In　order　to　solve　the　problem　of　lacking　of　accurate　foundaion　data　which　results　in　fuzziness　and　randomness　in　evaluating　cities＇　ability　of　reducing　earthquake　disasters,　a　comprehensive　evaluation　method　based　on　cloud　models　is　expatiated　in　this　paper.　A　new　mathematical　representation　of　qualitative　concepts　is　represented　by　cloud　models.　With　the　new　models,　mapping　between　qualities　and　quantities　becomes　much　easier　and　interchangeable.　Fuzziness　and　randomness　are　integrated　using　digital　characteristics,　such　as　expected　value　Ex,　entropy　En　and　hyper　entropy　He.　The　fuzziness　and　randomness　are　not　only　complementary,　but　also　inseparable.　It　starts　with　the　uncertainty　of　the　system,　then　the　theory　of　Analytic　Hierarchy　Process　for　reference　is　used.　And　it　makes　the　result　of　evaluation　more　accurate　when　used　to　analyze　cities＇　ability　of　reducing　earthquake　disasters.