A hybrid simulation-assignment modeling framework for crowd dynamics in large-scale pedestrian facilities

This paper presents a hybrid simulation-assignment modeling framework for studying crowd dynamics in large-scale pedestrian facilities. The proposed modeling framework judiciously manages the trade-off between ability to accurately capture congestion phenomena resulting from the pedestrians’ collective behavior and scalability to model large facilities. We present a novel modeling framework that integrates a dynamic simulation-assignment logic with a hybrid (two-layer or bi-resolution) representation of the facility. The top layer consists of a network representation of the facility, which enables modeling the pedestrians’ route planning decisions while performing their activities. The bottom layer consists of a high resolution Cellular Automata (CA) system for all open spaces, which enables modeling the pedestrians’ local maneuvers and movement decisions at a high level of detail. The model is applied to simulate the crowd dynamics in the ground floor of Al-Haram Al-Sharif Mosque in the City of Mecca, Saudi Arabia during the pilgrimage season. The analysis illustrates the model’s capability in accurately representing the observed congestion phenomena in the facility.     

Waiting zones for realistic modelling of pedestrian dynamics: A case study using two major German railway stations as examples

The value of a pedestrian stream simulation depends on its ability to reproduce natural behaviour of pedestrians in different situations. Most models assume that pedestrians are single-minded and constantly move towards their destinations. However, our observations at two major German railway stations made during field experiments and our analysis of video recordings at one of these stations revealed that in virtually every setting a significant proportion of pedestrians do not walk continuously. Instead, they occasionally change their route in order to visit certain locations and stand there for a period of time. By waiting, they often block walking pedestrians and thereby influence the overall dynamics.In this paper, we evaluate the impact of waiting pedestrians and propose a model for waiting pedestrians based on cellular automata. The model is able to reproduce the observed pedestrian behaviour. We illustrate the model with simulations of several real life scenarios for a major German railway station and show that during rush hour standing pedestrians may prolong walking time by up to nearly 20%. We also demonstrate how the developed model can be used for the analysis of infrastructures, and prediction of problematic areas in public spaces.     

Classification modeling approach for vehicle dynamics model‐integrated traffic simulation assessing surrogate safety

To assess safety impacts of untried traffic control strategies, an earlier study developed a vehicle dynamics model‐integrated (i.e., VISSIM‐CarSim‐SSAM) simulation approach and evaluated its performance using surrogate safety measures. Although the study found that the integrated simulation approach was a superior alternative to existing approaches in assessing surrogate safety, the computation time required for the implementation of the integrated simulation approach prevents it from using it in practice. Thus, this study developed and evaluated two types of models that could replace the integrated simulation approach with much faster computation time, feasible for real‐time implementation. The two models are as follows: (i) a statistical model (i.e., logit model) and (ii) a nonparametric approach (i.e., artificial neural network). The logit model and the neural network model were developed and trained on the basis of three simulation data sets obtained from the VISSIM‐CarSim‐SSAM integrated simulation approach, and their performances were compared in terms of the prediction accuracy. These two models were evaluated using six new simulation data sets. The results indicated that the neural network approach showing 97.7% prediction accuracy was superior to the logit model with 85.9% prediction accuracy. In addition, the correlation analysis results between the traffic conflicts obtained from the neural network approach and the actual traffic crash data collected in the field indicated a statistically significant relationship (i.e., 0.68 correlation coefficient) between them. This correlation strength is higher than that of the VISSIM only (i.e., the state of practice) simulation approach. The study results indicated that the neural network approach is not only a time‐efficient way to implementing the VISSIM‐CarSim‐SSAM integrated simulation but also a superior alternative in assessing surrogate safety. Copyright © 2014 John Wiley & Sons, Ltd.