help-glpk
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Help-glpk] How TO interpret information from results


From: Andrew Makhorin
Subject: Re: [Help-glpk] How TO interpret information from results
Date: Wed, 15 Mar 2006 19:58:31 +0300

>  I would like to ask that from the result output
> of the model file for example  shown below i 
> would like to get information that what exactly is 
> what 
>   1. What does "St" stand for n what are "B" "NL"
>       denoted.
>   2. What does it mean when the marginal value is "positive" and 
>      "esp" and what if its "Negative value"
> Problem:    tspmatrixfin
> Rows:       71
> Columns:    191
> Non-zeros:  850
> Status:     OPTIMAL
> Objective:  path = 4002.331 (MINimum)
> 
>    No.   Row name   St   Activity     Lower bound   Upper bound    Marginal
> ------ ------------ -- ------------- ------------- -------------
> -------------
>      1 path              B        4002.33                             
>      2 Travelled_distance[0]
>                             NL             2             2                 
> 263.609
>      3 Travelled_distance[1]
>                             NL             2             3                 
> esp
>      4 Travelled_distance[2]
>                             NL             2             2                 
> -83.12

For rows:

St = B means the corresponding constraint is inactive, i.e. its
auxiliary variable is basic.

St = NL means the corresponding constraint is active, i.e. its
auxiliary variable is non-basic (on the lower bound).

If marginal value is d > 0, the corresponding constraint is active
and increasing its lower bound by one will lead to increasing the
objective function by +d (until the basis will have been changed).
Analogously, if d < 0, the corresponding constraint is active and
decreasing its upper bound by one will lead to increasing the
objective by -d.

If marginal value is printed as eps, this means that its value is
close to zero, i.e. the corresponding solution is degenerative.

P.S. I recommend you to consult any textbook on linear programming.

Andrew Makhorin





reply via email to

[Prev in Thread] Current Thread [Next in Thread]