Prediction based traffic management in a metropolitan area

In recent years, modern metropolitan areas are the main indicators of economic growth of nation. In metropolitan areas, number and frequency of vehicles have increased tremendously, and they create issues, like traffic congestion, accidents, environmental pollution, economical losses and unnecessary waste of fuel. In this paper, we propose traffic management system based on the prediction information to reduce the above mentioned issues in a metropolitan area. The proposed traffic management system makes use of static and mobile agents, where the static agent available at region creates and dispatches mobile agents to zones in a metropolitan area. The migrated mobile agents use emergent intelligence technique to collect and share traffic flow parameters (speed and density), historical data, resource information, spatio-temporal data and so on, and are analyzes the static agent. The emergent intelligence technique at static agent uses analyzed, historical and spatio-temporal data for monitoring and predicting the expected patterns of traffic density (commuters and vehicles) and travel times in each zone and region. The static agent optimizes predicted and analyzed data for choosing optimal routes to divert the traffic, in order to ensure smooth traffic flow and reduce frequency of occurrence of traffic congestion, reduce traffic density and travel time. The performance analysis is performed in realistic scenario by integrating NS2, SUMO, OpenStreatMap (OSM) and MOVE tool. The effectiveness of the proposed approach has been compared with the existing approach.