Journal of Rail Transport Planning & Management最新文献

Freight train is selected to pick up and hang refrigerated containers according to railway freight train diagram in China. A multiobjective optimization model of railway cold-chain transportation route is established with total, time, and reliability costs as objective functions on the basis of dynamic train information. Different weights of three objective functions are obtained and weighted sum method is used to transform the multiobjective problem into a single-objective problem according to different transport demands of railway cold-chain transportation participants. An example of k short-circuit optimization algorithm based on genetic algorithm (GA) is designed to prove the feasibility and effectiveness of the proposed model. The empirical analysis showed that different transport times can be obtained by adjusting weights of various optimization objectives in the model to meet diverse needs of railway cold-chain transport participants and selecting differentiated shifts and transfer stations on the same route to provide a variety of transportation time limit options. Results of this study can provide guidance to decision makers in choosing railway transportation schemes.

Crowd at metro stations is usually a mixture of individuals and small groups of families or friends. However, limited research has focused on small group behaviours for metro safe evacuation evaluation and planning. In this study, a field observation at metro stations and a questionnaire survey were conducted to reveal the small group behaviour characteristics with different decision patterns and compactness. A cellular automaton (CA) based simulation model was proposed to reproduce small group behaviours of independent or joint decision pattern, with loose or close contact, reflecting the real-time trade-off between individual efficiency and group coherence. Impacts of small group behaviours on crowd dynamics were investigated by simulation experiments under diverse scenarios. Simulation experiments revealed that joint decision pattern and close contact of small groups were more likely to lead to longer evacuation time, lower average speed and stronger interference on the individuals. Deviations of estimated evacuation time due to small group behaviours were investigated and found to be common and widespread with different group decision pattern and compactness, congestion levels, proportions of groups in the crowd and exit layouts.

Fault text classification is a prerequisite task for railway engineers based historical train operation data to diagnose vehicle on-board equipment (VOBE) faults and formulate maintenance strategies. Aiming at the low efficiency and accuracy of manual fault text classification, based on Bidirectional Gated Recurrent Unit (BiGRU) and improved attention mechanism (IAtt), an intelligent VOBE fault text classification method is proposed in this paper. Combining the characteristics of the VOBE faults text, also called application event log (AElog) files, the Labeled-Doc2vec is used to generate sentence embedding to realize the vectorized representation of the fault texts, then input sentence embedding into BiGRU to extract the fault text features as the improved attention mechanism layer. Finally, the high-dimensional fault text features outputted by hidden are input into Softmax to complete the fault text classification. The experimental results show that the proposed method can analyze the semantics of fault text according to the train running state before and after the fault time, that is, it can realize text classification by combining context. Compared with other methods, the method in this paper obtains the optimal accuracy, precision, recall and F1-score, which shows that the proposed method can be applied to fault text classification of VOBE, effectively reduces the labor cost of fault text classification in practice, and improves the efficiency of fault text classification of VOBE.

The classic optimal train control problem is to drive a train on a track with known gradient over a fixed distance and within a specified time in such a way as to minimize tractive energy consumption. On level track the optimal strategies take two basic forms—a truncated strategy of optimal type with phases of maximum acceleration, coast and maximum brake which is typical of shorter metropolitan journeys, and an extended strategy of optimal type with phases of maximum acceleration, speedhold at the optimal driving speed, coast to the optimal braking speed, and maximum brake which is typical of longer journeys by freight trains and intercity passenger trains. The cost of these optimal strategies is uniquely determined by the journey distance and journey time. In this paper we extend a previously known formula for the partial rate of change of cost with respect to journey time to a formula for the full rate of change of cost that also incorporates the partial rate of change of cost with respect to journey distance.

Every day, railways experience disturbances and disruptions, both on the network and the fleet side, that affect the stability of rail traffic. Induced delays propagate through the network, which leads to a mismatch in demand and offer for goods and passengers, and, in turn, to a loss in service quality. In these cases, it is the duty of human traffic controllers, the so-called dispatchers, to do their best to minimize the impact on traffic. However, dispatchers inevitably have a limited depth of perception of the knock-on effect of their decisions, particularly how they affect areas of the network that are outside their direct control. In recent years, much work in Decision Science has been devoted to developing methods to solve the problem automatically and support the dispatchers in this challenging task. This paper investigates Machine Learning-based methods for tackling this problem, proposing two different Deep Q-Learning methods(Decentralized and Centralized). Numerical results show the superiority of these techniques respect to the classical linear Q-Learning based on matrices. Moreover the Centralized approach is compared with a MILP formulation showing interesting results. The experiments are inspired on data provided by a U.S. class 1 railroad.

Energy-saving driving is crucial during railway operation, especially in case of disturbances that require timetable rescheduling. This paper presents a method for the dynamic onboard tuning of energy-efficient speed profiles after the real-time train rescheduling process under a fixed block signaling system for mixed traffic. Similar to the idea of connected driver advisory systems, trains constantly communicate with a central traffic management system. After the rescheduling process, this system provides recommended time corridors for the passing of block signals that depend on the predicted clearing times of the block sections. These recommendations are then used for individual energy optimization of single train runs by avoiding unnecessary braking in front of block signals and maximizing cruising distances. The method is tested and evaluated on a representative line segment of the ELVA, the Railway Signaling Lab at RWTH Aachen University under realistic conditions. Energy consumptions are compared for different prediction time horizons at which the recommendations are available to the train's onboard system. In the examined test case of two trains, the energy-consumption could be decreased by up to 53% compared to operation without any rescheduling system. Thus, the proposed method is able to reduce energy consumption significantly.

This study investigated the climate effect under consecutive winters on the arrival delay of high-speed passenger trains. Inhomogeneous Markov chain model and stratified Cox model were adopted to account for the time-varying risks of train delays. The inhomogeneous Markov chain modelling used covariates weather variables, train operational direction, and findings from the primary delay analysis through stratified Cox model. The results showed that temperature, snow depth, ice/snow precipitation, and train operational direction, significantly impacted the arrival delay. Further, by partitioning the train line into three segments as per transition intensity, the model identified that the middle segment had the highest chance of a transfer from punctuality to delay, and the last segment had the lowest probability of recovering from delayed state. The performance of the fitted inhomogeneous Markov chain model was evaluated by the walk-forward validation method, which indicated that approximately 9% of trains may be misclassified as having arrival delays by the fitted model at a measuring point on the train line. With the model performance, the fitted model could be beneficial for both travellers to plan their trips reasonably and railway operators to design more efficient and wiser train schedules as per weather condition.

Traffic management is crucial for improving the punctuality and reliability of train operations, enabling train operating companies (TOCs) to maintain their competitiveness and further increases the share and profits. A common goal of the train rescheduling problem is to minimize train delays, which fails to examine the results from the perspective of passengers. Moreover, focusing only on the punctuality performance overlooks how the delay is distributed among entities (i.e., trains, passengers, and train operating companies).

We study the train rescheduling problem with the inclusion of passenger choices and the equity concerns. A mixed-integer linear programming (MILP) model is proposed to find the optimal train schedules and the best route for passengers at the same time, with respect to the demanded equity level. Passengers choose a sequence of train services to complete their trip with the least amount of costs (i.e., delays). To evaluate the equity performance of the system, we define equity by means of Gini Coefficient and Maximal Deviation, included in the MILP model as constraints.

Experiments are conducted to explore the impacts of the objective change, i.e., from reducing train delays to reducing passenger delays, and to compare the system performance of using the two equity measures in terms of punctuality and equity. According to the results, the average passenger delay decreases by 34% when minimizing passenger delays, compared with that of minimizing train delays. Moreover, the Gini Coefficient yields less cost of equity (i.e., less increase of delays), compared to that of the Maximal Deviation.

Railway freight wagons have a significant share in the total capital assets of a railway freight company. Due to this fact, the main objective of every company is to maximize the utilization of these resources and on that way minimize its size. In this paper we consider a real-world freight wagon fleet management problem and propose a decomposition approach for optimization of the heterogeneous fleet of flat wagons. The approach has four steps: random container weight generation, optimal container to wagon assignment, empty wagon repositioning and optimal wagon fleet sizing. For the purpose of validation, real-life experiments were conducted based on a rail network composed of 911 origin-destination links and the yearly demand of more than 3 × 10⁶ empty and loaded 20-foot and 40-foot containers. Experimental results show that proposed approach has a practical applicability and that in comparison with existing experience-based practice it represents a significant improvement for the flat wagon fleet management.

Classification yards represent network nodes in the single-wagonload transport system. The processes are complex due to a high number of involved resources and restrictive dependencies. Decisions on job sequencing and resource allocation have a major impact on outbound delays and thus on the quality of service in the network. Due to permanent updates of arrival times and resource availabilities, a constant revision of decisions is necessary. In many cases, considering multiple qualifications of the personnel is crucial for efficient operations. This paper presents an approach for the rescheduling of processes and the assignment of resources in classification yards, which allows to determine best working schedules based on current data such that the cumulative outbound delay of all trains is minimized. Therefore, the paper presents a mixed integer program that includes all essential components (tracks, locomotives and personnel with individual skill patterns). For the real-time capable solution of the optimization problem, four different heuristic approaches based on priority rules are presented. The performance of these approaches is evaluated by a gap analysis with respect to the solutions found by CPLEX. For this purpose, real example data of an operation day of a large classification yard in Germany are used.