Traffic flow breakdown is the abrupt shift from operation at free-flow conditions to congested conditions and is typically the result of complex interactions in traffic dynamics. Because of its stochastic nature, breakdown is commonly predicted only in a probabilistic manner. This paper focuses on using stationary aggregated traffic data to capture traffic dynamics, developing and testing machine learning (ML) approaches for traffic breakdown prediction and comparing them with the traditionally used probabilistic approaches. The contribution of this study is three-fold: it explores the usefulness of temporally and spatially lagged detector data in predicting traffic flow breakdown occurrence, it develops and tests ML approaches for traffic breakdown prediction using this data, and it compares the predictive power and performance of these approaches with the traditionally used probabilistic methods. Feature selection results indicate that breakdown prediction benefits greatly from the inclusion of temporally and spatially lagged variables. Comparing the performance of the ML methods with the probabilistic approaches, ML methods achieve better prediction performance in relation to the class-balanced accuracy, true positive rate (recall), true negative rate (specificity), and positive predictive value (precision). Depending on the application of the prediction approach, the method selection criteria may differ on a case-by-case basis. Overall, the best performance was achieved by the neural network and support vector machine approaches with class balancing, and with the random forest approach without class balancing. Recommendations on the choice of prediction approaches based on the specific application objectives are also given.

This study investigates the prediction and mitigation of the phenomenon of traffic flow breakdown when affected by varying weather conditions. First, the probability of breakdown occurrence is examined using a survival analysis approach to obtain distributions of pre-breakdown flow rates under different weather conditions. Second, pre-breakdown flow rate distributions were applied in breakdown prediction for the implementation of breakdown mitigation strategies. In the first part, a set of data from the network of Kansas City was used to demonstrate the applicability of the Kaplan–Meier Product Limit method to estimating the breakdown probability under various weather conditions. Then, using simulated data on the network of Chicago, the K-M approach was used again to obtain survival likelihood distributions, which in turn yield breakdown probability, for 13 different weather cases as combinations of weather categories for different levels of visibility, rain, and snow precipitation. In the second part, continuing with the simulated data, dynamic speed limits (DSL) were applied to demonstrate the effectiveness of the prediction method presented. A sensitivity analysis of the threshold probability and upstream distance at which DSL should be implemented was performed for clear and inclement weather conditions. In clear weather the performance of the strategy is better at a lower probability threshold and farther upstream location, whereas in inclement weather the performance is better at a lower probability threshold and closer upstream location. The paper demonstrates the effect of changing weather conditions on the likelihood of breakdown occurrence and the implementation of breakdown mitigation strategies.

No abstract

Transportation research has increasingly focused on the modeling of travel time uncertainty in transportation networks. From a user’s perspective, the performance of the network is experienced at the level of a path, and, as such, knowledge of variability of travel times along paths contemplated by the user is necessary. This paper focuses on developing approaches for the estimation of path travel time distributions in stochastic time-varying networks so as to capture generalized correlations between link travel times. Specifically, the goal is to develop methods to estimate path travel time distributions for any path in the networks by synthesizing available trajectory data from various portions of the path, and this paper addresses that problem in a two-fold manner. Firstly, a Monte Carlo simulation (MCS)-based approach is presented for the convolution of time-varying random variables with general correlation structures and distribution shapes. Secondly, a combinatorial data-mining approach is developed, which aims to utilize sparse trajectory data for the estimation of path travel time distributions by implicitly capturing the complex correlation structure in the network travel times. Numerical results indicate that the MCS approach allowing for time-dependence and a time-varying correlation structure outperforms other approaches, and that its performance is robust with respect to different path travel time distributions. Additionally, using the path segmentations from the segment search approach with a MCS approach with time-dependence also produces accurate and robust estimates of the path travel time distributions with the added benefit of shorter computation times.

This study used logistic regression to determine whether there is a significant difference in pedestrian–vehicle interaction severity at side street green and exclusive phase pedestrian signals, and to evaluate whether waiting- and crossing times are useful predictors of pedestrian–vehicle conflict. To do this, data related to the physical characteristics of each intersection and to the crossing experience of every pedestrian were gathered at 32 signalized intersections in Connecticut. At each intersection, conflicts between pedestrians and vehicles were classified into four distinct severity levels: undisturbed passage, potential conflict, minor conflict, and serious conflict. After interpreting the results of six different logistic regression refinements, it was determined that waiting time, crossing time, the number of lanes, annual average daily traffic, pedestrian compliance, phasing, and the presence of crosswalks were all useful predictors. Based on the results of this analysis, low wait time and crossing time values were associated with a decrease in the odds of a conflict. It was also determined that exclusive phase signals reduced the odds of a conflict by 85%. Future research should investigate variables associated with different land development patterns and demographic information. Along with this, the crossing experience of individuals who do not comply with pedestrian signals should also be evaluated.