Hybrid lectern and poster session on research related to railway capacity and delay modeling. Authors will introduce their work with a 5-minute lectern presentation. Following presentations on each of the 10 papers, posters will be displayed for the remainder of the session with authors available to further discuss the details of their work.
Simulation-Based Method of Capacity Utilization Evaluation to Account for Uncertainty in Recovery Time
Yung-Cheng Lai, National Taiwan UniversityShow Abstract
Kuan-Ting Chen, National Taiwan University
Tzu-Hao Yan, National Taiwan University
Ming-Hua Li, National Taiwan University
A typical objective of a railway agency is to design an efficient and reliable service plan. Although enhancing the level of capacity utilization may increase efficiency, the stability of the service plan may decrease due to stochastic railway disruptions. Past studies used recovery time to evaluate the stability of a service plan; however, the variation of recovery time was not fully examined. Therefore, we developed a Monte Carlo simulation framework to address this issue and proposed four capacity-based indices for the evaluation of efficiency and stability. These indices are capacity utilization efficiency, mean recovery time, dispersion of recovery time, and probability of unacceptable recovery time. The latter two new indices explored the variation of recovery time and probability of unacceptable long recovery time, which is undesired for railway operation. We conducted a real-world case study on the Taiwan Railways Administration before and after the service plan revision. Consequently, the northbound service plan needed a deliberate review and revision before implementation, whereas the southbound plan successfully increased capacity usage while not necessarily worsening stability. This result demonstrated that our study introduced uncertainty analysis into the evaluation framework and provided flexible information on the service plan. These developments can provide improved support in the decision-making of railway agencies to strike a balance between service level and asset usage.
Simultaneous Optimization of Railcar Itinerary and Train Formation Plan
Boliang Lin, Beijing Jiaotong UniversityShow Abstract
Jiaxi Wang, Beijing Jiaotong University
Ruixi Lin, CloudMinds
Chang Liu, Beijing Jiaotong University
Jie Xiao, Beijing Jiaotong University
Siqi Liu, Beijing Jiaotong University
Jianping Wu, Beijing Jiaotong University
An essential issue a freight transportation system faced is how to deliver shipments (OD pairs) on a capacitated physical network optimally; that is, to determine the best physical path for each OD pair and assign each OD pair into the most reasonable freight train service sequence. Instead of pre-specifying or pre-solving the railcar routing beforehand and optimizing the train formation plan subsequently, which is a standard practice in China railway system and a widely used method in existing literature, this paper proposes a non-linear binary programming model to address the integrated railcar itinerary and train formation plan optimization problem. The model comprehensively considers various operational requirements and a set of capacity constraints, including link capacity, yard reclassification capacity and the maximal number of blocks a yard can be formed, while trying to minimize the total costs of accumulation, reclassification and transportation. An efficient simulated annealing based heuristic solution approach is developed to solve the mathematical model. We use a penalty function method to tackle the difficult capacity constraints and a customized method for the operational requirements. To demonstrate the effectiveness and efficiency of the proposed approach, a large-scale real-world case study consisting up to 127 yards and 14,440 shipments after preprocessing is carried out. Computational results validate our method and we have provided the results to railway practitioners as a useful decision support.
A Data-Driven Method for Delay Duration Estimation of High-Speed Train
Meng-Cheng (Jason) Ni, City University of Hong KongShow Abstract
K. L. Tsui, City University of Hong Kong
Yang Zhao, City University of Hong Kong
Punctuality is commonly acknowledged as a key performance indicator for the railway industry. However, the train delays are usually unavoidable due to various kinds of disruptive events. In this paper, data mining and analysis of historical delay describer records is used to analyze and discover causes for major delay events, which is essential for effective operation strategy planning and improving rail management. We develop a data driven method to model the relationship between delay duration and cause, and statistically estimate future delay duration. The estimation results are critical for rail control center’s first-response procedures, as well as for travelers’ information.
Operational Schedule Flexibility, Train Velocity, and the Performance Reliability of Single-Track Railways
Taskin Sehitoglu, HNTB CorporationShow Abstract
Darkhan Mussanov, University of Illinois, Urbana Champaign
Tyler Dick, University of Illinois, Urbana Champaign
Freight shippers and travelers demand a consistent level of service from transportation systems, including railways. Inventory and train connections are more efficiently managed when the total time for railways to move freight or passengers between origin and destination is predictable. One approach to achieve consistency is structured operations where trains are dispatched according to predetermined schedules. By precisely planning the meet and pass interactions to match available track infrastructure, train delay is minimized, train velocity is increased and total runtimes are highly predictable. In North America, the economics of transporting many freight commodities requires a certain amount of schedule flexibility that results in sub-optimal train conflicts and introduces additional train delay. Theoretically, the same total running time and train velocity can be achieved under these flexible operations by increasing the maximum allowable train speed on the line to compensate for the delays. To investigate the equivalency of structured operations at lower speeds versus flexible operations at higher speeds, a representative single-track route was simulated with Rail Traffic Controller. From a baseline minimum-delay schedule, the experiment design increased both maximum allowable speed and schedule flexibility to examine the interaction between these factors and the distribution of runtime and train velocity response for various traffic volumes. Simulation results suggest a slight shift from structured to flexible operations requires a substantial increase in operating speed to maintain runtime and velocity. Decreasing schedule flexibility to facilitate reduced maximum operating speed (and fuel and motive power savings) shows little return until operations become completely structured.
Analysis of Causes and Effects of Primary Delays in a High-Speed Rail System
Chao Wen, University of WaterlooShow Abstract
ZhongCan Li, Southwest Jiaotong University
Javad Lessan, University of Waterloo
Liping Fu, University of Waterloo
Ping Huang, Southwest Jiaotong University
Chaozhe Jiang, Southwest Jiaotong University
Matthew Muresan, University of Waterloo
This paper presents the results of a case study on the causes and effects of typical service disruptions in a high-speed rail (HSR) system in China - Wuhan-Guangzhou High-speed railway (WH-GZ HSR) – a 1096-kilometre HSR line. Ten months of train operation records were used to evaluate the major events causing train service disruptions or primary delay and their cascading or knock-on effects on the operations of other trains. Seven major types of delay events are identified and their impacts in terms of primary delay and number of delayed trains are analyzed. The analysis shows that, regardless the causes of the disruptions, the primary delays follow approximately similar distribution patterns. The overall impact of the disruptions, as measured by the number of trains being delayed, is shown to be largely dependent on disruption type and location. The analysis results from this research provide insights into one of the critical concerns of HSR operations – service disruptions, which is essential for developing robust train schedules and service management strategies.
Statistical Analysis of High-Speed Railway Capacity Utilization and Passenger Distribution in China: A Case Study of Wuhan–Guangzhou High-Speed Rail
Jie Li, Southwest Jiaotong UniversityShow Abstract
Yuxiang Yang, Southwest Jiaotong University
Jing Gan, Southwest Jiaotong University
Chao Wen, University of Waterloo
Ping Huang, Southwest Jiaotong University
Qiyuan Peng, Southwest Jiaotong University
Train operation plan is subject to the distribution of passengers flow, as well as the capacity utilization of railway. The paper concentrates on the capacity utilization and passenger flow characteristics of Wuhan-Guangzhou high-speed railway（WG-HSR）based on the real-word trains operation data.
Firstly, the number of trains departing from each station along WG-HSR is presented, which reflect the service frequency of line. The capacity utilization of each section is analyzed, and the spatial-temporal propagation of capacity utilization is proved.
Then, the spatial-temporal characteristics of passenger flow are presented. In order to get a systematic recognition of passenger flow characteristics, we investigate the passenger volume during different time periods and between several origin and destination (OD) pairs to characterize travelers’ spatial-temporal preferences. The passenger distribution and passenger turnover on some long-distance trains are shown to get accurate the number and proportion of cross-line passengers running on the WG-HSR.
Finally, load rates of trains running on WG-HSR are displayed, which reflect the seats capacity utilization and the economic benefit of trains. The relationship between train load rate and the travelling distance is explored, which can contribute to better passenger organization program and better train operation schedule.
Keywords： Capacity utilization; Passenger distribution；Wuhan-Guangzhou high-speed railway； Load rate
Societal Benefits in Capacity Allocation Mechanisms and Infrastructure Charging Systems for Vertically Separated Railways: Incorporating Economic Assessment into Capacity Allocation and Infrastructure Charging Policies
Aleksandr Prodan, CPCS Transcom Ltd.Show Abstract
Paulo Teixeira, Instituto Superior Tecnico
Capacity allocation mechanisms and infrastructure access charges in vertically-separated railways generally consider only the points of view of the train operator and infrastructure manager. External and other societal costs are not considered in setting these policies and in evaluating their impact on each player. This work proposes a methodology for evaluating capacity allocation and infrastructure pricing policies that uses a Cost-Benefit Analysis together with European Railway Project Appraisal Guidelines to test the impact of a proposed policy on infrastructure managers, train operators, society, and other players in the industry.
A case study of the ScanMed corridor is used, looking at a congested section over the Oresund Bridge. The case study evaluates proposed capacity allocation mechanisms that prioritizing either passenger or freight traffic. The results of this evaluation show the total impact of a particular policy on each player, including society. External costs are also considered in this evaluation. This approach can be by decision makers to make more-informed decisions when setting infrastructure charging and capacity allocation policy.
A Markov Chain Model for Measuring Robustness of Train Schedules
Ismail Sahin, Yildiz Technical UniversityShow Abstract
Trains are subject to random delay-causing influences that may get them deviate from their scheduled paths (primary delay). Trains also interfere with each other causing further delays due to maintaining feasibility of their movements (secondary delay). Time supplements are inserted to train schedules to make up small (primary) delay and to prevent /limit (secondary) delay propagation between trains. Because of the stochastic nature of delay occurrences in schedule realization, a Markov chain model can be developed to represent the process. Ones the one-step transition matrix is developed, the mean first passage times can be computed. They are the indicators how fast the train schedule deteriorates or recovers given its current state. The results can then be interpreted as a measure for timetable robustness. We provide a numerical example to illustrate the workability of the approach.
Simultaneous Rerouting and Rescheduling on Rail Networks Under Weather Impact
Ying Wang, No OrganizationShow Abstract
Raymond S. K. Kwan, University of Leeds
Weather events can cause significant perturbations to the operation of train services. In the UK, weather related train delays run into tens of thousands of hours each year. With the rapid advances in weather forecasting and emerging information technology systems, it is now possible to obtain highly accurate real-time forecasting data. This information can be utilised to improve the operational responses to weather events. In this paper, we discuss the disruptions of dynamic weather fronts on the railways, and map the moving weather fronts in terms of their temporal and spatial distributions on the network and their severity. Depending on the severity of the weather events, the disruption is modelled in terms of reduced speed restriction or track blockage. The weather will only affect the trains which travel through the impacted segment during the impacted time period. We formulate a mixed integer programming model to simultaneously model the rerouting and rescheduling trains under weather impact. We present a case study on a section of the UK East Coast Main Line railway with a thunderstorm moving front which causes disruption. We show that the proposed method could find out optimized result in feasible time.
Predicting Delay Occurrence at Freight Rail Sidings
Juan Martinez Mori, Cornell UniversityShow Abstract
William Barbour, University of Illinois, Urbana Champaign
Shankara Kuppa, CSX Corporation, Inc.
Daniel Work, University of Illinois
Human dispatchers make freight rail dispatch decisions in real-time based on factors such as network traffic, network topology, and train characteristics. These decisions have significant influence on train delays and rail capacity, which motivates the development of tools to predict their effects. This article presents a machine learning framework to predict the occurrence of delay-inducing meets at sidings using an encoding of network state that incorporates information available to dispatchers at the time of prediction. Support vector classifiers (SVCs) are trained and predictions are compared to a simple deterministic baseline technique that uses only location information and treats trains equally. Testing is performed using historical data from a rail network in Tennessee, USA. Preliminary findings indicate that SVCs are able to exploit critical information beyond just train locations that is present in the network state to predict the occurrence of delays at sidings. The SVCs far outperform the baseline technique to which we compare and show that factors such as train length, train priority, and track occupancy have significant influence on delay occurrence.