Skip to main content
Article
SIMPL: A System for Integrating Optimization Techniques
J.-C. Régin and M. Rueher, eds., Conference on Integration of AI and OR Techniques in Constraint Programming for Combinatorial Optimization Problems (CPAIOR 2004)
  • Ionuţ Aron, Carnegie Mellon University
  • John N. Hooker, Carnegie Mellon University
  • Tallys H. Yunes, Carnegie Mellon University
Date of Original Version
1-1-2004
Type
Book Chapter
Abstract or Description

In recent years, the Constraint Programming (CP) and Operations Research (OR) communities have explored the advantages of combining CP and OR techniques to formulate and solve combinatorial optimization problems. These advantages include a more versatile modeling framework and the ability to combine complementary strengths of the two solution technologies. This research has reached a stage at which further development would benefit from a general-purpose modeling and solution system. We introduce here a system for integrated modeling and solution called SIMPL. Our approach is to view CP and OR techniques as special cases of a single method rather than as separate methods to be combined. This overarching method consists of an infer-relax-restrict cycle in which CP and OR techniques may interact at any stage. We describe the main features of SIMPL and illustrate its usage with examples.

DOI
10.1007/b96957
Citation Information
Ionuţ Aron, John N. Hooker and Tallys H. Yunes. "SIMPL: A System for Integrating Optimization Techniques" J.-C. Régin and M. Rueher, eds., Conference on Integration of AI and OR Techniques in Constraint Programming for Combinatorial Optimization Problems (CPAIOR 2004) Vol. LNCS 3011 (2004) p. 21 - 36
Available at: http://works.bepress.com/jnhooker/48/