Disaster risk evaluation using factor analysis: a case study of Chinese regions

Abstract

Regional risk to natural disasters is a critical multi-criteria decision-making (MCDM) problem in the literature due to the complicated and usually conflicting evaluation index system. Although a variety of MCDM methods can be applied to deal with the problem, the prior study primarily focused on the ranking of alternatives with little investigation on the influence of indicators. In this paper, an integrated approach is proposed by combining factor analysis and MCDM techniques to evaluate the thirty-one Chinese regions in terms of twenty-eight indicators. The advantage of factor analysis is demonstrated in extracting the dominant factors in an interpretable manner. Two commonly used MCDM techniques, namely TOPSIS and VIKOR, are then employed to evaluate the comprehensive risk of regions to natural hazards. The proposed approach not only provides the ranking of regions but also reveals the influence of indicators on the regional risk.