A comparison study on landslide prediction through FAHP and Dempster–Shafer methods and their evaluation by P–A plots

Abstract

One of the important issues in identifying susceptible landslide locations is to select effective factors in the landslide of studied areas. The complexity of this issue is due to the fact that each of the factors is a part of the desired criteria, and the selection among them is a multi-criteria decision-making problem, requiring a structured and systematic approach. In the study area, the presence of large outcrops, as well as relatively thick soils formed on the formations, has caused a significant level of the province to potentially have the conditions for the movement of the tides. In addition, the situation is exacerbated by inappropriate use of natural resources, and construction of numerous rural and forest roads. In this paper, multi-criteria decision-making (MCDM) such as the fuzzy analytical hierarchy process (FAHP) and Dempster–Shafer (DeS) method was used to detect susceptible landslide locations, and output maps were verified applying the prediction–area (P–A) method for landslide purposes. The study has utilized P–A plot to evaluate and compare the obtained results of FAHP and DeS methods. The results indicated that the normalized density value achieved from P–A plots was significantly higher for the FAHP than the corresponding value for the DeS. In other words, FAHP, compared to DeS, resulted in the successful detection of landslide occurrences and more accurate classification of the study area.