A model for predicting AND detecting vehicular congestion to achieve uniform vehicular density

Abstract

Economic development and social uplifting has significantly increased vehicle population. Increased vehicles put stress on road infrastructure. Building new roads, bridges or reconditioning infrastructure is most of the times not the best possible solution. Congestion reduces speed of road travel, impacting turn-around-time and efficiency. Reduced speeds, frequent stoppages and longer idling increases fuel consumption and CO2 emission. Economy and global warming are the two main parameters driving researchers to evolve a long term, defensible solution to traffic congestion. newlineThe research dwells into examining studies of mobility systems, from amongst analytical, mathematical, statistical models as well as contemporary systems. Important being fluid mechanics where a relationship between density and flow finds favor with vehicular traffic. The research also evaluates simulation systems only to deduce that the problem is too complicated to resolve based on static or quasi static learning. Contemporary systems are studied to appreciate the limitations and identify prevailing research gap. The solution lies in developing affordable, scalable, real-time system which can be installed quickly without specialized domain knowledge, can be maintained effortlessly, has minimal or near nil operating cost and is non-disruptive. newlineThe proposed implementation is done on multiprocessor cloud computing environment and the vehicle software is embedded on a GPS device. The server houses road infrastructure database enabling the computation of road capacity and occupancy, it computes signal phases and communicates with signaling system, the alternate route algorithm advises an alternate route in case of higher congestion. newline newline