Using CP propagation as preprocessor

When the constraints in a CP model are posted to the solver, they often immediately trigger some reductions to the domains of the involved variables. In certain (easy) cases, it may even happen that the constraint propagation is sufficient to obtain a solution without having to start an enumeration.

The domain reductions obtained through the constraint propagation can be passed on to an LP or MIP model, thus replacing, or re-inforcing the preprocessing algorithms that are used by LP and MIP solvers.

The example description in the following sections is taken from Section 7.1 of the book `Applications of optimization with Xpress-MP'.