A bayesian dynamic linear model approach for real-time short-term freeway travel time prediction

This paper presents a Bayesian inference-based dynamic linear model (DLM) to predict online short-term travel time on a freeway stretch. The proposed method considers the predicted freeway travel time as the sum of the median of historical travel times, time-varying random variations in travel time, and a model evolution error, where the median is employed to recognize the primary travel time pattern while the variation captures unexpected supply (i.e. capacity) reduction and demand fluctuations. Bayesian forecasting is a learning process that revises sequentially the state of a priori knowledge of travel time based on newly available information. The prediction result is a posterior travel time distribution that can be employed to generate a single-value (typically but not necessarily the mean) travel time as well as a confidence interval representing the uncertainty of travel time prediction. To better track travel time fluctuations during non-recurrent congestion due to unforeseen events (e.g., incidents, accidents, or bad weather), the DLM is integrated into an adaptive control framework that can automatically learn and adjust the system evolution noise level. The experiment results based on the real loop detector data of an I-66 segment in Northern Virginia suggest that the proposed method is able to provide accurate and reliable travel time prediction under both recurrent and non-recurrent traffic conditions.     

Induced road traffic in Spanish regions: A dynamic panel data model

Distinguishing between traffic generated exclusively from the expansion of the road network (induced demand) and that resulting from other demand factors is of crucial importance to properly designed transport policies. This paper analyzes and quantifies the induced demand for road transport for Spain’s main regions from 1998 to 2006, years that saw mobility in Spain attain its highest growth rate. The lack of research in this area involving Spain and the key role played by the sector, given its high level of energy consumption and the negative externalities associated with it (accidents, noise, traffic congestion, emissions, etc.), endow greater relevance to this type of research. Based on a Dynamic Panel Data (DPD) reduced-form model, we apply alternative approaches (fixed and random effects and GMM-based methods) for measuring the induced demand. The results obtained provide evidence for the existence of an induced demand for transport in Spain, though said results vary depending on the estimating method employed.     

A novel variable selection method based on frequent pattern tree for real-time traffic accident risk prediction

With the availability of large volumes of real-time traffic flow data along with traffic accident information, there is a renewed interest in the development of models for the real-time prediction of traffic accident risk. One challenge, however, is that the available data are usually complex, noisy, and even misleading. This raises the question of how to select the most important explanatory variables to achieve an acceptable level of accuracy for real-time traffic accident risk prediction. To address this, the present paper proposes a novel Frequent Pattern tree (FP tree) based variable selection method. The method works by first identifying all the frequent patterns in the traffic accident dataset. Next, for each frequent pattern, we introduce a new metric, herein referred to as the Relative Object Purity Ratio (ROPR). The ROPR is then used to calculate the importance score of each explanatory variable which in turn can be used for ranking and selecting the variables that contribute most to explaining the accident patterns. To demonstrate the advantages of the proposed variable selection method, the study develops two traffic accident risk prediction models, based on accident data collected on interstate highway I-64 in Virginia, namely a k-nearest neighbor model and a Bayesian network. Prior to model development, two variable selection methods are utilized: (1) the FP tree based method proposed in this paper; and (2) the random forest method, a widely used variable selection method, which is used as the base case for comparison. The results show that the FP tree based accident risk prediction models perform better than the random forest based models, regardless of the type of prediction models (i.e. k-nearest neighbor or Bayesian network), the settings of their parameters, and the types of datasets used for model training and testing. The best model found is a FP tree based Bayesian network model that can predict 61.11% of accidents while having a false alarm rate of 38.16%. These results compare very favorably with other accident prediction models reported in the literature.     

A methodology to estimate travel time using dynamic traffic assignment (DTA) under incident conditions

This paper presents results from a research case study that examined the distribution of travel time of origin–destination (OD) pairs on a transportation network under incident conditions. Using a transportation simulation dynamic traffic assignment (DTA) model, incident on a transportation network is executed under normal conditions, incident conditions without traveler information availability, and incident conditions assuming that users had perfect knowledge of the incident conditions and could select paths to avoid the incident location. The results suggest that incidents have a different impact on different OD pairs. The results confirm that an effective traveler information system has the potential to ease the impacts of incident conditions network wide. Yet it is also important to note that the use of information may detriment some OD pairs while benefiting other OD pairs. The methodology demonstrated in this paper provides insights into the usefulness of embedding a fully calibrated DTA model into the analysis tools of a traffic management and information center.     

Stochastic cell transmission model (SCTM): A stochastic dynamic traffic model for traffic state surveillance and assignment

The paper proposes a first-order macroscopic stochastic dynamic traffic model, namely the stochastic cell transmission model (SCTM), to model traffic flow density on freeway segments with stochastic demand and supply. The SCTM consists of five operational modes corresponding to different congestion levels of the freeway segment. Each mode is formulated as a discrete time bilinear stochastic system. A set of probabilistic conditions is proposed to characterize the probability of occurrence of each mode. The overall effect of the five modes is estimated by the joint traffic density which is derived from the theory of finite mixture distribution. The SCTM captures not only the mean and standard deviation (SD) of density of the traffic flow, but also the propagation of SD over time and space. The SCTM is tested with a hypothetical freeway corridor simulation and an empirical study. The simulation results are compared against the means and SDs of traffic densities obtained from the Monte Carlo Simulation (MCS) of the modified cell transmission model (MCTM). An approximately two-miles freeway segment of Interstate 210 West (I-210W) in Los Ageles, Southern California, is chosen for the empirical study. Traffic data is obtained from the Performance Measurement System (PeMS). The stochastic parameters of the SCTM are calibrated against the flow-density empirical data of I-210W. Both the SCTM and the MCS of the MCTM are tested. A discussion of the computational efficiency and the accuracy issues of the two methods is provided based on the empirical results. Both the numerical simulation results and the empirical results confirm that the SCTM is capable of accurately estimating the means and SDs of the freeway densities as compared to the MCS.