Utilizing a fuzzy inference system (FIS) and modified analytical hierarchical analysis for forest road network planning in afforested lands

Abstract

Afforestation activity needs serious feasibility in as much as it is a work of establishing a new facility. Moreover, the process of selecting the land to be afforested; determining site conditions; selecting suitable tree species, provenance, and clones; the planting or sowing techniques; and planning of the environmental transport network are all costly. Determining suitable areas for environmental road network installation, which has an important place in afforestation areas, is discussed in this study. In this context, characteristics of an afforestated area have been evaluated by virtue of the fuzzy inference system (FIS) and modified analytical hierarchical analysis (M-AHP) methods on a sample area established as an afforestation area. The main objective of this study is planning in a way that is sensitive to nature and strives for ecological balance on environmental forest road networks for forested and afforested areas by utilizing multiple decision support methods with a view to realizing maintenance on afforestation areas in short- and long-term processes. A total of eight factors were determined in the study. The best twelve models, obtained as a result of analyses carried out with these factors, are presented. The success of the models was evaluated with receiver operating characteristic (ROC) analysis and the best model was obtained by M-AHP, which was Model 4 M-AHP with 71.2% area under curve (AUC) value. We observed that the model obtained through the M-AHP method is more successful in on-site road network planning during the phase of creating the afforestation projects.