The article explores the problem of train rescheduling based on the actual situation. The proposed stochastic model uses specific distributions of operating times which are dependent on the current traffic conditions. The arrival time distribution is considered as a result of adjusting the train trajectory by speed control. The results of modelled arrival distributions correspond well with the experimental data received at the russian railways. The proposed model is used for prevention of sequence-of-trains conflicts and violations of connections. The basis of deviation prediction is two-train model of mesa-level which uses actual features of scattering of the operation times both at sites and at stations. The article also proposed a new measure of arrival delay which considers the share of satisfied passengers.

Over the last twenty years, the railways in Britain have seen rapid growth, with patronage doubling and the level of train service increasing by 50%. However, these successes have also led to challenges. In particular real unit costs are estimated to have increased by 50%, with particularly marked increases in infrastructure renewal and enhancement costs. Against this background, the University of Southampton have been leading the Track 21 and Track to the Future projects, funded by the Engineering and Physical Sciences Research Council. These projects are assessing how to make ballasted track systems more durable given higher traffic levels and how to reduce the costs of maintenance and renewal. Track 21 considered a number of engineering interventions of which this paper focusses on one, under sleeper pads (USPs). The cost implications of this intervention are assessed for the South West Main Line using an industry-specific model, VTISM (vehicle track interaction strategic model), adapted in two ways. First, the results of laboratory experiments from specialist test rigs are incorporated into the model. Secondly, the wider effects of increased reliability, improved ride quality and reduced vibration are also considered within a cost-benefit analysis framework. It is found that USPs can lead to substantial financial savings and, depending on assumptions concerning noise impacts, are likely to have substantial wider social benefits.

Promoting rail systems can represent a useful policy for rebalancing modal choices and reducing private car use, especially in high-density contexts. Obviously, an increase in passenger numbers is only possible if generalized costs (i.e. a weighted sum of times and monetary costs) associated to public transport are abated. According to the recent literature and current professional practice, most strategies for achieving this objective are based only on infrastructural interventions which may be unfeasible or inadequate in densely populated contexts. Likewise, the adoption of policies based on replacing existing fleets or reducing fare levels entails increases in national or regional subsidies, which would be difficult to achieve in the current economic climate. Hence, our proposal is based on investigating effects on travel demand arising from the replacement or upgrading of existing signalling systems (both in terms of trackside and on-board equipment). Indeed, the recent European Union policy to create a single transnational interoperable rail network imposes the development of innovative signalling systems. In this context, since cost–benefit analysis has to be implemented to verify the economic and environmental feasibility of the proposed intervention strategy, an appropriate method should be developed to estimate passenger flows according to future configurations. In this article, we propose a method to determine travel demand in current and future contexts by appropriately processing data from Italy’s national census on mobility, population growth forecasts and turnstile counts. The proposed approach is applied to the regional Naples–Sorrento rail line serving the metropolitan area of Naples in southern Italy in order to show its feasibility.

The choice of university studies is one of the main socio-economic categories that is affected by the quality of public transport services and in particular rail transport. This category is characterized to cover medium to long distances to reach, every day, the place of study. for this reason, among the urban sustainable policies, the mobility choices of the university students have long been analysed in literature. The aim of this paper was to estimate the effect of railway accessibility on the choice of university studies.

The university of Naples federico II (Italy) is taken as the case study. In the spring of 2014, using the database obtained from the registry office of the university the following were investigated:

– the origin of the daily trips (residential address);

– the trip frequency and the departure/arriving time;

– the modes or sequence of modes used;

– the perceived public transport quality (e.g. travel time, comfort).

To evaluate the catchment area of the university and for analysing what attributes influence the choice of university studies (in particular the ‘weight’ of a specific railway accessibility), a regression model was used. among the model attributes educational course variables, home–university distance and a constant-specific attribute were also considered.

The result of the study shows the surprising role of railway accessibility in the choice of university studies; to travel from a station within 900 m from home is equivalent (in terms of perceived utility) to travel every day a distance greater than 28 km to reach the university.

The costly construction and operation of urban rail transit have become escalating problems for cities worldwide, especially in developing countries. Reliable measures of residential property appreciation for urban rail transit can provide suggestions for policy-making of value capture to fund transit improvements. Using GIS techniques and residential property price data, the relationships between accessibility improvement value and residential property appreciation are analysed in Shanghai. The impacts of urban rail transit on residential property values are classified into traffic effect and agglomeration effect, both of which are measured by the accessibility improvement model. The results indicate that the goodness-of-fit of the model is greater than 93%. Traffic benefit is greater than agglomeration benefit in the suburb, which is completely different in the city centre. furthermore, the residential property appreciation is about 5 times the accessibility improvement value per year. This study contributes to the evidence of capitalization impacts of public transit from a booming and transitional economy and provides suggestions for land use planning of areas surrounding stations.

In Japan, seating plans of reserved/non-reserved seats in intercity express trains do not necessarily correspond to passenger demands which may vary depending on many factors such as calendar days and operating sections. This mismatch possibly causes passengers’ inconvenience due to passenger congestions on board, and railway operators can lose revenue. Optimizing seating plans is expected to be an effective solution for these problems, where predicting both potential passenger demands and reflections to the potential demand caused by the passenger congestions is necessary. For establishing the prediction methods, we conducted some demand analyses and some questionnaire surveys. according to these results, we establish demand estimation models of reserved/non-reserved seats which can consider potential demands and behaviour changes if the passengers face congestion. We developed a prototype of seating plans optimization system applying these models, simulation of passengers’ behaviour and genetic algorithm. This paper describes these models, the optimization system and the result of applying this system to some existing intercity express trains as a case study. We got the optimized seating plan which is expected to improve both of the passengers’ convenience and the railway operators’ revenue.

A new era of automation in rail has begun offering developments in the operation and maintenance of industry standard systems. This article documents the development of an architecture and range of scenarios for an autonomous system for rail maintenance planning and scheduling. The Unified Modelling Language (UML) has been utilized to visualize and validate the design of the prototype. A model for information exchange between prototype components and related maintenance planning systems is proposed in this article. Putting forward an architecture and set of usage mode scenarios for the proposed system, this article outlines and validates a viable platform for autonomous planning and scheduling in rail.

It is important for railway operators to make suitable timetables. On the assumption that passengers’ train paths are invariant even if the timetable changes, the timetables are determined according to the various statistical data under the current timetable. In planning the timetable, it is difficult to grasp the change of passengers’ paths due to timetable modifications because it is too complex. In this paper, we propose a framework of timetabling with due consideration of the path change and propose a practical timetabling system which can quickly estimate the congestion of each train by using passengers’ origin–destination (OD) data collected by automatic ticket checkers. With the system, timetable planners can interactively make and modify timetables by trial and error while confirming the congestion, and finally, they can reach the most preferable one. In order to realize such an interactive system, it is important to develop a fast estimation algorithm of train congestion. We developed a new shortest path search algorithm to determine which trains each passenger gets on. The algorithm devised based on the Dijkstra method has two features. First, the shortest path search from each node to all the destination stations in the composed graph network is executed only once. Second, overlapping the path searches are omitted using stored information of the shortest paths which have already been searched. By applying this algorithm, it took only about 10 s to estimate the train congestion under a timetable of the whole day on an urban commuter line.

This article describes how to maximize the number of freight trains through global networks. Nowadays, the capacity of railway lines and also of railway nodes can be calculated by using analytic algorithms. Currently there is no generally accepted method to allocate the overall capacity of lines and nodes of such networks by using analytic algorithms. However, to maximize the number of train runs, the rail passenger service will be fixed on the scheduled train course so that the rail freight service can use the remaining capacity on a planned or re-routed train course. Existing or detected bottlenecks could be eliminated by means of sensible re-routing due to optimization, which will reveal the best train paths through the network. The article concludes with an illustrative computation for a generic railway sub-network to exhibit how the optimization is working.

Train rescheduling means a transient situation to correct a train diagram in a suspended state due to traffic accidents or disasters. Automatic (or half manual) rescheduling has been studied, and such previous research has shown promising results.

In this paper, we describe a train timetable rescheduling method. In Japanese urban areas, some private companies operate on each other’s tracks. If vehicle types have limitations due to ground facilities or company rules, the vehicle has to be operated under these limitations even when running on a rescheduled timetable. Even if the ground facilities of different companies have uniform conditions, local and rapid trains must be operated in a distinct manner.

Therefore, we suggest a rescheduling method. With this method, each vehicle type and its vehicle routes based on the track layout are registered, and rescheduling diagrams are composed with the route combinations.

Important conditions to decide the combinations are vehicle location at the operation resumption time and the introduction of same-type vehicles at an originally unscheduled timing. We compare the traffic effects for some combinations of the latter situation where originally unscheduled same-type vehicles are introduced for rescheduling. The evaluation values are average headway time and its standard deviation at all stations on the timetable.

We apply our rescheduling method to a theoretical line and timetable modelled on existing urban lines in Japan where trains go and come back on double tracks, and indicate the efficacy of our rescheduling method.

Railway operation can be perceived as a complex system consisting of thousands of constituent elements, including demands for transportation and a variety of transportation resources, such as tracks, track blocks, stations, sidings, trains, crews, etc. Overall behaviour of a railway operation is characterized by unpredictable disruptive events such as changes in availability of resources due to failures, weather conditions or human errors. In addition transportation demands tend to change over time whilst changing transportation resources, such as tracks, is not always possible or practical. The key task of railway management is the allocation of transportation resources to transportation demands with a goal of achieving a complete match, ensuring a smooth operation. The difficulty of this task primarily depends on the variability of demand and reliability of resources, in particular, tracks, trains and human resources. The paper describes how to design complex adaptive railway schedulers, which allocate resources to demands in real time and ensure rapid rescheduling in reaction to unpredictable disruptive events.

The aim of this article is to study how selected variables influence delays in train traffic. Data has been collected on train movements, timetables, weather and capacity utilization on a highly utilized single-track railway line in southern Sweden during 2014. Based on this dataset, we have analysed how different factors affect delays in passenger traffic. We measure delays in a novel way, as deviations from the scheduled duration for each line section and station stop, not as deviations from a published or operational timetable, and this allows us to identify when and where the delays first occur. Average delays were much larger at station stops. The most significant factor affecting delays was the scheduled duration time at station stops and the existence of margins on line sections. If trains arrive to a line section or station stop slightly delayed they speed up the activity, otherwise they are typically delayed. The influence of weather was less significant and somewhat contradictory: snow and cold temperatures increase delays on line sections but reduce them at station stops, while precipitation made no difference. Capacity utilization seems to have a negative correlation with delays, but we have too little variation in the levels to be confident. All studied variables, except for precipitation, have impacts that are statistically significant to a very high degree of confidence, using both t-tests and regression analysis. The results of this study have important practical implications for timetable construction; for instance we estimate that a reallocation of scheduled time at stations could reduce delays by as much as 80%.

In railway lines where trains are running densely, a small increase of a dwell time causes a delay and the delay tends to expand and propagate to many other trains. One idea to avoid such delays is to increase dwell times of trains in timetables. Increase of dwell times, however, may cause an expansion of headways and we may have to decrease the frequency, which is never allowed in a congested railway line. In this article, we propose a procedure to get a timetable in which each train is given an enough amount of dwell time without reducing the frequency of trains. Our key idea is to find a train performance curve which minimizes the departure–arrival interval of trains and to make use of the generated time to increase the dwell times. In this process, we do not stick to an idea that travel times of trains have to be minimized. Following this procedure, we can increase the dwell times without reducing the frequency. We have applied our procedure to actual timetables and confirmed that our procedure works very well.

In strategic subway scheduling stage, the conflict sometimes comes from different requirements of the subway operator. This study aims to investigate the significant factors concerning strategic subway scheduling problem and to develop an automatic procedure of feasibility analysis in subway scheduling. To this end, accurate simulation of train movement (via a simulator, named HAMLET) is applied first by considering the line geography, train performances, actual speed restrictions, etc. The critical elements of subway scheduling and their correlations are then studied and a bound structure of the critical factors is established. The feasibility of primary plan requirements is analysed with the restrictions of the bound structure. Infeasible aspects and possible adjustments are shortly discussed. Finally, the subsequent applications including schedule generation and optimization according to various objectives are indicated as well.

This paper deals with an optimization approach to railway track allocation, which is one of the most important decision problems in the presence of multiple train operating companies (TOCs). In fact there has been deep discussion about how to boost the competition environment in Korean railway since the functional reform in 2004, which at last resulted in introducing a new entrant to high-speed railway passenger transportation market. Finally, in August 2016, two operating companies will compete on the major routes in the Korean high-speed railway network. The infra manager, KR Network, who is responsible for allocating the slots, has been developing their own allocation procedure which partly uses an optimization model for adjusting the times of requested train-paths. But one of the TOCs’ concerns with respect to the adjustment is that their train-set routing plan could be in disorder by the adjustment of the arrival/departure times. Assuming TOCs submit their routing plan as well as their desired train-paths, we present an optimization model and algorithm for track allocation problem, considering the routing plan requested by TOCs. The model is developed on a time-space network, where a train-path can be described as the sequence of the arcs. Based on the network, we developed an column-generation approach to dynamically generate the promising train-paths for each requested one so as to maximize the total profit while preventing the routing plans from disrupting by means of setting up the arcs only among the two successive train-paths in the routing plan. Also we present the experimental results applied to the Korean high-speed railway network.

Convenience is an important requirement of a scheduled transit system. The authors proposed a new method to evaluate a train schedule by using entropy of the information theory. By the aspect of statistical mechanics, the randomness of a schedule is associated with the convenience of the schedule. However, the way of using entropy requires special conditions. In this paper, we show that we can relax the conditions by interpreting the entropy as information.

First, we explain the representation of a schedule by a tree structure. Introducing the tree structure, we recognize the more composite construction of schedules. The tree structure leads to the concept of conditional entropy. Using the concept of conditional entropy, we can evaluate schedules of the states of different places or times. Secondly, we can understand what the information of the schedule is. We can grasp that the information is the benefit of knowing the schedule. The increase of the entropy corresponds to the increase of the benefit. Moreover, it becomes possible to express an unexpected disbenefit by the entropy with the loss of information. In other words, the benefit is a positive entropy and the disbenefit is negative entropy. Finally, we show that we are able to deal with an influence of wasteful time in a schedule, such as an unnecessary waiting time and an extra duration by using the entropy. It is very important that we can evaluate negative effects, namely the discrepancy of supply and demand from the point of view of time.

Nowadays, the most commonly used configuration to supply high-speed trains is 2 × 25 kV power supply system. The location of ground faults in 2 × 25 kV power supply systems is a difficult task, since the use of distance protection relays to localize ground faults positions doesn’t work properly as the relation between the distance and the impedance seen by the distance protection relays is not linear and therefore the location is not accurate enough. A new simple and economical method to identify the subsection between autotransformers (ATS) and the conductor (positive or negative) where the ground fault is happening, based on the comparison of the angle between the current and the voltage of the positive terminal in each autotransformer, was developed recently. Consequently, after the identification of the subsection and the conductor with the ground fault, only this subsection where the ground fault is present, will be quickly removed from service, with the minimum effect on rail traffic. The high-speed trains demand a power about 12–16 MW, hence a significant current flow through the conductors of the 2 × 25 kV power system. This paper presents a study about the influence of the current consumed by the high-speed trains on this novel ground fault identification method. The operation of the method is correct even with the high-speed train currents in the section with a ground fault. This fact has been validated through numerous computer simulations, obtaining excellent results.

Electrification systems applied to railway systems have high complexity to achieve the optimum in several terms. These terms could be separated in energy supply and budget costs, such as direct and indirect costs, all of them derived from the implantation of the electrification system along the railway line. Starting from the experience in several projects carried out in CITEF (Railway Research Center in Technical University of Madrid), a new research way has grown up. The main goal has been an expert system, which is able to improve substantially the design process of power supply systems in railways, such as in AC as DC. Bearing in mind the specific constraints in terms of budget and electric standards, the expert system is able to determine a set of possible design scenarios. Each of them has the number, type and location for the main elements (catenary, substation, autotransformer, etc.). In this article a step forward is presented. This new step would apply a better way to choose between elements, hence taking into account the relationship between lifecycle versus element cost or losses avoided in the infrastructure. This has been applied to the main elements like traction substations, autotransformers, catenary and impact on maintenance and environmental zones. This new concept for evaluation has been integrated in the objective functions. These functions are the managers for the guidance of the optimization method, in this case genetic algorithm AMGA-II, in order to achieve the final Pareto Front. This Pareto Front contains the several possibilities which the designers must evaluate to accomplish their goals to power supply system.

Energy efficiency in railway operation has been a topic of high interest for years. Lowering energy costs can make railway operation not only more sustainable, but also more competitive towards other modalities. Energy efficiency can be reached in two ways: system efficiency and operational efficiency. Innovation can be found in combining the right system measures with the right operational measures. Introducing the right system measures can even lead to energy efficiency independent of operational strategies, such as coasting and regenerative braking.

Rail infra providers struggle with the question which innovations – system, operational or both – will bring them closer to their targets. To determine which measures will contribute best, the authors developed a new universal method based on energy efficiency at three levels: individual rolling stock performance, rolling stock performance in networks and operational strategies.

The authors used this three-level method to study the energy efficiency effect of a system change from the current 1,500 V_DC to a 3 kV_DC traction power supply system, including operational strategies. The results of the study revealed that in the Netherlands this system change will result in a return on investment (ROI) within 15 years and a reduction of energy consumption ranging up to 20%.

We consider energy-efficient scheduling of freight trains on a line with single-track sections, as is the case on many freight railways routes in North America. We determine the train speed that minimizes the energy consumption, taking into account the departure and arrival times and exploiting the waiting times. The proposed energy consumption model takes into account the ground topography, the speed, the number of axles the axle load and the type of locomotives. Some experiments are conducted in order to estimate the energy gain when adapting the train speed, using Canada Pacific Railway (CPR) data.

Stray current corrosion in direct current (dc) transit systems occurs because of the mechanism of current transfer between metals and a conductive electrolyte such as concrete, soil and water. Stray current reactions can be considered as a special case in that the anode (point of current discharge) may be at a considerable distance from the cathode (point of current pickup). The risk of stray current corrosion arising from the operation of dc-powered transit system is difficult to eliminate completely. However, suitable design of dc traction power systems and structures carrying the railways can significantly reduce the risk of corrosion both to the transit system structures and third-party structures.

Stray currents can cause safety risks, thus making the design of stray current mitigation, testing and maintenance an important element of the holistic design for a dc transit system. based on the results of the literature research, interviews with over two dozen dc rail transit systems, and testing of dc rail tracks, this paper presents and analyses various mitigation methods currently in use in the industry to control stray current corrosion. Recommendations for the testing (monitoring) and maintenance procedures to keep the stray current leakage and the related corrosion in control are then presented.

To find out the relationship between the track structure parameters and rail corrugation type and put forward a reasonable maintenance threshold for the amplitude of rail corrugation, surveys on rail corrugation by rail static and dynamic tests were carried out on Shanghai Metro recently. Based on the in situ data, the wavelength and amplitude of rail corrugation were calculated and analysed by different categories according to their service conditions. Investigation data and its analysis show that different track structure parameters, such as ballast, fastener, curve radius, etc., can affect the amplitude and wavelength in pretty different ways. In order to know the development trend in rail corrugation, statistical significance on the distribution of amplitude and dynamic data is calculated, the distribution of amplitude at certain wavelengths is figured out, and accordingly a reasonable threshold of the amplitude maintenance is put forward.

The article describes the preliminary results and conclusions of experimental monitoring of the thermal regime of railway track structure performed on the experimental stands (railway track models in 1:1). The purpose of this study was to optimize the structural thickness of the protective layer in the subgrade structure. The first part of the study describes the experimental stand design and methodology for monitoring the heat and humidity variations in the railway track structure. The following parts describe the results of verification of the thermal characteristics of the material incorporated into the model of the railway track and of its subgrade.

A computational model to solve for the dynamic interaction of a high-speed train and railway structure including derailment during an earthquake is given. An efficient mechanical model to express contact–impact behaviours between wheel and rail before derailment, between wheel and the track structure after derailment, and between the car body and railway structure during and after derailment is presented. The motion of the train with nonlinear springs and dampers is modelled in multibody dynamics. The railway structure is modelled with various finite elements. The combined dynamic response of the train and railway structure during an earthquake is obtained by solving the nonlinear equations of motions of the train and railway structure in the modal coordinates subjected to the interaction between the train and railway structure. Numerical examples of a high-speed train running on the railway structure after derailment during an earthquake are demonstrated.

Collapses due to underground cavities have been drastically increasing in urban areas of South Korea. This article establishes a statistical model to assess the risk potential of railroads with respect to under-ground cavities. The authors first identified the risk factors of the event based on case studies where the collapses of underground cavities took place. The database was then established, taking into account the risk factors, to come up with a statistical model that estimates the risk level. In this study, the maximum likelihood estimation (MLE) method was employed to estimate the parameters in a statistical model. Thorough the statistical analysis, the probability of underground cavity occurrences was found to be expressed in terms of the depth of alluvial layer, groundwater level, water and sewage utilities, and their age. Consequently, an attempt was made to generate a preliminary hazard map for a specific railway route by employing the statistical model.

All large railway networks use a mixture of outdated, modern and emerging signalling and train operation principles. There is a need to develop novel modelling and verification mechanisms to support mixed traffic scenarios, including, for example, mixing different types of signalling and driving. In our previous work we introduced the Unified Train Driving Policy (UTDP) formal modelling language for uniformly capturing diverse signalling principles and mixing, in a demonstratively safe manner, at the node and/or network-level novel and legacy signalling principles. This paper describes our work on making UTDP practical and useful for the engineers.

In recent times, the transportation industry has generated a number of developments involving new technology in signalling. Important developments have involved the production of light by means of light-emitting diode (LED). Unfortunately, the maximum light output, dominant wavelength, reliability and the lifetime of LEDs are all closely related to the junction temperature. Accordingly, LED performance largely depends on the ambient temperature of its operating environment. Consequently, adequate heat-sinking is required to maintain a long lifetime and compliance. Today again, designing an LED railway signal that is able to meet compliance over wide range of temperatures is problematic. In particular, for yellow aspects compliance is difficult to achieve at extreme ambient temperature because the shift in the colour of yellow LEDs over the specified temperature range can be greater than the magnitude of colour specification. Use of phase change materials (PCMs) can provide practical solution of these problems. PCMs can store and release thermal energy during the process of melting and freezing (changing from one phase to another). When such a material freezes, it releases large amounts of energy in the form of latent heat of fusion or energy of crystallization. Conversely, when the material is melted, an equal amount of energy is absorbed from the environment as it changes from solid to liquid. This paper presents studies that enable the design of railway signals to meet, in a rigorous way, colorimetric compliances on a large range of temperature change. Furthermore, the proposed system is applicable, in a simple way and without operational changes, to the existing systems. It is sufficient to insert inside of the signals a suitable amount of a specific PCM.

The railway industry has taken a great effort and is currently focused on exploitation of global navigation satellite system (GNSS) for the European train control system (ETCS). It has been assessed that replacement of track balises, used for safe train location determination, with virtual balises (VBs) detected by means of GNSS will significantly reduce the track-side infrastructure and operational costs. However, this innovated ETCS can be put into operations only in the case when detection of VBs by means of GNSS will achieve the same safety integrity level (SIL 4) and availability as it is required for physical balise groups (BGs).

This paper describes a novel travelling virtual balise (TVB) concept, which was proposed to meet the demanding ETCS safety requirements for GNSS using the existing European Geostationary Navigation Overlay Service (EGNOS) safety-of-life (SoL) service. The TVB concept profits from the basic feature of GNSS – i.e. the ability of abundant train position determination in GNSS service volume, which cannot be realized by current track balise groups (BGs) with a spacing of hundreds of metres or more. The frequent GNSS train positions are utilized for (1) fast diagnostics of on-board location determination system (LDS), (2) introduction of reactive fail-safety into LDS and (3) derivation and justification of the ETCS safety requirements for EGNOS.

The TVB concept brings one significant advantage to ETCS in contrast to the static VBs – i.e. the safety requirement for LDS doesn’t depend longer on the distance between successive VBs. It means that the existing spacing between physical BGs (up to 2.5 km) can be also preserved in case of TVBs. It can significantly improve the availability of LDS. Further it was found that a less demanding tolerable hazard rate (THR) requirement for GNSS of about 1e-7/1 h or more still enables to meet the ETCS THR requirement for VB determination, i.e. THR_VB of 0.67e-9/1 h. Thus the ETCS TVB concept opens the door for efficient use of the EGNOS SoL service, originally developed for aviation.

The planning of a rail system requires the definition of travel demand in terms of passenger (or freight) flows for sizing physical and technological elements (such as number of trains, signalling system type, length and width of platforms). Moreover, once a system has been set up and functional elements have been acquired, system management in terms of services and related timetables requires knowledge of travel demand flows. Much has been written about the methods and techniques for estimating travel demand by means of analytical models (calibrated by surveys), statistical processing of survey data and/ or correcting model results by using properly collected traffic counts. However, whatever the adopted approach, it is necessary to proceed with survey campaigns to acquire experimental data. Obviously, the greater the number of detected data (and related acquisition costs and times), the greater the accuracy of travel demand estimations. Hence, in real cases, a fair compromise between survey costs and estimation accuracy has to be struck.

In this context, we propose an analytical methodology for identifying space–time relations between passenger counts to reduce the amount of data to be surveyed without affecting estimation accuracy. In particular, our proposal is based on defining analytical functions to provide boarding and alighting flows depending on the station (space component) and the time period (time component) in question. Finally, in order to show the feasibility of the proposed methodology and related improvements with respect to traditional approaches, we applied our proposal to the case of a real metro line in Naples (Italy) by comparing different levels of detail in passenger surveys.

new railway line for commuter trains through central Stockholm, the ‘City line’, is under construction and will open in 2017. The line will have two new underground stations and these will be equipped with platform screen doors in order to enhance air quality and prevent suicides. Several stations along the line will have only two platform tracks and this will limit capacity.

Neither trains nor platform doors will be fully automatized and thus the drivers will have to stop the trains in exact position to match the platform doors. Consequently, adding extended times for opening and closing doors will prolong the dwell time by approximately 12 s/stop.

an extensive capacity evaluation has been carried out as a RailSys simulation in order to find out whether a capacity of 24 trains/hour/ direction can be reached. The core part of the study was to prepare dwell time distributions for all stations along the line. These distributions include several terms, namely additional braking time due to a more careful and precise braking, time to correct stopping position (stochastic), door opening time, passenger exchange time (stochastic), door closing time, door obstruction time (stochastic) and driver’s reaction time before departure.

The simulation results indicate that the extended dwell times caused by the platform screen doors limit capacity to 22 trains/hour in one direction and 24 trains/hour in the other direction. This difference comes from the fact that the station with the longest dwell times is located adjacent to a station with four platform tracks where trains can recover delays. Thus, the delay level differs significantly between the two traffic directions. This delay difference influences the useful capacity in such a way that the most punctual direction has a higher capacity.