Re: sorted constraint

From: Joachim Schimpf <j.schimpf_at_icparc.ic.ac.uk>
Date: Tue 26 Aug 2003 03:59:08 PM GMT
Message-ID: <3F4B83CC.1A2DDE65@icparc.ic.ac.uk>
> I have a coule of questiona about the sorted constraint described in
> http://www.icparc.ic.ac.uk/eclipse/doc/doc/bips/lib/fd_global/sorted-2.html
> 
> (1) What level of consistency is achieved?

The idea was to have bounds-consistency.

> (2) Is there any reference to its propagation algorithm?

There is no separate publication about it. It is roughly like this:

The constraint wakes up whenever any bounds changed, and propagates
these bound changes. This is done in two passes, one along ascending
minima, and one along descending maxima of the unsorted variables.
The complexity of each propagation step is dominated by the sorting
and therefore N*log(N).


-- 
 Joachim Schimpf              /             phone: +44 20 7594 8187
 IC-Parc                     /      mailto:J.Schimpf@imperial.ac.uk
 Imperial College London    /    http://www.icparc.ic.ac.uk/eclipse
Received on Tue Aug 26 17:02:18 2003

This archive was generated by hypermail 2.1.8 : Wed 16 Nov 2005 06:07:25 PM GMT GMT