Abstract: Researches in landslide hazard zonation are of important practical significance in urban planning decision-making. With the statistical form, cluster analysis can be used to classify the data with similar characteristics, and the quantitative assessment of the spatial distribution of landslide hazard may be realized. Based on landslide hazard data in Badong County, representative landslide hazard risk factors have been chosen as index in the zoning evaluation. Entropy method and AHP method have been combined to give a comprehensive evaluation of the index weight. On the basis of the obove results, MapGIS is used as a platform and C# programming language is used to achieve a faster clustering algorithm. 86216 units in this study area are classified according to landslide hazard properties and are identified automatically to different risk levels. The results of this research are reliable and the method based on weight analysis and cluster model has innovative and practical value in processing highvolume and multiattribute data in field studies.