The objective of this work was to find systems of agricultural production that optimize the resources of familiar farmers of the North and northwest regions of the Rio de Janeiro, using the linear programming method. Farmers of the PRONAF groups (A, C and D) were interviewed, what allowed obtaining the information about resources endowment, production costs and gross margins. When the price risks were not included in the model, the guava was the best choice for the optimization of resources and income generation. When considered the price risks, through the MOTAD model, it was concluded that production systems involving the simultaneous culture of passion fruit and guava, mainly, can reduce the risks, without causing a great loss of the income. It was evidenced, besides, that the PRONAF resources contributed to raise the use of the land, to increase the return of the activities and for the rural employment.
family farming; linear programming; PRONAF
