[SCIP] SCIP gives the negative dual when a <= constraint is active at rhs

[Marco Lübbecke]

Hi Liding Xu,

adding to what Ambros said:


>    I am using scip to implement a branch and price algorithm for unsplittable multi-commodity flow problem:
>         min c^T x       (1)
>               Ax <= 1,   (2)
>               Ex = 1,     (3)
>               x is binary.
>
> ...
>
> In my knowledge, the dual for <= constraint can never be negative!
>
>


To the best of my knowledge (confirmed by a google picture search on
"dual linear program table"), when the primal is a minimzation
problem, the dual of a "<=" constraint is always non-positive.

All the best, and happy pricing, Marco





-- 
Prof. Dr. Marco Lübbecke
RWTH Aachen University
Chair of Operations Research
Kackertstrasse 7
D-52072 Aachen
Germany

fon / fax: +49 241 80-93362 / 92369
marco.luebbecke at rwth-aachen.de
www.or.rwth-aachen.de/luebbecke