Abstract:  Constraint Programming (CP) provides a powerful and intuitive framework for modeling combinatorial optimization problems such as Vehicle Routing Problems (VRPs). However, classical CP models based on successor variables are not well suited to handling optional visits or insertion-based search strategies, which are central to many state-of-the-art routing heuristics. In this talk, I introduce sequence variables, a modeling abstraction designed to better capture ordered structures such as routes. Sequence variables naturally support optional visits and insertion-based heuristics, including insertion-driven Large Neighborhood Search. I will present their formal definition, discuss the associated domain and update operations, and outline how they can be efficiently integrated into existing trail-based CP solvers. I will also introduce dedicated global constraints for routing problems. This work is in collaboration with Augustin Delecluse and Pascal Van Hentenryck.

 

 

Bio: Pierre Schaus is a Professor of Computer Science at UCLouvain. He obtained his Ph.D. in 2009 under the supervision of Yves Deville, working on global constraints and bin packing in Constraint Programming. After research stays at Brown Universityunder the supervision of Pascal Van Hentenryck and industrial experience at Dynadec and N-SIDE in Belgium, he returned to UCLouvain in 2012 as a professor. He has contributed to the development of several open-source CP solvers, including OscaR, MiniCP, and MaxiCP. His recent research focuses on decision-diagram-based optimization and exact learning algorithms.