This paper presents a multicriteria procedure for forest planning problems. The basic ingredients of this procedure are multigoal programming, compromise programming, and results connecting the best-compromise solutions with the concept of utility optima. The main feature of the procedure lies in the easy and transparent utility interpretation of some single best-compromise solutions even when the utility function, as is usual in a forestry context, is virtually unknown. The theoretical developments are applied to a timber harvest scheduling problem in Spain. The criteria considered are: (a) the net present value of the forest over the planning horizon, (b) the equality of harvest volume in each cutting period, (c) the area control criterion which looks for an ending regulated or even-aged forest, and (d) the ending inventory restraint. The preferential weights incorporated into the models are derived from the application of the Analytical Hierarchy Process (AHP) method to a group of experts in forestry matters. Several utility formulations are tested: (1) a separable and additive utility function linear in each criterion (i.e., a compromise solution for metric 1); (2) a Rawlsian utility function where the maximum deviation is minimized (i.e., a compromise solution for metric ∞) and (3) a family of quasi-Rawlsian utility functions. For. Sci. 44(1):47-57.
No Supplementary Data
No Article Media