Climate Change Adaptation for GeoRisks Mitigation of Railway Turnout Systems

Climate Change Adaptation for GeoRisks Mitigation of Railway Turnout Systems

Available online at ScienceDirect Procedia Engineering 189 (2017) 199 – 206 Transportation Geotechnics and Geoecology, TGG 201...

862KB Sizes 0 Downloads 58 Views

Available online at

ScienceDirect Procedia Engineering 189 (2017) 199 – 206

Transportation Geotechnics and Geoecology, TGG 2017, 17-19 May 2017, Saint Petersburg, Russia

Climate Change Adaptation for GeoRisks Mitigation of Railway Turnout Systems Serdar Dindara, Sakdirat Kaewunruena*, Joseph M. Sussmanb b

a School of Civil Engineering,The University of Birmingham, Birmingham, UK Department of Civil and Environmental Engineering, Massachusetts Institute of Technology, Cambridge, USA

Abstract To enhance rail operational flexibility, railway turnouts are special track systems, which are designed to divert or change a train from a particular direction or a particular track onto other directions or other tracks. In reality, the railway turnout is commonly built on complex track geometry and graded terrain, which makes it one of the most unique and critical railway infrastructures. The physical constraints and complexity of turnout systems cause various risks and uncertainty in rail operations. This study critically analyses emerging geotechnical risks on turnout systems considering all aspects that can potentially result in impaired reliability, availability, maintainability and safety (RAMS) of the turnout systems. The annual derailment incidents have been evaluated to identify emerging risk factors. Not only do these incidents yield operational downtime and financial losses, but they also give rise to the casualties and sometimes the loss of lives across the world. In particular, the climate change risks on geotechnical aspects of the turnout systems have been highlighted. This paper thus presents how turnout components work as a system, the diversity of emerging risks considering natural hazards and global warming potential to the system. In addition, it highlights the climate change adaptation strategies for georisk mitigation of the railway turnout systems in order to improve RAMS of the railway turnouts and crossings, focusing on trackbed failures on the systems. 2017The TheAuthors. Authors. Published by Elsevier Ltd. is an open access article under the CC BY-NC-ND license ©©2017 Published by Elsevier Ltd. This Peer-review under responsibility of the scientific committee of the International conference on Transportation Geotechnics and ( Geoecology. Peer-review under responsibility of the scientific committee of the International conference on Transportation Geotechnics and Geoecology Keywords: natural hazards; bayesian network; railway turnout; switch and crossing; trackbed failures.

* Corresponding author. Tel.: +44 (0) 121 414 2670; fax: +44 (0) 121 414 3675. E-mail address: [email protected]

1877-7058 © 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license

( Peer-review under responsibility of the scientific committee of the International conference on Transportation Geotechnics and Geoecology



Serdar Dindar et al. / Procedia Engineering 189 (2017) 199 – 206

1. Introduction

As a key asset of railway infrastructure, turnouts represent the junctions in trackwork where lines diverge or converge. These infrastructures are constructed to enable a rolling stock to divert one track to another. A turnout consists of a number of mechanical parts, electric or hydraulic installments, concrete or wooden ties and trackbeds. In order to ensure a smooth turnout operation, all turnouts should be working in harmony in various operational conditions, yielding both complexity and variable risks for railway operators. Turnouts were revealed to be systems vulnerable to the environment; thus, the diversity of emerging risks associated with natural hazards and global warming requires investigation [1]. The environmental impact varies from one part of the system to another [2]. In other words, the question regarding to what extent the vulnerability of a system’s parts can be influenced through environmental impact should be clarified in order to maximise operational productivity and smoothness of railway turnout operation. One of these vulnerable systems is the trackbed. In this paper, trackbed failures of turnouts were observed at various sites wherein different climate patterns are summarised and revealed. To determine the solid relation between environmental patterns and railway turnout trackbed, all derailment cases suspected as being caused by poor trackbed and reported as derailment caused by climate/weather are investigated. The results of investigation—causal relations—are presented through this paper. Thus, this study allows such relations to be modelled and predicted by numerous risk analysis techniques, each of which helps to minimize the risk in a particular part of turnouts [3]. The paper concludes with an analysis of possible scenarios and suggestions for countries with different climate patterns to reveal geo-risks by which challenges can be countered. 2. Railway turnouts

The word "turnout" in this paper refers to all track formations enabling rolling stocks to be guided from one track to another at a railway junction so as to avoid “switches" or "points", both of which terms in literature might sometimes be confusing. There is a large variant of railway turnout design. 2.1. Turnout types


Right hand turnout

Diamond switch

Y (wye) switch

Fig. 1 Common types of turnouts, the most used in railway operation.

3-way turnout

Slip switch


Serdar Dindar et al. / Procedia Engineering 189 (2017) 199 – 206

Apart from the standard left-hand and right-hand turnouts, railway turnouts also commonly exist in various combinations of configurations. According to different usages and structures, such turnouts are named as illustrated in Figure 1. Moreover, all turnouts, whether or not they are named in the same way, consist of three parts: railway switch, connection part, and frog and guard rail. 2.2. Turnout trackbed Similarly, a wide variety of turnout forms and systems incorporate some form of concrete support or base which is necessary to build ballast. Almost all of these are used to construct high-speed lines (HSLs), requiring less depth of construction than ballasted turnouts on regular lines often up to 200km/h. In the UK, British Rail Class 43, aka HST125, with a regular service speed of roughly 201km/h, is accepted as entry level of high-speed trains. Even entry level requires changes in the design of turnouts to achieve smooth operation at higher speeds. Thus, as the design of turnouts on HSLs does not frequently need ballast and demands quite different design, turnouts on HSLs are left to future research.

Slipper Rail


Fig. 2 Trackbed design of a desirable turnout.

As designed in the other parts of a rail line, the trackbed is constructed through almost the same engineering techniques. Figure 2 represents an advanced trackbed design formed of three layers: Ballast, Blanket and Subgrade. Turnout Ballast, packed between, below, and around the sleepers by specialised equipment, is frequently made from crushed fines-free granite. This layer allows for facilitating drainage of water, distribution of live loads on rails and suppressing vegetation on or around turnouts. As for Blanket, its role is to reduce traffic-induced stress and distribute it to the subgrade. This layer is mostly a graded-sand layer used to prevent the upward movement of fine sub-grade particles. The graded-sand layer might be covered by geotextiles such as kinds of membranes and/or filter fabrics, which makes the layer durable to the abrasion and point loading of ballast. The undermost layer is subgrade with numerous conditions in accordance with the geometry and drainage of the turnout structure which influences the performance of the turnout subgrade. 3. Failures Mechanisms The quality of the turnout trackbed is quite important to meet the requirements of the modern operational standards, as turnouts are so often used in a railway operation. For instance, Network Rail, the infrastructure manager of most of the rail in the UK, has 21,000 track miles and 19,000 turnouts [4] . In other words, there is one turnout per 1,14 track mile. In addition, considering the increasing demand for any kind of railway transportation through the world, turnouts will need more attention to mitigate the arising risk of failure, especially under traffic levels and increased speed.


Serdar Dindar et al. / Procedia Engineering 189 (2017) 199 – 206

Although a desirable turnout is illustrated in pervious section, the operator of lines encounters turnouts in various conditions. In the USA, there are nine classes which cover freight and high- and regular-speed passenger trains. Such numbering is developed by Federal Railroad Administration (FRA) to classify track quality. Therefore, turnouts with different designs than those illustrated in Figure 2 frequently appear in developed countries. Moreover, railway infrastructure operators often do not use geotextiles, due to economic concerns, which makes the system more vulnerable to environmental conditions [5]. Turnouts are also rarely well protected against water ingress because of a lack of drainage systems or poorly maintained drainage. As they consist of a permeable ballast surface, seasonal deformation behaviour is highly likely to present on railway turnouts [6]. In addition, where track drainage is somewhat impeded and, as a result, exacerbation of trackbed problems as well as concentrations of water are ongoing, track problems and progressive deformation of the subgrade might become worse [7]. Table 1 Environmental-based Failure Mechanism Resulting in Derailment on Turnouts. Failure type

Environmental reason


Progressive shear failure

High water content

Squeezing near subgrade surface

Excessive plastic deformation

Repeated freezing and thawing

Ballast pocket

Attrition with mud pumping

High water contact at subgrade surface

Poor drainage

Depression under ties

Muddy ballast Inadequate sub-ballast Frost action

Low temperature

Often occurs winter/spring season

Frost susceptible soil Swelling/Shrinkage

Changing moisture content

Rough track surface


A heavy downpour of rain

Turnout geometry problems

Slope erosion

High wind

Poor designed trackbed

Subsurface water

Soil washed or blown away

Table 1 includes accident reports from the Federal Railroad Administration (FRA) along from publications and reports in the literature. The table also illustrates common reported failure types associated with their environmental reasons and features. However, the table is prepared only for derailment accidents at turnouts. It might be expected to identify more or less different failures for the other kind of accident such as collision. The relation between environmental conditions and failure types is also revealed in the table. For instance, when high water content exists in trackbed layers for some reason, it is highly likely that this contributes to irregularities in track geometry, resulting in progressive shear failure, attrition, washout etc. The table, thus, provides information including causal relation, failure features, and relevant environmental reasons. Figure 3 shows a few frequently encountered case studies of environmental-based failure mechanisms in railway turnout operation. As trackbed accidents by extreme environmental conditions can be said to be rare events at railway turnout systems, pictures (a) and (b) in the figure are derived from similar accidents occurred on the main line. However, symptoms through such cases might be expected to be akin to each other even if the impact area of these failures on turnout design is larger than that of main line. Besides, these impacts might be observed to vary from a type of turnout to another. As already seen the figure 1 and discussed in previous section, the design of trackbed is likely to be different in accordance with its purpose, operator choices as well as traffic on it. On the other hand, considering only derailment cases on which this study concentrates, the impacts should be expected so destructive that a turnout, whether or not it is of a specific design, might as well be failure to function properly. As a result of these deductions, it can be expressed that much risk is taken geotechnically and geographically on by turnouts.


Serdar Dindar et al. / Procedia Engineering 189 (2017) 199 – 206






Fig. 3. (a) Washout; (b) Progressive shear failure; (c) Attrition with mud pumping; (d) Frost action; (e) Excessive plastic deformation

As for illustrations in the figure 3, a washout (a) occurs due to a heavy downpour of rain or other sudden flooding. This kind of failure type might destroy a turnout`s right-of-way, leaving it suspended in mid-air across the formed gap as seen in the Fig.3 (a). Progressive shear failure (b) develops at the subgrade surface of a turnout as the soil is remoulded and sheared due to repeated loading by rolling stock. The surface of the subgrade subjected to this type of failure gradually squeezes upward and outward following the path of least resistance as seen in the yellowish soil on left side on Figure 3 (b). This failure type is the most common problem in the railway industry [8]. Attrition with mud pumping (c), also known as the migration of fines, leads to the ballast of a turnout becoming contaminated, and, as a result, decreasing its ability to carry live loads, eventually causing turnout geometry problems often along with the formation of wet spots.This problem is widespread in UK rail lines; for instance, Ghataora et al. [9] , revealed roughly three wet spots per kilometre of track. Frost action (d), also known as heave and softening, is rarely seen in winter/spring seasons and is caused by periodic freezing. This failure gives rise to rough tracks as seen in Fig.3. (d). Excessive plastic deformation (e) consists of a vertical component of progressive shear deformation and the vertical deformation caused by the consolidation of subgrade soils and progressive compaction under cyclic over-stressing by traffic loads. This plastic deformation produced at railway turnouts might be essentially negligible in a way that water is trapped between slippers under normal conditions (see Fig.3 [e]).


Serdar Dindar et al. / Procedia Engineering 189 (2017) 199 – 206

4. Accident analysis This research uses quantitative and qualitative methods to examine geographic and geological risks incurred at turnouts in accordance with the nature`s role in derailment. The qualitative approach is given in Sec. 2 and 3, considering and deducing the investigation of main line based accidents. This section presents quantitative approach for better understanding to what degree a turnout is susceptible to the nature and possible consequences. This research uses quantitative and qualitative methods to examine geographic and geological risks incurred at turnouts in accordance with nature`s role in derailment. The qualitative approach is given in Sec. 2 and 3, considering the investigation of main line-based accidents. This section presents the quantitative approach for a better understanding of what degree a turnout is susceptible to nature and possible consequences. FRA railway accidents and incidents over the last 10 years from 2005 to 2015 have been chosen as this year base covers the most recent period for which a complete set of derailment accident data are available. The overall aim of this analysis is to determine the number, type, damage cost (consequences) and characteristics of reportable turnoutrelated accidents where nature is involved. An additional purpose of the analysis is to identify the utility of such a database in determining the role of nature, which is challenging for reporters. Complete raw accident reports include train derailments on all track types (yard, main, etc.) and caused by all type of failures and errors such as human or operational. This approach has resulted in a total of almost 18,000 unique cases. After sorting out all derailments wherein nature is involved, the researcher reduced the number down to 742. Of these 742 accidents, 126 have been identified as occurring on railway turnouts, given the description of the derailment in the report including solid information; otherwise, the report is omitted. Lastly, of 126 cases, 14 are determined to have resulted from various trackbed failures, as illustrated in Table 1. In a nutshell, a review of each of the 14 narratives in almost 18,000 cases revealed any kind of trackbed failures which contributed to a derailment at turnouts due some environmental reason. In other words, 14 out of 187 trackbed failures caused by harsh environment occurred at turnouts (less than one-tenth of the all). 40.0 Progressive shear failure

35.0 Attrition with mud pumping



Temperature (C0)

25.0 20.0



Progressive shear failure

Progressive shear failure


Progressive shear failure Washout


Progressive shear failure



Attrition with mud pumping

0.0 0










-5.0 Frost action


Average Precipitation (mm)

Fig. 4. Derailment cases by various trackbed failures.

The cases are illustrated in Figure 4, corresponding to the type of failures presented in Table 1. The temperature on the Y-axes presents the degree at the accident site at the time of the accident and are taken from accident reports, while the precipitation values are the average monthly precipitation of the accident site in the month when the accident occurred. The precipitation values are obtained from the National Weather Service, the official U.S. department concerning meteorology.

Serdar Dindar et al. / Procedia Engineering 189 (2017) 199 – 206

Washout cases accumulate on the right side of the figure. This means that a rainy area signified risk for washout cases. On the other hand, although the number of progressive share failure cases was limited, this type of failure generally occurred within a temperature range of between 50 and 200 C. Lastly, according to the figure, frost action failures are likely not to be a relation between the amount of precipitation and the failure itself. Due to the limited number of cases and the absence of previous studies in this domain, it is not possible to comment on the other type of failures. Although slope erosion is assumed to address the failure contributing to a derailment at turnouts in Table 1, such a case could not be observed in the years between 2005 and 2015 in the USA. Figure 5 shows a bar chart along with a pie chart. The former is drawn to show how much a failure type costs on average, while the latter illustrates the distribution of total property damage costs for all crash types. Therefore, the burden of washout accidents, including all equipment damage along with track, signal, way and structural damage, costs over half a million dollars, as it destroys not only superstructure but also the infrastructure of a turnout. Washout failures also dominate the distribution of total damage costs as illustrated in Fig.5 (b). Therefore, considering the catastrophic financial consequences of washout failures, attention should be paid to this kind of failure in the risk management of a turnout. Although the damage cost of progressive shear failures is reasonably low in comparison with the other three types, it is still ranked second in terms of the total damage cost carried by operators because this kind of failure occurs more frequently.


600 USD Thousands

500 400 300

200 100 0


Fig. 5. a) damage cost per accident, b) damage cost distribution by accident types



Serdar Dindar et al. / Procedia Engineering 189 (2017) 199 – 206

5. Concluding Remark This research achieved the first steps of risk management chain in investigating the role of nature in turnoutrelated derailments at railway turnout systems. Additionally, emerging geotechnical risks on turnout systems is analysed critically emerging geotechnical risks on turnout systems considering all aspects. As this investigation focused on a unique area, it is the first attempt to reveal possible failure mechanisms as a result of such conditions. A number of different mechanisms regarding railway turnout failure have been discussed to comprehend to which degree climate-related softening of a trackbed can induce derailment. Then, the dataset created by the FRA covering all railway accidents and incidents from 2005 to 2015 has been investigated. As a result of this investigation, washout is observed to be likely the most crucial element to consider in risk management as not only does it occur more often than the others, but accidents of this kind also give rise to the considerable damage cost of the burden.

Acknowledgements The authors would also like to thank British Department for Transport (DfT) for Transport - Technology Research Innovations Grant Scheme, Project No. RCS15/0233; and the BRIDGE Grant (provided by University of Birmingham and the University of Illinois at Urbana Champaign). The second author gratefully acknowledges the Japan Society for the Promotion of Science (JSPS) for his JSPS Invitation Research Fellowship (Long-term), Grant No L15701, at Track Dynamics Laboratory, Railway Technical Research Institute and at Concrete Laboratory, the University of Tokyo, Tokyo, Japan. The authors are sincerely grateful to the European Commission for the financial sponsorship of the H2020-RISE Project No. 691135 “RISEN: Rail Infrastructure Systems Engineering Network”, which enables a global research network that tackles the grand challenge of railway infrastructure resilience and advanced sensing in extreme environments ( [10]. References [1] S. Dindar, S. Kaewunruen, M. An and O. Mohd , "Natural Hazard Risks on Railway Turnout Systems," Procedia Engineering, vol. 161, p. 1254–1259, 2016. [2] M. I. Fitri , S. Dindar and S. Kaewunruen, "Safety-based maintenance for geometry restoration of railway turnout systems in various operational environments," Songkhla, THAILAND, Proceedings of The 21st National Convention on Civil Engineering. [3] S. Dindar, S. Kaewunruen and M. An, "Identification of appropriate risk analysis techniques for railway turnout systems," Journal of Risk Research, pp. 1-22, 2016. [4] Halcrow Group Limited , "Independent Reporter A Reporter Mandate - Coal Dust Spillage Costs Final Report," Network Rail, London, 2008. [5] P. Wang, Design of High-Speed Railway Turnouts: Theory and Applications, Academic Press, 2015. [6] A. O'Brien, "Rehabilitation of Urban Railway Embankments-Investigation, Analysis and Stabilisation.," in Proceedings of the 14th International Conference Soil Mechanics and Geotechnical Engineering, Madrid,, 2007. [7] E. T. Selig and J. M. Waters, Track geotechnology and substructure management, London: Thomas Telford, 1994. [8] K. Usman, B. Michael and G. Gurmel , "Railway Track Subgrade Failure Mechanisms Using a Fault Chart Approach," Procedia Engineering, vol. 125, p. 547–555, 2015. [9] K. R. Rushton, G. Ghataora and V. Diyaljee, "Discussion: Design for efficient drainage of railway track foundations," Proceedings of the Institution of Civil Engineers - Transport, vol. 169, no. 2, pp. 118-119, 2016.