Campus Event Calendar
Event Entry
What and Who
On global capacity constraints in scheduling
Dr. Hendrik C. R. Lock
Universitaet Karlsruhe, Institut fuer Programmstrukturen und Datenorganisationen
DFKI-Kolloquium
AG 1, AG 2, AG 3, INET, AG 4, AG 5, D6, RG1, SWS
AG Audience
English
Note: We use this to send email in the morning.
Date, Time and Location
Tuesday, 2 July 96
14:00
60 Minutes
45 - FB14
015
Saarbrücken
Abstract
In disjunctive and cumulative scheduling, global capacity constraints
achieve stronger consistency among problem variables, and therefore
stronger pruning during search. The idea of such global constraints is
to relate groups of activities, their capacity consumption, and the
capacity provided by the resource over varying time intervals.
achieve stronger consistency among problem variables, and therefore
stronger pruning during search. The idea of such global constraints is
to relate groups of activities, their capacity consumption, and the
capacity provided by the resource over varying time intervals.
>From those grouping new bounds to problem variables can be inferred.
As a result, larger problems (up to
300 activities per resource) can be handled, and optimization problems
can be tackled more effectively.
As a second aspect, I will discuss the impact of selection heuristics
in the branching scheme, and in particular the use of randomized
selection. As a result, the average deviation between best solution
and lower bound was found to be 3.63% for a benchmark suite of 36
medium size problems.
Contact
Ruth Seeger (Sekretariat Gert Smolka)
Christian Schulte (schulte@dfki.uni-sb.de)
Peter Van Roy (vanroy@ps.uni-sb.de)
--email hidden
passcode not visible
logged in users only