The modal satisfiability problem is solved either by using a specifically designed algorithm, or by translating tire modal logic formula into air instance of a different class of problem, such as a first-order logic problem, a propositional satisfiability problem, or, more recently, a constraint satisfaction problem. In the latter approach, the modal formula is translated into layered propositional formulae. Each layer is translated into a constraint satisfaction problem which is solved rising a constraint solver. We extend this approach to the modal logics KT and S4 and introduce a range of optimizations of the basic prototype. The results compare favorably with those of other solvers, and support the adoption of constraint programming as implementation platform for modal and other related satisfiability solvers.
