International Summer School on Constraints
in Computational Logic
Building Industrial Applications With Constraint Programming
Helmut Simonis
In this tutorial we will give an overview of real-life applications
developed with constraint logic programming. Constraint logic
programming (CLP) combines declarative logic based programming with
specialized constraint solving methods from artificial intelligence,
Operations Research and mathematics. It allows clear and concise
expression of a wide class of combinatorial problems together with their
efficient solution. In parallel with ongoing research in this field, CLP
is now increasingly used to tackle real world decision making problems.
In a first part of the talk, we will briefly present the methods and
tools used for CLP and describe typical application areas. We introduce
the concepts of global constraints, meta-heuristics and constraint
visualization, which are central to large scale constraint solving..
A second part of the presentation will give a classification of problem
domains suitable for the constraint programming approach and discusses
various large scale applications which have been developed using the
CHIP CLP system. These applications were co-developed by COSYTEC with
specialists in the respective application domain.
-
The TACT system is an integrated transport planning and scheduling
system developed for a large food-industry company in the UK. The system
plans the delivery of raw materials to different processing factories
with a fleet of lorries and a variety of other resources.
-
The OPTI-PLAN application is used by the French TV/Radio company RFO to
plan and control the assignment of hundreds of journalists and
technicians to different emissions. The tool is now used world-wide in 8
installations.
-
The COBRA system is used to develop working plans and schedules for
train drivers and conductors at NWT in the UK. The work plans must cover
several ten-thousand activities in a weekly time table, making sure that
all working rules and regulations are respected.
The last part of the presentation will concentrate on lessons learned
from this application development, showing interesting new research
fields and practical problems in developing state-of-the-art decision
support systems.
ccl99@lri.fr
CCL'99 main page
Last change: Jan 18, 1999.