Predicting speeds on urban streets using real-time GPS data

The study aimed at developing a mathematical model for predicting traffic speeds using bus speeds obtained from real time GPS data. Unlike the traditional approach of collecting speed data manually or with the help of sensors which is uneconomical, the present study looked at collecting speed data using GPS technology. A model was developed using multiple regression analysis, taking other roadway characteristics such as number of lanes per traveling direction and posted speed limits into consideration. Such a model can be used as a tool in real time traffic maps, displaying the details of speed and travel time information for a particular roadway section.