Urban Rail Transit (URT) accidents not only seriously affect the safety and reliability of its operations, but also reduce service level to passengers. Based on historical URT accident data, this study develops a hybrid ensemble learning model based on a Convolutional Neural Network (CNN) and Bidirectional Encoder Representations from Transformers (BERT) for predicting accident consequences in URT. The CNN is employed to capture spatial patterns from the diverse accident data, while the BERT is applied to learn complex relations in accident text descriptions. The results of the two models are combined for classifying accident consequences. The proposed hybrid ensemble learning model was applied to predict accident consequences in Chongqing’s URT using historical accident records. It achieved a prediction accuracy of 0.805 on testing data set, which is at least 20% higher than that of commonly used machine learning models, including multilayer perceptrons, support vector machines, and Bayesian networks. Furthermore, the reapplication of the proposed model to historical accident records of the URT in Chengdu demonstrates the generalizability and reusability of the model. This study forecasts the consequences of URT accidents with high accuracy using limited historical data, which supports operators in identifying high-frequency and high-impact accidents. Consequently, targeted maintenance and timely emergency response strategies can be developed to decrease accident rates and mitigate the impacts.

Railway alignment design is a crucial but difficult task that should trade off many objective factors. Currently, although several Multi-objective Intelligent Alignment Optimization (M-IAO) methods have been proposed, previous methods may still be limited by low convergence performance with more than three objectives. In response, a Many-objective Intelligent Alignment Optimization (Ma-IAO) method is presented in this paper. First, a six-objective Ma-IAO model is built considering three kinds of objective factors in railway design, namely economic, geologic and ecologic factors. Then, a Particle Swarm Optimization with Strengthened Pareto Dominance Analysis (SPDA-PSO) is developed to solve the model. Two types of external archives are primarily designed to store and update nondominated solutions during optimization with a specifically-proposed strengthened pareto dominance criterion (known as LC-dominance). Afterward, two elite selection operators are devised to search for Pareto corner and knee points within external archives as the best particle individuals for guiding SPDA-PSO evolution. Lastly, the proposed method has been applied to a complex real-world railway example. Through comparisons with a contemporary M-IAO method and the manual work of human designers, its effectiveness is confirmed via detailed data analyses.

Railway major technical standards (RMTSs) are prerequisites for subsequent railway design processes, which have fundamental influences on controlling railway transport capacity and investment. However, the traditional manual decision process is laborious and time-consuming. It also has difficulties in quantifying the interrelations among different standard parameters. In this paper, we propose a so-called voting election framework for efficient RMTS decision-making, in which three stages, namely constituency partition, constituency election and ballot summary, are developed. Then, for constituency election, four types of multi-task neural networks, namely, cascaded, overall parallel, grouped parallel and hybrid multi-task neural networks, are designed to identify the interrelations among six kinds of RMTSs. Besides, hyperparameter fine-tuning experiments are conducted to optimize the performance of the proposed networks. Finally, based on a dataset comprising realistic railway cases, it is found that the grouped parallel multi-task neural network performs the best by achieving an average decision accuracy of 86.9% compared to manual decision results and by outperforming the previous single-task neural network method by 10.7% in average decision accuracy.

Roads, railways and rail transit lines play major roles in transporting passengers and freight. In the design process for such transportation infrastructures, the alignment determination greatly affects their life-cycle performances. However, due to the great complexity associated with alignment design, conventional manual work is very time-consuming and laborious, which consequently prompts the flourishing development of intelligent alignment optimization (AO). Hitherto, with considerable invested efforts, numerous remarkable AO studies have been reported. In this work, we present the first-known literature review paper in this field from the past 25 years. First, a bibliometric visualization is conducted to show the overall characteristics of AO research. Then, existing mathematical models that formulate AO problems are reviewed. Particularly, GIS-based applications for AO are also analyzed in that part. Afterward, the intelligent solution methodologies for solving AO models are discussed. Moreover, two specific research topics, i.e., concurrent optimization of railroad alignments and station locations as well as the existing alignment recreation and redesign, are provided. Finally, we propose 12 possible research extensions and directions in this realm.

Station elevations and vertical alignment between adjacent stations often influence each other. Therefore, they should be optimized concurrently. However, existing studies mainly focus on optimizing station elevations or links separately. In this paper, a concurrent optimization model of stations and links is developed, in which the vertical points of intersection in stations and links are design variables. The comprehensive cost including construction, energy consumption, and travelling time, is the optimized objective function. Particle swarm optimization (PSO) is a method for searching in continuous space and is widely used for alignment optimization. Considering the characteristics of concurrent optimization for stations and links, the PSO algorithm is improved by modifying the updating formulas for particles and designing two strategies for particle updating, namely “Stations before Links” and “Links before Stations” strategy. A dynamic adaptive feasible region is proposed to handle the complex constraints during optimization. This method is applied here to a real-world case. The applications demonstrate that this method can automatically generate vertical alignments which jointly optimize the locations of stations and links and satisfy all the complex constrains.

The flow‐capturing problem (FCP) consists of locating facilities to maximize the number of flow‐based customers that encounter at least one of these facilities along their predetermined travel paths. The FCP literature assumes that if a facility is located along (or “close enough” to) a predetermined path of a flow of customers, that flow is considered captured. However, existing models for the FCP do not consider targeted users who behave noncooperatively by changing their travel paths to avoid fixed facilities. Examples of facilities that targeted subjects may have an incentive to avoid include weigh‐in‐motion stations used to detect and fine overweight trucks, tollbooths, and security and safety checkpoints. This article introduces a new type of flow‐capturing model, called the “evasive flow‐capturing problem” (EFCP), which generalizes the FCP and has relevant applications in transportation, revenue management, and security and safety management. We formulate deterministic and stochastic versions of the EFCP, analyze their structural properties, study exact and approximate solution techniques, and show an application to a real‐world transportation network. © 2014 Wiley Periodicals, Inc. NETWORKS, Vol. 65(1), 22–42. 2015