GraphPlan-like and SATPLAN-like planners have shown to outperform classical planners for most of the classical planning domains. However, these two propositional approaches do not exhibit good results for large-size problems due to the graph size they have to handle. In this paper we show a new approach for planning as an attempt to combine the advantanges of a graph-based analysis and a partialorder planner. The fast responses obtained in the experimental results show that the application of this technique can report significant benefits in terms of a reduction in search space and that the average performance of this planner is much better: it takes a bit longer to solve some easy problems but it is capable to easily solve large problems.