Microscopic Simulation Model for Pedestrian Flow at Signalized Crosswalks

Signalized crosswalks are considered an important pedestrian facility, where the pedestrian flow possesses unique characteristics. This study used an improved cellular automaton model, which incorporated social forces to describe the interactions between pedestrians, to model the bidirectional pedestrian flow at a crosswalk. The simulation model was capable of estimating crossing time for various levels of pedestrian demand. The observation field data extracted from the video record validated the model's ability to estimate average crossing speed. It also found that pedestrian crossing time and speed are correlated to pedestrian demand. In addition, the proposed model presented the characteristics of pedestrian flow at the crosswalk visually and showed that the interaction between the conflicting flows contributed to an increase in crossing time and a reduction in crossing speed. The benefits of this model can be used to assess the design of signalized crosswalks, including their geometry and pedestrian cycle length.