Section: New Results

Rule-based Modeling Language for Constraint Programming

Participants : François Fages, Raphaël Martin, Thierry Martinez, Sylvain Soliman.

Rules2CP is a rule-based modeling language which allows easy modeling of constraint satisfaction problems, together with specifications for search strategies and heuristic choice criterias by pattern matching. in [23] , we study a new compilation scheme for Rules2CP which allows us to deal with dynamic ordering criteria and to generate procedural constraint programming code instead of flattened constraints. The comparison with the static expansion of Rules2CP models shows that the overhead at runtime is limited, with a gain in the size of the generated program which could be exponentially larger by static expansion.

The language Rules2CP is currently extended to deal with hybrid discrete and continuous domains and packing problems with complex shapes, in the framework of the ANR Net-WMS-2 project with KLS-Optim and EMN Inria EPI TASC. The compiler of Rules2CP is currently rewritten in Java in the framework of a collaboration with KLS-Optim supported by Inria DTI.