On the diversity-basedweighting method for risk assessment and decision-making about natural hazards

34Citations
Citations of this article
22Readers
Mendeley users who have this article in their library.

Abstract

The entropy-weighting method (EWM) and variation coefficient method (VCM) are two typical diversity-based weighting methods, which are widely used in risk assessment and decision-making for natural hazards. However, for the attributes with a specific range of values (RV), the weights calculated by EWM and VCM (abbreviated as WE and WV) may be irrational. To solve this problem, a new indicator representing the dipartite degree is proposed, which is called the coefficient of dipartite degree (CDD), and the corresponding weighting method is called the dipartite coefficient method (DCM). Firstly, based on a large amount of statistical data, a comparison between the EWM and VCM is carried out. It is found that there is a strong correlation between the weights calculated by the EWM and VCM (abbreviated as WE and WV); however, in some cases the difference between WE and WV is big. Especially when the diversity of attributes is high, WE may be much larger than WV. Then, a comparison of the DCM, EWM and VCM is carried out based on two case studies. The results indicate that DCM is preferred for determining the weights of the attributes with a specific RV, and if the values of attributes are large enough, the EWM and VCM are both available. The EWM is more suitable for distinguishing the alternatives, but prudence is required when the diversity of an attribute is high. Finally, the applications of the diversity-based weighting method in natural hazards are discussed.

Cite

CITATION STYLE

APA

Chen, P. (2019). On the diversity-basedweighting method for risk assessment and decision-making about natural hazards. Entropy, 21(3). https://doi.org/10.3390/e21030269

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free