Analyzing passenger train arrival delays with support vector regression

We propose machine learning models that capture the relation between passenger train arrival delays and various characteristics of a railway system. Such models can be used at the tactical level to evaluate effects of various changes in a railway system on train delays. We present the first application of support vector regression in the analysis of train delays and compare its performance with the artificial neural networks which have been commonly used for such problems. Statistical comparison of the two models indicates that the support vector regression outperforms the artificial neural networks. Data for this analysis are collected from Serbian Railways and include expert opinions about the influence of infrastructure along different routes on train arrival delays.     

Integrated modeling of high performance passenger and freight train planning on shared-use corridors in the US

This paper studies strategic level train planning for high performance passenger and freight train operations on shared-use corridors in the US. We develop a hypergraph-based, two-level approach to sequentially minimize passenger and freight costs while scheduling train services. Passenger schedule delay and freight lost demand are explicitly modeled. We explore different solution strategies and conclude that a problem-tailored linearized reformulation yields superior computational performance. Using realistic parameter values, our numerical experiments show that passenger cost due to schedule delay is comparable to in-vehicle travel time cost and rail fare. In most cases, marginal freight cost increase from scheduling more passenger trains is higher than marginal reduction in passenger schedule delay cost. The heterogeneity of train speed reduces the number of freight trains that can run on a corridor. Greater tolerance for delays could reduce lost demand and overall cost on the freight side. The approach developed in the paper could be applied to other scenarios with different parameter values.     

Rail track infrastructure ownership: investment and operational issues

As privatisation of railway systems reach the political agendas in a number of countries, the separation of track infrastructure from train operations is seen as providing the vehicle which will improve profitability within the rail industry. This paper deals with three main issues related to such separation within a freight railway focus, namely: investment appraisal; track standards and maintenance; and train operating performance. The conflicts of interest between the owners of track and their client operators are discussed in detail. Costs related to track capacity and congestion need to be taken into account, given that additional trains are likely to lead to increased risk of delays to existing services. The paper discusses the use of a travel time reliability model to estimate the additional costs imposed on the system through the introduction of specific train services.It is concluded that investment in individual elements of railway infrastructure must be integrated with the overall financial and customer service strategies of both operators and owners. As an alternative to current practices, a hybrid model of track ownership is put forward here. Under such a model, a joint-venture company with equity from the main ÒplayersÓ would be owner of track. This would allow some of the benefits of vertical integration to be retained, whilst providing fair access to new operators.     

Service-cost tradeoffs for carload freight traffic in the U.S. rail industry

Railroad technology permits a single train to move a large number of individual freight cars. However, cars which are not in dedicated unit train or intermodal service experience considerable delay due to the consolidation and breakup of trains. Rail operations thus involve a tradeoff between the economies of shipment consolidation, and the resulting delays. More direct and/or more frequent train connections will increase costs, but reduce transit times. This article quantifies the cost of providing a range of transit times for general carload traffic for several representative U.S. rail systems. It shows that significant reductions in transit time will require a large increase in the number of train connections and operating cost. Changes in labor contracts to reduce train crew cost will provide some incentive for higher service levels, but reductions in crew cost alone cannot be expected to dramatically improve the performance of the carload segment of the industry.     

Development of a differential route share model for railroad freight traffic

This paper summarizes work undertaken towards development and calibration of a model to predict the distribution of rail freight traffic among competing routes. The model is designed for use in analyzing the traffic effects of changes in the level-of-service on selected rail lines. The model predicts route shares based on the overall network configuration of each railroad participating in a given market. The model selects feasible routes, discards those routes which appear to be too circuitous or costly, and then assigns traffic to the remaining routes in accordance with several network characteristics. It is designed to be sensitive to level-of-service changes, and to simulate the response of shippers and railroads to a competitive environment. A multiple route-finding algorithm was used to find possible routes based on the number of railroads operating at the originating and terminating end of a market. Multiple routes were determined and matched with observed traffic flows from the ICC One-Percent Waybill Sample. Physical network characteristics for each route, including distance, junction frequency, and “impedance,” were calculated from the network model and were correlated with the traffic share observed on each route in the market. A two-stage model was developed to find feasible routes from the set of possible routes and to allocate traffic to feasible routes based on levels-of-service. The model was calibrated on 9,793 routes from 1,199 markets with twenty or more carloads from the 1977 One-Percent Carload Waybill Sample. Model calibration supported the hypothesis that network route characteristics did indeed influence shipper choice of route, and that a normative model could be used to assess relative attractiveness of routes under various railroad corporate ownership restructuring scenarios.     

Train design and routing optimization for evaluating criticality of freight railroad infrastructures

Freight transportation by railroads is an integral part of the U.S. economy. Identifying critical rail infrastructures can help stakeholders prioritize protection initiatives or add necessary redundancy to maximize rail network resiliency. The criticality of an infrastructure element, link or yard, is based on the increased cost (delay) incurred when that element is disrupted. An event of disruption can cause heavy congestion so that the capacity at links and yards should be considered when freight is re-routed. This paper proposes an optimization model for making-up and routing of trains in a disruptive situation to minimize the system-wide total cost, including classification time at yards and travel time along links. Train design optimization seeks to determine the optimal number of trains, their routes, and associated blocks, subject to various capacity and operational constraints at rail links and yards. An iterative heuristic algorithm is proposed to attack the computational burden for real-world networks. The solution algorithm considers the impact of volume on travel time in a congested or near-congested network. The proposed heuristics provide quality solutions with high speed, demonstrated by numerical experiments for small instances. A case study is conducted for the network of a major U.S. Class-I railroad based on publicly available data. The paper provides maps showing the criticality of infrastructure in the study area from the viewpoint of strategic planning.     

Closing the loop in real-time railway control: Framework design and impacts on operations

Railway traffic is heavily affected by disturbances and/or disruptions, which are often cause of delays and low performance of train services. The impact and the propagation of such delays can be mitigated by relying on automatic tools for rescheduling traffic in real-time. These tools predict future track conflict based on current train information and provide suitable control measures (e.g. reordering, retiming and/or rerouting) by using advanced mathematical models. A growing literature is available on these tools, but their effects on real operations are blurry and not yet well known, due to the very scarce implementation of such systems in practice.In this paper we widen the knowledge on how automatic real-time rescheduling tools can influence train performance when interfaced with railway operations. To this purpose we build up a novel traffic control framework that couples the state-of-the art automatic rescheduling tool ROMA, with the realistic railway traffic simulation environment EGTRAIN, used as a surrogate of the real field. At regular times ROMA is fed with current traffic information measured from the field (i.e. EGTRAIN) in order to predict possible conflicts and compute (sub) optimal control measures that minimize the max consecutive delay on the network. We test the impact of the traffic control framework based on different types of interaction (i.e. open loop, multiple open loop, closed loop) between the rescheduling tool and the simulation environment as well as different combinations of parameter values (such as the rescheduling interval and prediction horizon). The influence of different traffic prediction models (assuming e.g. aggressive versus conservative driving behaviour) is also investigated together with the effects on traffic due to control delays of the dispatcher in implementing the control measures computed by the rescheduling tool.Results obtained for the Dutch railway corridor Utrecht–Den Bosch show that a closed loop interaction outperforms both the multiple open loop and the open loop approaches, especially with large control delays and limited information on train entrance delays and dwell times. A slow rescheduling frequency and a large prediction horizon improve the quality of the control measure. A limited control delay and a conservative prediction of train speed help filtering out uncertain traffic dynamics thereby increasing the effectiveness of the implemented measures.     

Probabilistic analysis of the release of liquefied natural gas (LNG) tenders due to freight-train derailments

Liquefied natural gas (LNG) has emerged as a possible alternative fuel for freight railroads in the United States, due to the availability of cheap domestic natural gas and continued pursuit of environmental and energy sustainability. A safety concern regarding the deployment of LNG-powered trains is the risk of breaching the LNG tender car (a special type of hazardous materials car that stores fuel for adjacent locomotives) in a train accident. When a train is derailed, an LNG tender car might be derailed or damaged, causing a release and possible fire. This paper describes the first study that focuses on modeling the probability of an LNG tender car release incident due to a freight train derailment on a mainline. The model accounts for a number of factors such as FRA track class, method of operation, annual traffic density level, train length, the point of derailment, accident speed, the position(s) of the LNG tender(s) in a train, and LNG tender car design. The model can be applied to any specified route or network with LNG-fueled trains. The implementation of the model can be undertaken by the railroad industry to develop proactive risk management solutions when using LNG as an alternative railroad fuel.     

Rail freight efficiency and competitiveness in Australia

In 1994/95, the Australian rail freight task was approximately 100 billion tonne kilometres (btkm). This freight task included some 37 btkm for the haulage of iron ore in Western Australia, 28 btkm for coal haulage in Queensland and New South Wales and about 16 btkm for interstate rail freight. This paper mainly concerns how improvements can be made to the efficiency and competitiveness of interstate rail freight services through the upgrading of sections of mainline track that currently have severe speed‐weight restrictions. Recent improvements in rail freight efficiency are discussed, with emphasis on two indicators: average unit revenues (cents per net tonne km), and average energy efficiency (net tonne km per MJ). Rail freight efficiency is high for the Western Australia iron‐ore operations, Queensland coal operations and Adelaide‐Perth general freight operations. However, between Australia's three largest cities of Melbourne, Sydney and Brisbane, some 36% of the mainline track fails to meet basic fast freight train standards with a ruling grade of 1 in 66 and no curve radius less than 800 m. The contraints on efficient rail freight operations imposed by severe terrain, and how the effects of terrain may be reduced by improved track alignment, are discussed. Some economically warranted rail track investment measures are outlined, including those identified for a National Transport Planning Taskforce. These measures have the potential to reduce liquid fuel use by over 250 million litres a year. Factors affecting competitive neutrality between road and rail freight that are outside of the present scope of Australia's National Competition Policy are broadly considered. These factors include the extensive upgrading of the National Highway System with full Federal funding, and low levels of road cost recovery from heavy trucks operating over long distances.     

Stochastic prediction of train delays in real-time using Bayesian networks

In this paper we present a stochastic model for predicting the propagation of train delays based on Bayesian networks. This method can efficiently represent and compute the complex stochastic inference between random variables. Moreover, it allows updating the probability distributions and reducing the uncertainty of future train delays in real time under the assumption that more information continuously becomes available from the monitoring system. The dynamics of a train delay over time and space is presented as a stochastic process that describes the evolution of the time-dependent random variable. This approach is further extended by modelling the interdependence between trains that share the same infrastructure or have a scheduled passenger transfer. The model is applied on a set of historical traffic realisation data from the part of a busy corridor in Sweden. We present the results and analyse the accuracy of predictions as well as the evolution of probability distributions of event delays over time. The presented method is important for making better predictions for train traffic, that are not only based on static, offline collected data, but are able to positively include the dynamic characteristics of the continuously changing delays.     

Assessing the effects of several variables on freight train fuel consumption and performance using a train performance simulator

The impact of several variables on freight train fuel consumption and performance are assessed using a train performance simulator. These variables include: wind, precipitation, number of cars, and number and type of locomotives. The input to the train performance simulator includes data related to train characteristics and data related to external conditions such as weather. The simulator output represents fuel consumption expected under a given set of conditions. Graphical and cost/benefit approaches were used to assess operations alternatives.     

Automated real-time railway traffic control: an experimental analysis of reliability,resilience and robustness

Railway transportation provides sustainable, fast and safe transport. Its attractiveness is linked to a broad concept of service reliability: the capability to adhere to a timetable in the presence of delays perturbing traffic. To counter these phenomena, real-time rescheduling can be used, changing train orders and times, according to rules of thumb, or mathematical optimization models, minimizing delays or maximizing punctuality. In the literature, different indices of robustness, reliability and resilience are defined for railway traffic. We review and evaluate these indices applied to railway traffic control, comparing optimal rescheduling approaches such as Open Loop and Closed Loop control, to a typical First-Come-First-Served dispatching rule, and following the timetable (no-action). This experimental analysis clarifies the benefits of automated traffic control for infrastructure managers, railway operators and passengers. The timetable order, normally used in assessing a-priori reliability, systematically overestimates unreliability of operations that can be reduced by real-time control.     

A dynamic approach to road freight flows modeling in Spain

This paper presents an innovative approach to analyzing road vehicle freight traffic that uses a dynamic panel data specification derived from a gravity model. This dynamic approach, which has recently been employed in international goods trade models in lieu of the traditional static specification, is applied to the case of Spain using data for the countrys 15 NUTS-3 regions between 1999 and 2009. Using the system general method of moments approach, we obtained significant evidence that the flow of vehicles carrying commodities by road has a strong persistence effect when controlling for unobserved heterogeneity. We also found that the quality of road transport infrastructure has a significant impact on vehicle trips. According to our findings, we suggest that this type of specification be employed in distribution models in which fixed effects and lags of the dependent variable are included to account for unobserved heterogeneity and persistence effects, respectively.     

The Benefits for Rail Freight Customers of Vertical Separation and Open Access

Abstract

This paper reviews and analyses the benefits for rail freight customers of the two principal models for introducing competition in main line railway networks: (1) the vertical separation of infrastructure from operations; and (2) the introduction of competition providing other operators with open access to the network. The paper shows that traffic growth has generally been higher in those European countries that have liberalized most. It then examines in detail the impact of restructuring in Great Britain in terms of the competition, freight rates and traffic growth. The paper concludes that vertical separation benefits freight customers more than just open access.     

Large-scale network distribution of pooled empty freight cars over time,with limited substitution and equitable benefits

A set of models and procedures is described for finding the optimal distribution of empty freight cars owned by the railroads participating in a pooling agreement. A distinction is drawn between a system focus, in which the emphasis is on minimizing total cost, and a company focus, in which the benefits of the agreement to the individual railroads are emphasized. Limited car substitution is accounted for by combining interchange costs with distribution costs, and incorporating interchange possibilities and prohibitions into the network structure. Temporal variations in car supply and demand levels are also taken into account. A large-scale network algorithm is used in conjunction with decomposition to obtain solutions which show for a given time horizon how much equity can be achieved in the balance of savings among the railroads involved and at what cost. Results using actual operating data are reported.     

Direct to your destination: the size,scope and competitive status of express coach carriers in the United States

The advent of express coach bus lines offering guaranteed seating and emphasizing curbside pickup and drop-off is contributing to a revival in intercity bus travel in the United States. Relatively little is known, however, about the scale and geographic scope of these carriers or the competitive landscape in which they operate. To fill this void, this study evaluates the service networks operated by the two largest express coach operators in the country, BoltBus and Megabus, and evaluates a data set of 4775 fares sold on megabus.com. The results show that these carriers cumulatively serve 127 intercity segments and operate about 52.9 million bus miles per year. Together, these carriers have grown to about one third of the size of Amtrak, with Megabus and BoltBus providing 3.3 billion and .69 billion seat miles of service, respectively, compared to Amtrak’s 12.8 billion. With respect to the types of routes it serves and the competition it faces, Megabus has evolved into a carrier quite different than Boltbus; more than one third of Megabus’ bus miles are operated on segments without Amtrak service, while virtually all of BoltBus’ miles face this competition. The analysis of Megabus’ pricing shows that fares rise modestly within 2 weeks of departure, while the per-mile costs are much less ($.08/mile) for 300–399 mile trips than for those 50–99 miles ($.22/mile). Nevertheless, the dispersion of fares tends to fall as the departure date nears, regardless of distance. Together, these prominent bus lines serve 66 of the 100 most heavily traveled U.S. city pairs that have characteristics suitable for intercity bus service—which is more than Amtrak. With further growth on the horizon, planners, federal regulators, and researchers should collaborate on establishing reporting requirements for this expanding sector.

     

Reinforcement learning approach for train rescheduling on a single-track railway

Optimal rail network infrastructure and rolling stock utilization can be achieved with use of different scheduling tools by extensive planning a long time before actual operations. The initial train timetable takes into account possible smaller disturbances, which can be compensated within the schedule. Bigger disruptions, such as accidents, rolling stock breakdown, prolonged passenger boarding, and changed speed limit cause delays that require train rescheduling. In this paper, we introduce a train rescheduling method based on reinforcement learning, and more specifically, Q-learning. We present here the Q-learning principles for train rescheduling, which consist of a learning agent and its actions, environment and its states, as well as rewards. The use of the proposed approach is first illustrated on a simple rescheduling problem comprising a single-lane track with three trains. The evaluation of the approach is performed on extensive set of experiments carried out on a real-world railway network in Slovenia. The empirical results show that Q-learning lead to rescheduling solutions that are at least equivalent and often superior to those of several basic rescheduling methods that do not rely on learning agents. The solutions are learned within reasonable computational time, a crucial factor for real-time applications.     

Prediction of arrival times of freight traffic on US railroads using support vector regression

Variability of travel times on the United States freight rail network is high due to large network demands relative to infrastructure capacity, especially when traffic is heterogeneous. Variable runtimes pose significant operational challenges if the nature of runtime variability is not predictable. To address this issue, this article proposes a data-driven approach to predict estimated times of arrival (ETAs) of individual freight trains, based on the properties of the train, the properties of the network, and the properties of potentially conflicting traffic on the network. The ETA prediction problem from an origin to a destination is posed as a machine learning regression problem and solved using support vector regression trained and cross validated on over two years of detailed historical data for a 140 mile section of track located primarily in Tennessee, USA. The article presents the data used in this problem and details on feature engineering and construction for predictions made across the full route. It also highlights findings on the dominant sources of runtime variability and the most predictive factors for ETA. Improvement results for ETA exceed 21% over a baseline prediction method at some locations and average 14% across the study area.     

Conflict Resolution through Negotiation in a Railway Open Access Market: a Multi-agent System Approach

Abstract

Open access reforms to railway regulations allow multiple train operators to provide rail services on a common infrastructure. As railway operations are now independently managed by different stakeholders, conflicts in operations may arise, and there have been attempts to derive an effective access charge regime so that these conflicts may be resolved. One approach is by direct negotiation between the infrastructure manager and the train service providers. Despite the substantial literature on the topic, few consider the benefits of employing computer simulation as an evaluation tool for railway operational activities such as access pricing. This article proposes a multi-agent system (MAS) framework for the railway open market and demonstrates its feasibility by modelling the negotiation between an infrastructure provider and a train service operator. Empirical results show that the model is capable of resolving operational conflicts according to market demand.     

Railway infrastructure disturbances and train operator performance: The role of weather

In this paper, we estimate the effects of weather conditions such as wind, temperature and precipitation on railway operator performance of passenger train services. We distinguish between the direct effects of weather conditions and the indirect effects through disturbances in infrastructure. We show that certain types of bad weather mainly affect train operators’ performance indirectly, through their effect on infrastructure. Furthermore, we show that the welfare losses for passengers confronted with increased cancellations of trains and decreased punctuality in The Netherlands due to one standard deviation increase in infrastructure disruptions amount to about €80 million per year.