Information-based predictive congestion control for intelligent transportation management

Third world cities are overpopulated and the number of vehicles plying on roads far exceeds the available traffic handling facilities resulting in frequent traffic congestions. Billions of dollars are being spent by the governments to construct over bridges for traffic management. This work proposes a unique solution scheme in which current traffic information is gathered through inductive loop detectors, piezoelectric sensors and wireless sensor networks coupled with mobile telephony to predict traffic flow and control traffic dynamics. The device placements in turn are dependent on information concerning traffic history of the place. The scheme is modeled on the basis of load-based packet switching as in computer network theory and is cost-effective, business-oriented and hence sustainable and replicable. This work proves the effectiveness of the solution both mathematically and through simulated traffic dynamics for Jadavpur and Gariahat , two heavily congested test cases in Calcutta, India. This is the first information-based decision making scheme which predicts the future for current congestion control using an integrated architecture of in-vogue techniques, sensor and cellular networks.