A GIS-based multi-criteria evaluation framework for uncertainty reduction in earthquake disaster management using granular computing

Abstract

One of the most important steps in earthquake disaster management is the prediction of probable damages which is called earthquake vulnerability assessment. Earthquake vulnerability assessment is a multi-criteria problem and a number of multi-criteria decision making models have been proposed for the problem. Two main sources of uncertainty including uncertainty associated with experts’ point of views and the one associated with attribute values exist in the earthquake vulnerability assessment problem. If the uncertainty in these two sources is not handled properly the resulted seismic vulnerability map will be unreliable. The main objective of this research is to propose a reliable model for earthquake vulnerability assessment which is able to manage the uncertainty associated with the experts’ opinions. Granular Computing (GrC) is able to extract a set of if-then rules with minimum incompatibility from an information table. An integration of DempsterShafer Theory (DST) and GrC is applied in the current research to minimize the entropy in experts’ opinions. The accuracy of the model based on the integration of the DST and GrC is 83%, while the accuracy of the single-expert model is 62% which indicates the importance of uncertainty management in seismic vulnerability assessment problem. Due to limited accessibility to current data, only six criteria are used in this model. However, the model is able to take into account both qualitative and quantitative criteria.