The selection of trees for thinning with the fuzzy reasoning model
1996
Yamasaki, H. (Kyoto Univ. (Japan). Coll. of Agriculture) | Yoshimura, T. | Kanzaki, K.
To help forest managers start a long-term, high-quality timber production system, we constructed a fuzzy reasoning model, which selects trees for thinning by several simple rules. The model is supposed to make a judgment that only an experienced expert can make. Inputting the data on quality of the stem, liveliness of the tree, and location of the tree with regard to other trees, the model makes a decision about which trees should be cut. We applied the model to all of the trees on a plot of 0.05 ha and obtained the following percentage of discriminant accuracy: 86% over all for the trees and 63% for the trees to be cut. The discriminant accuracy was a little better than that derived from a discriminant analysis with the same explanatory variables. Furthermore, it was proved that the fuzzy reasoning model easily could be modified for adjustment. The result of the present study implies that the fuzzy reasoning system can reproduce the judgment of an experienced expert quite well, which contributes to forest management planning. The result also showed that such variables as the locations of trees and the relative quality of the adjoining trees and important for the judgment
Показать больше [+] Меньше [-]