Dynamic modelling and optimisation of transportation systems in the connected era

Sensing, information, and communication technologies have become fundamental components to transportation system management in the connected era. We have been witnessing an increasing number of instrumented networks and cooperative transportation systems deployed around the globe. The subsequent increasing availability of traffic measurements and data offer opportunities for research and development of advanced holistic methodologies and tools for dynamic transport systems modelling and optimisation. Recent advances in sensing, vehicle communication and automation technologies have generated new kinds of measurements, vehicles and infrastructure communications, and control actuation capabilities. There is a genuine and timely need for developing novel modelling and optimisation techniques in order to harness the opportunities offered by these technological achievements and consequently enhance the performance of transportation systems.

This special issue focuses on the development of quantitative modelling and optimisation approaches for transportation systems management with the use of advanced sensing, communication, and information technologies. Our goal is to consolidate and publish original and high-quality research papers addressing innovative approaches to improve traffic and transportation systems modelling and optimisation with applications of state-of-the-art technologies. Among all the received manuscripts, six papers have been selected for publication on this special issue after a rigorous peer review process, as follows:

Yang and Li investigate the dynamics of a bi-agent logistics system on an arbitrary complex network. Large-scale agent-based numerical simulations are carried out to explore realistic and complicated systems, with interesting emergent behaviours that can be well understood from analytical studies. Using the taxi system with occupied and empty taxis, they illustrate two dynamical phases with distinct behaviours, separated by a phase boundary that can be identified as the optimal number of taxis for a particular taxi system. These features and the tuning of the optimal number of taxis can be applied to various situations including real-time ride-sharing.

Xie et al. propose two cooperative driving strategies for connected vehicles (CVs) that move in mixed traffic with regular vehicles (RVs) and CVs, namely CDS-L and CDS-G. CDS-L is a local strategy that considers only the traffic surrounding the subject CVs. CDS-G is a global strategy that considers the conditions of all CVs on a road. The basis of the strategies is to incorporate the spatial and temporal averages of the driving status of surrounding CVs. A parameter that is a function of the traffic density is introduced to balance traffic efficiency and fluctuations. Simulation results demonstrate that the proposed strategies well suppress traffic fluctuations and thus promote traffic efficiency. The proposed strategies are dedicated to traffic congestion mitigation by using low-level configuration CVs.

Jiang and Gao propose a layout for Displaced Left-turn (DLT) intersection by combining the conventional DLT intersection with protected left-turn phases, based on the safe-crossing requirements of pedestrians and non-motorized vehicles. The delay and stops are weighted to form an integrated performance index (PI) in a real-time vehicle-to-infrastructure communication environment. The PI models, pertaining to all vehicles in non-DLT lanes, DLT lanes and shared through-right lanes are established based on the improved DLT intersection in unsaturated traffic conditions. In addition, a dynamic optimised method of traffic signal timing parameters is constructed based on minimising average PI per vehicle and considering the traffic signals’ coordination at the main intersection and left-turn crossovers. The operational performance of the optimised method is validated using data collected at an intersection in Harbin, China.

Sahebgharani et al. extend the concept of reliable space-time prisms (STPs) to non-normally distributed and spatially correlated networks. In doing so, a method is elaborated for computing such a prism, and several cases are carried out to show applicability of the proposed model. Outputs depicted different prism size in the correlated and independent networks and ability of the model for representing spatial extent of an individual considering various on-time arrival probability preferences.

Gao et al. propose a data-driven lane change detection (DLCD) system using deep learning techniques. Firstly, DLCD system explores to modelling driving context in spatial domain instead of traditional temporal domain. Secondly, DLCD has an ability of extracting innovative features, i.e. vehicle dynamics feature, lane boundary based distance feature and visual scene-centric feature from multi-modal input data efficiently. Finally, an improved focal loss-based deep long short-term memory (FL-LSTM) network is introduced to learn co-occurrence features and capture the dependencies within lane change events simultaneously. The experimental results on a real-world driving data set show that the DLCD system can learn the latent features of lane change behaviours and significantly outperform other advanced models.

Zhang et al. investigate how variation in value of time for AVs will reshape the commuting dynamics in the short-run and the implication on AV-related policies in the long run. They find that in the short run, the adoption of AV can create more congestion delays since delay becomes cheaper for commuters. In the long run, a number of external factors such as ownership cost and safety concerns may affect commuters’ preference for AVs as against to traditional vehicles (TVs). This will influence the AV penetration, which in turn affects the daily commuting equilibrium. Multiple long-run equilibria with different AV penetrations may exist, depending on the additional cost/benefit of AVs with respect to TVs. Government subsidies may be needed to drive the system from inefficient long-run equilibrium to a more efficient one.