Calibration of SUMO Microscopic Simulation for Heterogeneous Traffic Condition: The Case of the City of Khulna, Bangladesh

Abstract

In the context of Bangladesh, Efficient modeling of vehicular traffic is a challenging task. Simulation is one of the approaches for traffic modeling. As cities become more urbanized, intelligent transportation systems (ITS) are becoming progressively more important to the development of future smart cities. The purpose of this research is to compare simulation data with field data to calibrate the SUMO simulation model tool. In this study, a microscopic traffic simulation model for some CBD areas of Khulna City has been calibrated. However, the focus area of Powerhouse is strategically located at the intersection of four key places in Khulna City, encompassing the central business district (CBD) and the city's railway station. During the peak hour, congestion frequently occurs at this crossroads. This research aims to provide alternative scenarios to reduce queue and travel time in the Powerhouse intersect. Types of observation data are flow, length of queue, and travel time that are observed during field survey. The calibration process is done by minimizing the root mean square error (RMSE) of queue, and travel time, and combining both of them between observation and calibrated model. After the calibration process, the sigma and tau values came out to be 0.3 and 1.4, respectively. By using this calibrated value, the travel time output may accurately reflect real-world scenarios, allowing for experimentation with various scenarios. This study presented two alternate scenarios aimed at enhancing the performance of Powerhouse. Alternate: 1 is more of a geometric modification with the installment of Channelization, it will reduce travel time substantially for each lane, nearly 10 % to 30% of its current time. Alternate: 2 is more of applying strict and fast regulations for the traffic which is Traffic light management, it is a more convenient option that appears. The results generated by utilizing SUMO can be reliable and helpful in developing and optimizing urban transportation systems in the future. It is essential to understand that traffic simulation with SUMO is merely a transportation decision-making and planning tool and must be linked with field observations and reliable data for suitable and efficient transportation solutions.