Campus Event Calendar
Event Entry
New for: D1, D2
What and Who
Challenges and Innovations in Industrial Constraint
Dr. Ulrich Junker
ILOG SA
Talk
AG 1, AG 2
AG Audience
Note: We use this to send email in the morning.
Date, Time and Location
Friday, 28 November 97
14:00
-- Not specified --
43.1 - DFKI
Turing (+1.01)
Saarbrücken
Abstract
Recent years have shown a rapid progress in optimization methods
such as column generation, simulated annealing, genetic algorithms,
iterated improvement techniques, and constraint programming. However,
solving real-world problems successfully does not only require
powerful optimizers, but also powerful modelling capabilities.
Constraint programming (CP) provides a general framework that allows
to increase modelling and problem solving capabilities depending on
the needs of a problem. (Global) constraints for complex mathematical
structures (e.g. paths, trees) can easily be integrated into the
basic propagation mechanism. The chronological search procedure
can treat arbitrary search programs. The declarativity of constraint
programs, furthermore, facilities the decomposition of the complete
problem into subproblems. Different subproblems can then be solved
by different optimization method. This method of combining CP with
the other techniques mentioned above has been crucial for solving
large problems in the areas of production scheduling, air
such as column generation, simulated annealing, genetic algorithms,
iterated improvement techniques, and constraint programming. However,
solving real-world problems successfully does not only require
powerful optimizers, but also powerful modelling capabilities.
Constraint programming (CP) provides a general framework that allows
to increase modelling and problem solving capabilities depending on
the needs of a problem. (Global) constraints for complex mathematical
structures (e.g. paths, trees) can easily be integrated into the
basic propagation mechanism. The chronological search procedure
can treat arbitrary search programs. The declarativity of constraint
programs, furthermore, facilities the decomposition of the complete
problem into subproblems. Different subproblems can then be solved
by different optimization method. This method of combining CP with
the other techniques mentioned above has been crucial for solving
large problems in the areas of production scheduling, air
traffic control, crew scheduling, configuration, and so on. On
the one hand,
we profit from the modelling capacities of CP and its flexibility
in programming search. On the other hand, we solve subproblems by
state-of-the-art algorithms and we control the size of subproblems
solved by CP.
Contact
Christian Schulte
302-5340
--email hidden
passcode not visible
logged in users only
Uwe Brahm, 04/12/2007 12:01 -- Created document.