Clustering Crash Hotspots to Organize Police Dispatch Routes Using GIS

Traffic crashes events are known threats to public safety. Identifying high-crash-risk road segments provides safety specialists with an insight to better understanding of crash patterns and enhancing road safety management. In this work the main aim is to identify hotspots of severe crashes and reduce police dispatch time thereby reducing crash severity. In this crash monitoring and identifying approach, the objects and its positions tend to change periodically due to its dynamicity and mobility by which there should be a provision to monitor its behavior and position regularly. This will help to study the traffic flow and object behavior in the study locality. Based on the object movement in the region the position and time will be updated based on which the speed can be calculated. The frequency of speed will helps to identify the traffic situation of the region. Instead of monitoring individual node movement and speed the system will analyze the region based on multi-object speed analysis. It also defines the minimum and maximum speed and average speed flow of the object in a particular region. After a certain period of time, the object selects a new random path based on the distance and maximum speed path. In this process a novel Position-based speed updating process is proposed, in which several moving objects are recognized by the system. A dynamic segmentation is done to classify regions based on crash severity and minor clustering of hotspots is done. Finally kernel density is analyzed and frequency is calculated by which the data will be analyzed for further classification. ArcGIS was used for analyzing the data and displaying the results.