Development of Bidirectional Mixed Traffic Simulation Model for Urban Roads

Abstract

Most of the cities in developing and emerging countries (e.g. India, China) consist of large proportion of undivided roads which carry mixed traffic with non-lane discipline. Vehicular manoeuvers on such roads are complex due to high lateral interactions between the vehicles moving in both directions in the absence of divider (median). Traffic congestion in cities during peak hours prolongs for longer periods each day which reduces the capacity of roads and increases delay and pollution. Possible ways to reduce these problems are to improve the operation of existing road systems through better traffic control and management measures. Microscopic simulation model is identified as a widely used tool to evaluate traffic control and management measures. The overall objective of this research work is to develop a microscopic simulation model for bi- directional traffic on urban undivided roads in mixed traffic conditions. The simulation model consists of three major logics: vehicle generation, vehicle placement, and vehicle movement. Longitudinal and lateral movements of vehicles are modeled together to mimic the real world conditions. The concept of influence area is introduced to identify the most influencing leader vehicle in vehicle movement logic. The simulation model is implemented in MATLAB programming language using Object-Oriented Programming (OOP) concepts. The model is calibrated and validated using internal and external datasets collected from two different urban cities (Bengaluru and Kollam) in India. The statistical validation indicates that the simulation model replicates the field conditions realistically. The developed model is applied to evaluate traffic management measures such as reversible lane (tidal flow) operations and modal shift of private vehicles towards public transport. A reversible roadway is one in which the direction of traffic flow in one or more lanes or shoulders is reversed to the opposing direction for some period of time to reduce congestion. The reversible lane operation is implemented in the model and the impact of it on capacity of roads is studied by using four different vehicular compositions commonly observed on major urban arterials in Indian cities. For this purpose, capacity of the road without and with reversible lanes are determined from simulation model. Simulation runs are performed for 24 scenarios without reversible lanes and 24 scenarios with it (total of 48 runs). Each simulation run represented one hour of traffic flow during peak period. Results show that there is animprovement in capacity during peak hours after implementing reversible lanes. For twowheeler dominant composition, the capacity in ongoing direction during morning peak hour is increased by 20.5% and similarly, capacity in opposing direction during evening peak hour is increased by 19.20%. The modal shift from private vehicles (two-wheelers and cars) to public transport system (bus) is implemented in the model and its impact on capacity, travel time and emissions are studied. Three different scenarios (shifting commuters, only from two-wheelers, only from cars, and both from two-wheelers and cars together) are evaluated using traffic data collected from study sections located in major urban cities such as Bengaluru, Delhi, and Mumbai. For this purpose, capacity, travel time and emissions before and after modal shift are obtained from simulation model. Simulation runs are performed for 9 scenarios before modal shift and 153 scenarios after modal shift (total of 162 runs). The optimum number of buses to be increased in these sections are also obtained. The results show that implementation of modal shift (for optimum increment in buses) improves the capacity by 16.9%, 17.8% and 45.8% for Bengaluru, Delhi, and Mumbai, respectively. Reduction in travel time observed for these sections are 17.2%, 26.9% and 19.5%, respectively. Also, there is a significant reduction in CO2 emissions in the range of 7.3 % - 12.6 %. The developed simulation model can serve as a tool for traffic engineers and policy makers to evaluate various other traffic control and management measures (e.g. exclusion of certain category of vehicles, implementation of lane discipline and lane segregation), which can be implemented on urban roads carrying mixed traffic as prevailing in developing and emerging countries.