[SCIP] propagation with set partitioning constraints

[James Cussens]

Hi folks,

Suppose I have two set partitioning constraints:

a + b + c + d = 1
a + b + e = 1

Suppose I were to branch on "e" and let's consider the branch where e=0
So locally we have a+b=1 and so both c and d can be set to 0 (even though we don't know yet which of a or b will be 1).

Does this propagation happen in SCIP? From my reading of the source code (and Achterberg's thesis) it seems that not. But perhaps I have missed something.

(And let's also assume that the relevant rows are not in the LP - I have posted the constraints with initial (and separating) to FALSE. I'm interested in whether domain propagation in the setppc constraint handler will do the job.)

James

James Cussens
Dept of Computer Science, University of Bristol
https://jcussens.github.io/
Funded PhDs available in Bristol in the following areas: Data Science<http://www.bristol.ac.uk/cdt/compass/>, Interactive AI<http://www.bristol.ac.uk/cdt/interactive-ai/>, Cyber Security<http://www.bristol.ac.uk/cdt/cyber-security/> or Digital Health<http://www.bristol.ac.uk/cdt/digital-health/>.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.zib.de/pipermail/scip/attachments/20210128/1f547815/attachment.html>