In this paper, we propose a novel framework for the parallel solution of combinatorial problems based on MODS theory (PARMODS) This framework makes use of metaheuristics based on the Deterministic Swapping (MODS) theory. These approaches represents the feasible solution space of any combinatorial problem through a Deterministic Finite Automata. Some of those methods are the Metaheuristic Of Deterministic Swapping (MODS), the Simulated Annealing Deterministic Swapping (SAMODS), the Simulated Annealing Genetic Swapping (SAGAMODS) and the Evolutionary Deterministic Swapping (EMODS) Those approaches have been utilized in different contexts such as data base optimization, operational research [1—3, 8] and multi-objective optimization. The main idea of this framework is to exploit parallel computation in order to obtain a general view of the feasible solution space of  any combinatorial optimization problem. This is, all the MODS methods are used in a unique general optimization process. In parallel, each instance of MODS explores a different region of the solution space. This allows us to explore distant regions of the feasible solution which could not be explored making use of classical (sequential) MODS implementations. Some experiments are performed making use of well-known TSP instances. Partial results shows that PARMODS provides better solutions than sequential MODS based implementations.