18 February 2009 in IEEE Transactions on Intelligent Transportation Systems
In this paper, an aggregation approach is proposed for traffic flow prediction that is based on the moving average (MA), exponential smoothing (ES), autoregressive MA (ARIMA), and neural network (NN) models. The aggregation approach assembles information from relevant time series. The source time series is the traffic flow volume that is collected 24 h/day over several years. The three relevant time series are a weekly similarity time series, a daily similarity time series, and an hourly time series, which can be directly generated from the source time series. The MA, ES, and ARIMA models are selected to give predictions of the three relevant time series. The predictions that result from the different models are used as the basis of the NN in the aggregation stage. The output of the trained NN serves as the final prediction. To assess the performance of the different models, the naive, ARIMA, nonparametric regression, NN, and data aggregation (DA) models are applied to the prediction of a real vehicle traffic flow, from which data have been collected at a data-collection point that is located on National Highway 107, Guangzhou, Guangdong, China. The outcome suggests that the DA model obtains a more accurate forecast than any individual model alone. The aggregation strategy can offer substantial benefits in terms of improving operational forecasting.
1 March 2010 in Journal of Transportation Engineering
In this paper, we develop a model of bidirectional pedestrian streams with an oblique intersecting angle that represents the crossing interaction of pedestrians moving in different directions on a walking facility. We formulate the model based on Drake’s model in the traffic flow literature, and calibrate it using results from a set of controlled experiments in which two groups of pedestrians were asked to walk in different directions and cross on designated walkways with different levels of pedestrian flow and different intersecting angles. Their movements were video recorded, and macroscopic quantities such as speed, density, and flow were extracted to calibrate the model of bidirectional pedestrian streams. All of the parameters of the calibrated model are statistically significant at the 0.1% level.
1 November 2010 in Physica A: Statistical Mechanics and its Applications
13 February 2012 in Physical Review E
This paper proposes a cellular automata model of pedestrian flow that defines a cost potential field, which takes into account the costs of travel time and discomfort, for a pedestrian to move to an empty neighboring cell. The formulation is based on a reconstruction of the density distribution and the underlying physics, including the rule for resolving conflicts, which is comparable to that in the floor field cellular automaton model. However, we assume that each pedestrian is familiar with the surroundings, thereby minimizing his or her instantaneous cost. This, in turn, helps reduce the randomness in selecting a target cell, which improves the existing cellular automata modelings, together with the computational efficiency. In the presence of two pedestrian groups, which are distinguished by their destinations, the cost distribution for each group is magnified due to the strong interaction between the two groups. As a typical phenomenon, the formation of lanes in the counter flow is reproduced.
1 March 2006 in Journal of Computational Physics
1 February 2010 in Physica A: Statistical Mechanics and its Applications