A Mathematical Model of Tree Selection in a Homogeneous Plantation during Improvement Thinning

Currently, the qualitative composition of merchantable forests of maturity age and approaching it is often unsatisfactory. This is largely due to the lack of improvement thinning or its improper carrying out. The productivity of logging operations during improvement thinning is low, and the resulting timber in most cases turns out to be unclaimed. Mechanization of improvement thinning is inconvenient from a production point of view and economically unprofitable. At the same time, in the Russian Federation there are ex amples of the implementation of advanced foreign practices of improvement thinning by machine using a harvester and a forwarder. The use of this method allows for a multiple increase in productivity. The unit costs (rub/m3) in this case are higher than the costs of clear cutting of mature and overmature plantations, but often lower than when felling forests. It is possible to further increase the efficiency of machine logging complexes during improvement thinning operations by automating the selection of trees for logging. The basis for such selection should be the main objective of this type of logging – obtaining the most optimal size and quality characteristics of plantations by the age of their maturity. This becomes possible when taking into account the competition of woody plants in the plantation before and after improvement thinning. The article introduces a mathematical formulation of the problem of assigning trees for logging in a homogeneous plantation. It is shown that the number of possible solutions is growing exponentially. An algorithm based on the greedy method is implemented, and an experimental comparison of this method with a random result, as well as with the removal of a tree with the closest neighbour among all trees, is performed. Testing the algorithm on experimental data has shown that it is more effective than the other two.