Journal of Rail Transport Planning & Management最新文献

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.

Train passenger demand fluctuates throughout the day. In order to let train services, such as the line plan and timetable, match this fluctuating demand, insights are needed into how the demand is changing and for which periods the demand is relatively stable. Hierarchical clustering on both regular and normalized origin–destination (OD) data is used to determine for each workday continuous time-of-day periods in which the passenger demand is homogeneous. The periods found for each workday are subsequently used as input in a clustering algorithm to look for similarities and differences between workdays. The methods for finding homogeneous periods during the day and week are applied to a case study covering a large part of the railway network in the Netherlands. We find large differences between the periods based on regular OD matrices and those based on normalized OD matrices. The periods based on regular OD matrices are more compact in terms of passenger volumes and average kms travelled and therefore more suitable to use as input for designing a service plan. Comparison of different workdays shows that mainly the peak periods on Friday are far away from Monday to Thursday, and hence could benefit from an altered service plan.

This paper systematically reviewed the slipping operation, which is a train separation at cruising speed. For this, we describe the historical and operational background of the operation scenario practiced for over 100 years. Based on the concept of slipping, we discuss the holistic potential to improve current railway operations, considering travel time saving, energy saving, the increase of capacity utilization, station topology, driver requirements, and vehicle usage. Finally, a simulation of a theoretical urban railway line with several scenarios quantifies the magnitudes of the improvements. Based on the slipping test cases, one parameter can improve enormously, e.g., up to −65 % energy saving, −33 % capacity usage, and travel time reductions. Otherwise, slipping can slightly improve several parameters simultaneously.

Train timetabling poses inherent challenges, prompting the need to enhance existing schedules by extracting valuable insights from the current timetable structure. By such philosophy, this paper studies the impact of timetable elements on the consumed capacity. These elements primarily encompass train operation parameters, including running times and stop plans. Their impact is defined as the deviation of consumed capacity relative to their variations. We initially establish a link between consumed capacity and timetable elements. This relationship is articulated as the signed sum of timetable elements along a designated ”critical path”. Then, the limited impact of any timetable element is clarified, namely changing an element can impact the consumed capacity with its neighbor trains in a combinatorial way. With this knowledge, we analyze the impact of a single element, using stop plans for example. This result is then generalized into analyzing the impact of several dependent and independent stop plans. The findings on capacity calculation and impact analysis of a single element are tested through real-world numerical computations and then extended to analyzing various capacity factors, such as average speed and heterogeneity.

Signal passed at danger (SPAD) is the most frequent cause of rail accidents as confirmed by relevant investigation in the field. The purpose of this paper was to identify the trends in SPAD research. Consequently, we aim to inspire and motivate future researchers, particularly academicians, to delve into this critical issue. Currently, the majority of researchers working on this topic are from the railway industry. For this reason, conducting a comprehensive review of existing research would yield significant value to both the academic and technical communities. This research examines the developments and accomplishments in this field concerning SPAD using the bibliometric library of R software from 1966 to the end of 2022. By bridging the gap in the existing literature, this research facilitates the global exchange of knowledge among railway experts, ultimately contributing to the reduction of safety risks and associated economic costs.

Over time, rail's share of the freight market has steadily decreased and, currently, is significantly lower than that of the road. This study explores what fosters shippers' rail dispreference. The study is conducted in the domain of outbound logistics in the steel-making industry in India. Twenty-one industry experts are interviewed in-depth to capture their perceptions, and their responses are analysed. Of these, seven are industry experts, and the remaining fourteen are logistics managers working across two steel plants, among which the annual output of one is about ten times that of the other. We find that a capacity shortage in the rail sector and the monopoly position of the rail transport provider together foster multiple factors that drive shippers' rail dispreference. Further, shipper firm size moderates the influence of some of these factors, influencing shippers' rail dispreference to a lesser extent in the larger firms than in the smaller ones. The study highlights the realization that while increasing rail capacity is necessary, it is not enough by itself, but must be complemented by targeted policy changes. The study brings to the forefront the roles played by rail capacity shortage, rail monopoly position, and shipper firm size in shippers' rail dispreference.

Railway operations may experience delays due to technical issues or weather conditions. Accurate prediction of such delays can enhance the quality of rail transport services and the effectiveness of railway operations. The study has developed the arrival delay prediction model using random forest regression based on the train operation data from the Ankara - Eskişehir high-speed train line in Turkey. The model can simultaneously predict arrival delays at all downstream stations on this line and continuously update these predictions as new information about train movements becomes available. The accuracy rates of the model vary from 76% to 99% under a 1-min prediction error. The results show that incorporating variables related to weather conditions and technical problems related to train control systems into the model improves prediction performance. The contribution of these variables to the model performance increases as the prediction horizon widens. The model results suggest that the model predictions may assist network managers in making better decisions about train operations. In order to evaluate the model's performance from the passengers' point of view, the study has proposed two methods: the proportion of late predictions and the stability of forecasts. The findings indicate that most trains (between 96.7% and 99%) have stable arrival delay predictions at target stations. The proportion of 2-min (or greater) late predictions, which means that the predicted delay exceeds the actual delay by 2 min or more, fluctuates from 14% to 0.5%, depending on the prediction horizon. Although the ratio for the short horizons (one station ahead) becomes relatively low, it is necessary to be cautious when using the model predictions to inform passengers because a prediction of more than 1 min late for short horizons might have negative consequences (e.g., misleading passengers to leave stations).

This paper evaluates the capacity effect of ERTMS/ETCS Hybrid Level 3 (HL3, also known as Hybrid Train Detection, HTD) on a conceptual level by looking at a scenario with two trains and on a network level. Key performance indicators help evaluate the results of implementing HL3: headway for the conceptual model and capacity utilization and punctuality for the network level. The study uses the simulation tool RailSys for both levels. A case study on the interaction between two trains examines how various lengths of virtual blocks affect the performance indicator headway. The network scale simulations use a real-world infrastructure and a complete timetable. Two cases examine how the performance indicators capacity utilization and punctuality are affected by the share of Level 2 and Level 3 trains in a HL3 system. Results from the conceptual two-train interaction show that HL3 slightly improves the headway, but it is similar for varying virtual block lengths. The results from the network model indicate the share of Level 2 and Level 3 trains has minimal effect on punctuality and capacity utilization. However, we identified some factors influencing the HL3 capacity evaluation, like stations and switches on the line, that affect the potential capacity gains.

Study on the platoon planning of virtual coupling (VC) is indispensable for both passenger and freight railway operation in terms of capacity, scheduling, and timetabling. As an advanced technology in train operation and control systems, VC can significantly improve the performance of rail operations, with regards to capacity, flexibility and robustness. In this study, we mainly focus on the former one, i.e., to maximize the capacity gains, by employing the relative braking distance based spacing policy for heterogenous train traffic. To our knowledge, this article is the first to address the VC intra-platoon planning problem for a railway network. The main function/purpose of this study is to decide a leader train and allocate an order to the heterogeneous follower trains for the virtually coupled platoon in the off-line mode. The objective is transformed to minimize and compare the sum of minimum safe distance for platoons on the common shared route sections in the railway network from various perspectives, so as to seek the desirable orders of potential train consists within the platoons from the perspective of the whole railway network. Our contributions are mainly threefold. First, the conditions for train platooning under virtual coupling are synthesized. Second, the solution flow charts and algorithms for simulation are proposed, including general framework for simulation, classification of prioritizing the shared common route section, and flow charts and algorithms for decision of coupling order to train platoons. Finally, the proposed methodology was tested and discussed on the designed computational experiment (i.e., numerical simulations of a typical case study with two sets of test examples) via NetLogo platform. The numerical simulation results series revealed some key findings and validated the effectiveness of the proposed methodology. Results of this study can provide guidance to decision makers in timetabling and potential capacity bottleneck identification under VC. We believe that it is worth investigating and further advancing this research direction in the future.

A key strategy for increasing railway capacity is utilizing infrastructure more efficiently. While much research has been completed on methods for assessing railway capacity, very little has focused on the details of capacity utilisation, such as assessing the various ways trains use capacity, the impacts of specific blocking time components, and how train dynamics (accelerating, cruising, braking, and dwelling) affect capacity.

This paper presents a methodology for comparing planned occupancy to actual occupancy under real operations and applies it in a case study. The methodology is based on identifying a critical path which represents an extension of bottleneck concept presented in UIC leaflet 406. The methodology was applied in a case study to determine the specific blocking times and train dynamics which cause a blocking time gap for a sequence of trains, both a-priori and a-posteriori, after considering the operational variations. The analysis of real operations with variations in train trajectories shows that capacity occupation is mostly influenced by train sequence heterogeneity in the original schedule. The varying effects of operations have a smaller but relevant impact. The methods developed in this paper can be used to help assess railway capacity under real operations.