Re: Two questions regarding lib(fd).

From: Ole Tranberg <proj15_at_imm.dtu.dk>
Date: Tue 10 Sep 2002 08:18:37 AM GMT
Message-Id: <200209100818.KAA25401@serv1.imm.dtu.dk>
Joachim Schimpf wrote:

>Ole Tranberg wrote:
>> 
>> I have looked at the "roster.ecl"-example from the ic-parc website. I have tried
>> to understand what is going on but there is one confusing clause.
>> 
>> >From the roster.ecl:
>> 
>>     try_day_off(X) :- X #= r of shift.
>>     try_day_off(X) :- X #\= r of shift.
>> 
>> Why does this give sense?
>
>Declaratively this is useless, of course. It just says
>"X is equal to r, or not".
>
>However, this predicate is used in the search part of the
>program, and here the operational behaviour is important:
>It _first_ tries to set X to 'r of shift', and only if
>this does not lead to a complete solution of the problem,
>it tries the other alternative.

Thank You, now it makes sense :-)

But now I got a new question regarding the "roster.ecl"-example. It is about the search 
strategies.

The method used looks like this:

min_max((search(Strategy,...),labeling_ff(Vars)), Cost, 0, 1000, 0, 60).

In the documentation:

min_max(?Goal, ?C, +Lower, +Upper, +Percent, +Timeout)


So "?Goal" is "(search(Strategy,...),labeling_ff(Vars))".

What does this mean? I got the point of the search strategies (or at least most of it) and I can 
see what is going on in labeling_ff/1. But how do these two work together?
Is the search strategy some kind of a "help" to the labeling_ff/1?

Ole Tranberg
Received on Tue Sep 10 11:24:36 2002

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