Clustering techniques applied to a large number of scenarios under uncertainty allow the selection of a reduced, however, representative set of the total scenario population. In other words, a sample that contains a smaller number of elements to the point of sufficiently reducing the total volume of data and obtaining significant gains in efficiency in processing the data, but that is able, above all, to preserve the characteristics of the stochastic process that originated it. To this end, the present work proposes a methodology for selecting stochastic scenarios using the classic Set Covering Model. Applied in the calculation of demand for tools and services in the construction of offshore oil exploration wells, a sensitivity analysis compares the results of the demands calculated with the scenarios selected by the Set Covering Problem (SCP) and the demand calculated with the universe of scenarios. The SCP was solved, in this application, in its classic version of the literature using an exact algorithm and a heuristic algorithm. The results show not representative differences in the final result of the demands calculated with reduced scenarios and with the total scenarios. The heuristic, even if it is first solution, presented a satisfactory result in relation to the performance gain versus reliability, and indicates the potential of the method if applied together with metaheuristic and local search algorithms.