Performance Evaluation of Urban Traffic Using Simulation: A Case Study in Brazil

As a result of a fast-growing population, an increasing number of vehicles on the road, and inadequate public policies, metropolitan areas in Latin America are dealing with significant traffic congestion problems. Most cities do not have real-time urban traffic control systems. Therefore, the use of simulation software is a cost-effective solution to evaluate and reduce congestion in metropolitan areas. This paper seeks to assess urban traffic performance using the Urban Mobility Simulator (SUMO) on Fernandes Lima Avenue, the most important thoroughfare in Maceio, Alagoas, Brazil, which features distinctive characteristics such as a dedicated lane for public transportation and three segments with pedestrian traffic lights. Comparing the real observations with the simulation results, it was confirmed that the model provided accurate estimates, with errors of less than 5% for vehicle traffic volume and 10% for total travel time. After conducting experimental studies on four different scenarios, including the current state (1), no blue lane (2), no pedestrian traffic lights (3), and no blue lane and no pedestrian traffic lights (4), it was found that significant improvements in efficiency indicators, such as travel time, waiting time, fuel consumption, and carbon dioxide emissions, could be achieved. Scenarios 2, 3, and 4 were particularly effective, resulting in volumetric increases of 9.95%, 7.88%, and 10.77%, respectively, in vehicle traffic.