We consider a heuristic approach for the solution of a location problem with economies of scale. The method chosen has a strong intuitive appeal, a prominent empirical track record, and is trivial to efficiently implement on parallel processors. We define the various components comprising this GRASP approach and perform a step-by-step development of such a heuristic for the location problem with concave costs. Computational results for problems of dimensions up to 100 × 1000 are reported.