For the development of a sustainable transportation system, a modal shift from automobiles to walking or transit is encouraged. In order to design a more convenient and comfortable walking environment, a sound modeling of pedestrian flow is necessary. Most of the previously developed pedestrian flow models well described the macroscopic features of unidirectional pedestrian flow. However, in pedestrian counter-flow, interactions among conflicting pedestrians are so complicated and existing flow models fall short in explaining some features of pedestrian behaviors. A spontaneous lane formation, which helps to reduce conflicts and increase travel speeds, is a commonly observed feature of a crowded pedestrian flow. This paper develops a social-force based pedestrian model, which can explain the lane formation phenomenon. From the simulation results, it turns out that the ‘following effect’ and ‘evasive effect’ mainly contribute to the lane formation. Higher capacity and travel speed are obtained when pedestrians are more congregated.
26 February 2021 in KSCE Journal of Civil Engineering
In a traffic flow, once congestion is initiated, the throughput to the downstream of the bottleneck tends to decrease from the maximum throughput in the free-flow state. It is known that the maximum throughput in the free-flow state varies according to the traffic conditions for the same study site. If we can correctly forecast the maximum throughput depending on the traffic conditions, traffic congestion can be prevented or mitigated through ramp-metering. In this study, the relationship between the maximum throughput and the on-ramp ratio is analyzed. It turns out that maximum throughput of each study site has different values with the change of on-ramp ratio and there exists a convex quadratic-polynomial relationship between them. The elasticity of the maximum throughput to the on-ramp ratio is influenced by the geometric highway design and operation factors. The results of this study confirm that the on-ramp ratio control can effectively minimize the capacity drop and prevent traffic congestion.
15 April 2023 in Transportation Research Record: Journal of the Transportation Research Board
At freeway weaving sections, traffic congestion and accidents are frequent because of the disturbance generated by merging or diverging vehicles. To reduce the disturbance and accordingly improve the level of service (LOS), an appropriate length of weaving section should be established. According to the US-HCM, the weaving section density, which is a surrogate for LOS, is determined by weaving length, traffic volume, and weaving volume ratio. However, it is not yet clear whether the US-HCM accurately estimates the density for various traffic conditions. The purpose of this study is to evaluate the effects of weaving length under various traffic conditions through microscopic traffic simulation VISSIM. The GA optimization method is applied to build the car-following and lane-changing models in VISSIM. The results show that weaving length is a critical determinant of LOS and short weaving length significantly increases traffic congestion. Increasing the length of the weaving section is a good countermeasure for alleviating traffic congestion. This study also proposes the appropriate weaving section length, with which the upstream LOS can be maintained in the weaving section. These results would serve as valuable information in the planning or design phase of a weaving section given that the flow demands are specified.