Journal of Rail Transport Planning & Management最新文献

This paper investigates the acceptance of a mega rail freight infrastructure as a sustainable alternative to road transport for domestic freight movements in India. The dedicated rail freight corridors (DFCs) are the freight-only rail corridors proposed by the Indian government to improve freight mobility from a sustainable outlook. A shipper/freight forwarder survey was conducted to gather information on mode attributes and their stated preferences toward DFCs. We employ discrete choice (binary logit) and machine learning algorithms (random forest and extreme gradient boosting) to analyse the choice behaviour. The machine learning methods exhibited higher prediction accuracy, while discrete choice models offered better interpretability. On-time performance and transport costs are crucial factors that influence mode choice. Large-scale companies are more willing to shift to DFCs compared to small and medium firms. The policy scenario analysis indicates that providing a better on-time performance can gain a substantial share of DFCs.

The study explores the relationship between the determinants and the ridership decrease incorporating spatial heterogeneity. ARIMA model is utilized to estimate the normal ridership assumed absence of COVID-19. Geography weighted regression (GWR) with Gaussian kernel function is constructed for regression. The K-means algorithm is applied to cluster the stations based on coefficients. Stations of Tokyo case are clustered into 2 groups: city area and western ward which represents mainly suburban areas. City stations are mainly influenced by the number of transfer lines, distance to the CBD, number of jobs and residents. In the western ward, the level of importance that residents place on public health primarily influences the ridership decrease. The implementation of work-from-home policies makes number of jobs a positive impactor on the decrease in ridership, with a greater impact observed on urban stations compared to suburban stations. City residents tend to engage in more travel than suburban residents because of less spacious living environments, which partially offsets the decrease in ridership. The findings offer parameters for predicting ridership of both city and suburban stations during public health emergency events, such as COVID-19. They can assist URT operators in developing strategies for balancing passenger demand and operational costs.

The contribution of this paper consists of a deep reinforcement learning (DRL) based method for autonomous train collision avoidance. While DRL applied to autonomous vehicles’ collision avoidance has shown interesting results compared to traditional methods, train-like vehicles are not currently covered. In addition, DRL applied to collision avoidance suffers from sparse rewards, which can lead to poor convergence and long training time. To overcome these limitations, this paper proposes a method for training a reinforcement learning (RL) agent for collision avoidance using local obstacle information mapped into occupancy grids. This method also integrates a network architecture containing a predictive auxiliary task consisting in future state prediction and encouraging the intermediate representation to be predictive of obstacle trajectories. A comparison study conducted on multiple simulated scenarios demonstrates that the trained policy outperforms other deep-learning-based policies as well as human driving in terms of both safety and efficiency. As a first step toward the certification of a DRL based method, this paper proposes to approximate the policy learned by the RL agent with an interpretable decision tree. Although this approximation results in a loss of performance, it enables a safety analysis of the learned function and thus paves the way to use the strengths of RL in certifiable algorithms. As this work is pioneering the use of RL for collision avoidance of rail-guided vehicles, and to facilitate future work by other engineers and researchers, a RL-ready simulator is provided with this paper.

Heavy-haul railways (HHRs) pose significant challenges due to their substantial traction weight, extended train length, and complex operational environments. Heavy-haul trains (HHTs), equipped with traditional pneumatic control braking systems, must adopt cycle braking strategies on long downhill slopes. The varying traction masses of HHTs on these railways lead to diverse maneuvering characteristics, presenting challenges for drivers and dispatchers in unforeseen circumstances. To enhance transportation efficiency and mitigate operational complexities, a trajectory optimization method is formulated for determining the optimal trajectory of HHTs with different traction masses under complex conditions, including long downhill slopes, temporary speed limit sections, and regular sections. It considers the dynamics of train traction, braking, and coasting at each phase, optimizing objectives such as train operation efficiency, energy consumption, and pneumatic braking times. A linear weight search algorithm ensures punctuality, and the model is linearized into a mixed-integer linear programming (MILP) form using segmented and stepwise functions to align with operational realities. Simulation experiments utilizing real data and various HHT configurations validate the efficacy of the proposed approach against alternative methods. This method offers precise trajectory optimization under complex conditions, providing valuable guidance for dispatchers and drivers in the heavy-haul railway sector.