META-HEURISTIC CLONAL SELECTION ALGORITHM FOR OPTIMIZATION OF FOREST PLANNING

Abstract

ABSTRACT It is important to evaluate the application of new technologies in the field of computational science to forest science. The goal of this study was to test a different kind of metaheuristic, namely Clonal Selection Algorithm, in a forest planning problem. In this problem, the total management area is 4.210 ha that is distributed in 120 stands in ages between 1 and 6 years and site indexes of 22 m to 31 m. The problem was modeled considering the maximization of the net present value subject to the constraints: annual harvested volume between 140,000 m3 and 160,000 m3, harvest ages equal to 5, 6 or 7 years, and the impossibility of division of the management unity at harvest time. Different settings for Clonal Selection Algorithm were evaluated to include: varying selection, cloning, hypermutation, and replacement rates beyond the size of the initial population. A generation value equal to 100 was considered as a stopping criteria and 30 repetitions were performed for each setting. The results were compared to those obtained from integer linear programming and linear programming. The integer linear programming, considered to be the best solution, was obtained after 1 hour of processing. The best setting for Clonal Selection Algorithm was 80 individuals in the initial population and selection. Cloning, hypermutation, and replacement rates equal to 0.20, 0.80, 0.20 and 0.50, respectively, were found. The results obtained by Clonal Selection Algorithm were 1.69% better than the integer linear programming and 4.35% worse than the linear programming. It is possible to conclude that the presented metaheuristic can be used in the resolution of forest scheduling problems.