%0 Research article
%T Application of ant colony optimization for the risk management of wind damage in forest planning
%A Zeng, Hongcheng
%A Pukkala, Timo
%A Peltola, Heli
%A Kellomäki, Seppo
%D 2007
%J Silva Fennica
%V 41
%N 2
%R doi:10.14214/sf.299
%U https://silvafennica.fi/article/299
%X Ant colony optimization (ACO) is still quite a new technique and seldom  used in the field of forest planning compared to other heuristics such  as simulated annealing and genetic algorithms. This work was aimed at  evaluating the suitability of ACO for optimizing the clear-cut patterns  of a forest landscape when aiming at simultaneously minimizing the risk  of wind damage and maintaining sustainable and even flow of periodical  harvests. For this purpose, the ACO was first revised and the algorithm  was coded using the Visual Basic Application of the ArcGIS software.  Thereafter, the performance of the modified ACO was demonstrated in a  forest located in central Finland using a 30-year planning period. Its  performance was compared to simulated annealing and a genetic algorithm.  The revised ACO performed logically since the objective function value  was improving and the algorithm was converging during the optimization  process. The solutions maintained a quite even periodical harvesting  timber while minimizing the risk of wind damage. Implementing the  solution would result in smooth landscape in terms of stand height after  the 30-year planning period. The algorithm is quite sensitive to the  parameters controlling pheromone updating and schedule selecting. It is  comparable in solution quality to simulated annealing and genetic  algorithms.