Abstract: In order to effectively evaluate the fire safety status of metro stations, the fire safety evaluation index system of subway stations is constructed from four dimensions. The weights of indexes are determined by the improved CM- AHP and factor analysis, and the combination optimization is based on the least square method. The cloud model is generated according to the modified digital characteristics, and then the floating comprehensive level model is integrated. The comprehensive evaluation cloud model is obtained by combining the grade membership degree and combination weight of the indexes. Through the cloud map, the degree of dependency on the comprehensive level cloud and the development trend of the fire safety state can be seen intuitively.

Key words:  , metro station, fire safety, CM- AHP, factor analysis, cloud model