3 + 0.615 x_2 + 0.9758 x_1

 - ~r_4 = -36000

 r_5: + x_5 + 10 x_4 + x_3 + 18.75 x_2 + x_1 <= 800

 r_6: + x_6 + 9.281400000008 x_5 + 2.548 x_4 + 0.17905 x_3 + 2.9625 x_2

 + 1.0491 x_1 >= 400

 r_7: - x_6 + 9.281400000008 x_5 + 2.548 x_4 + 0.17905 x_3 + 2.9625 x_2

 + 1.0491 x_1 <= 400

 r_8: + x_7 + 9.2814 x_5 + 0.744 x_4 + 0.02325 x_3 + 0.34875 x_2

 + 0.0651 x_1 >= 160

 r_9: - x_7 + 9.2814 x_5 + 0.744 x_4 + 0.02325 x_3 + 0.34875 x_2

 + 0.0651 x_1 <= 160

 r_10: + x_8 + 4e-12 x_5 + 0.369 x_4 + 0.10291 x_3 + 1.99875 x_2

 + 0.0082 x_1 >= 60

 r_11: - x_8 + 4e-12 x_5 + 0.369 x_4 + 0.10291 x_3 + 1.99875 x_2

 + 0.0082 x_1 <= 60

 r_12: + x_9 + 4e-12 x_5 + 1.435 x_4 + 0.05289 x_3 + 0.615 x_2

 + 0.9758 x_1 >= 180

 r_13: - x_9 + 4e-12 x_5 + 1.435 x_4 + 0.05289 x_3 + 0.615 x_2

 + 0.9758 x_1 <= 180

Bounds

 0 <= ~r_1 <= 160000

 0 <= ~r_2 <= 64000

 0 <= ~r_3 <= 24000

 0 <= ~r_4 <= 72000

 100 <= x_1 <= 400

 8 <= x_2 <= 16

 2 <= x_3 <= 400

 4 <= x_4 <= 20

 4 <= x_5 <= 30

Generals

 x_1

 x_2

 x_3

 x_4

 x_5

End

It does not seems to be large scale, and I am not sure if this is classified as hard problem? Also the interesting part is that on the older version of glpk it was solved without any problems.

Thanks for help.

--

S pozdravom Dušan Plavák