Simple constraint programming solver

It depends on what are you trying to do.

1- If you want to get started with Constraint Programming (CP) tools, you need first to understand modelling in this approach. For that, you have be able to define a given problem as a set of constraints then you post your model to a solver (that already knows how to deal with these constraints).

I personally find the G12_MiniZinc_Distribution (available via http://www.minizinc.org/) and its latest tutorial a very good way to start (the latest distribution cames with a lot of models). There is also online models if you want to see how a CP model looks like (w.r.t minizinc syntax) : http://www.hakank.org/minizinc.

2- If you have an advanced theoretical understanding of CP, you may directly see the implementation of some techniques (e.g. propagators) from available libraries since developing a CP Solver from scratch is (IMO) a foolish idea. These are some open-source solvers/libraries : G12, Choco, Mistral, Or-Tools.


FWIW, the OptaPlanner (open source, java) documentation has a in depth explanation of a few algorithms, such as Simulated Annealing and Tabu Search.

Here's an image from there:

Simulated Annealing