Spatio-Temporal Clustering of Road Accidents: GIS Based Analysis and Assessment

Road accident hot spots are evaluated and delineated in a South Indian city where inadequate development of land transport network often leads to traffic congestion and accidents. The patterns of localization and distribution of hotspots are examined with the help of geo-information technology to bring out the influence of spatial and/or temporal factors in their formation. Assessment of spatial clustering of accidents and hotspots spatial densities was carried out following Moran's I method of spatial autocorrelation, Getis-Ord Gi* statistics and point Kernel density functions. The accidents as a whole show a clustered nature while the comparison of spatio-temporal break ups indicates random distribution in certain classes. The Kernel density surface, estimated from the results of hotspot analysis delineates the road stretches as well as isolated zones where the hotspots are concentrated. The results can be effectively used by various agencies for adopting better planning and management strategies for improved traffic conditions as well as accident reduction.