Simulation of Classified Lane-Wise Vehicle Count at Toll Plazas Using Monte Carlo Simulation and Probability-Based Discrete Random Number Generation

The simulation-based prediction of traffic conditions based on current and past traffic observations is an important component in the intelligent transportation system (ITS) applications. Infrastructure, in the form of toll plazas, is inevitable for collection of revenue after the development of National Highways in India. Intelligent transportation systems utilize the advanced technologies and employ them in the field of transportation. The implementation of advanced traffic management systems (ATMS) at toll plazas will improve the toll plaza operations. A simulation model can help in the evaluation and optimization of toll operations of existing toll plazas as well as in the planning and design of similar systems. With this motivation, a lane-wise classified vehicle count prediction algorithm, which can simulate traffic conditions at any time interval, has been developed in this study based on Monte Carlo simulation (MCS). Vehicle arrival was modeled by assuming Poisson’s distribution, followed by classification. Lane selection was done using the probability-based discrete random number generation. Radio-frequency identification (RFID)-based electronic toll collection (ETC) system gives timely varying traffic counts observed at the toll plaza, which has been utilized to develop and validate the simulation model. The flexibility with respect to the probabilities of the proposed algorithm makes it more applicable in the area of ITS. The observed vehicle count for each lane has been compared with the simulated values. The results of statistical tests show that there is no significant difference between actual and simulated traffic for each lane.