Significance: Different traffic states experienced by commuters may be due to localized random jams or overall network states, which is decided by system response to perturbations. While for natural systems, the existence of multiple states is common, the evidence for multiple states in transportation has so far been unclear, which hinders our ability to predict and prevent the spread of traffic jams. Based on percolation theory and high-resolution GPS datasets, we find strong evidence for the existence of multiple metastable network states corresponding to distinct system performance regimes. Our results may provide deep insight for a city traffic manager seeking to design early warning signals to prevent the system from shifting to a state of severe congestion.

In this paper, we present a series of machine learning approaches for better understanding people’s travel mode choice. The widely used Logit model is dependent on the assumption that the utility items are independent, violating of this assumption caused inconsistent parameter estimations and biased predictions. To improve the prediction accuracy of mode choice, this paper employs the data fusion model based on stacking strategy and proposes a hybrid model of the unsupervised Denoising Autoencoder (DAE) combining with the supervised Random Forest (RF). A variety of features that may impact mode choice behavior are ranked and selected by using the feature selection algorithms. The proposed model, which is particularly useful and powerful in the choice behavior analysis and outperforms other widely used classifiers, is verified by travel diary data from Germany and Switzerland. The results can be used for better understanding and effectively modeling of human travel mode choice behavior.

Working from home becomes the norm; this trend has put added pressure on urban logistics, as large volumes of goods and services are required for domestic use. Meanwhile, public transport operators face a big challenge and trade-off due to higher labour and frequent cleaning costs, with lower passenger revenue over a longer period. Considering the collaborative urban public transport services achieve a seamless movement for both passengers and goods, and could reduce the adverse effects of the existing urban public transport systems. Therefore, a mixed-integer linear programming model introducing the concept of capacity matching is proposed to assist this collaborative urban freight service network in minimising total freight transport time at station hubs and not affecting passenger transportation in this paper. Moreover, an efficient improved optimisation algorithm based on the Artificial Bee Colony Algorithm (ABC) is designed, and the numerical examples and real cases are illustrated to demonstrate the feasibility and effectiveness of the proposed model and algorithm. The performance evaluations suggest that the coordinated operating strategy of the collaborative freight transportation system supports increasing mobility demands for freight, resulting in declining congestion levels and reducing transport emissions, while no influence in passenger transport, notably in urban areas.

This paper presents a cooperative traffic control strategy to increase the capacity of nonrecurrent bottlenecks such as work zones by making full use of the spatial resources upstream of work zones. The upstream area is divided into two zones: the regulation and the merging areas. The basic logic is that a large gap is more efficient in accommodating merging vehicles than several small and scattered gaps with the same total length. In the regulation area, a nonlinear programming model is developed to balance both traffic capacity improvements and safety risks. A two‐step solving algorithm is proposed for finding optimal solutions. In the merging area, the sorting algorithm is used to design lane‐changing trajectories based on the regulated platoons. A case study is conducted, and the results indicate that the proposed model is able to significantly improve work zone capacity with minor disturbances to the traffic.