See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/286413084

Evaluating Track Geometrical Quality through Different Methodologies

Article in International Journal of Technology · July 2010

DOI: 10.14716/ijtech.v1i1.35

CITATIONS READS

74 4,645

4 authors:

Abdur Rohim Boy Berawi Raimundo Delgado

25 PUBLICATIONS 379 CITATIONS

University of Porto
156 PUBLICATIONS 1,962 CITATIONS
SEE PROFILE
SEE PROFILE

Rui Calçada Cecília Vale

University of Porto University of Porto
431 PUBLICATIONS 8,297 CITATIONS 54 PUBLICATIONS 641 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Cecília Vale on 09 December 2015.

The user has requested enhancement of the downloaded file.

International Journal of Technology (2010) 1: 38‐47
ISSN 2086‐9614 © IJTech 2010

EVALUATING TRACK GEOMETRICAL QUALITY THROUGH DIFFERENT

METHODOLOGIES

Abdur Rohim Boy Berawi1*, Raimundo Delgado2, Rui Calçada2, Cecilia Vale2
1
Transportation System, MIT - Portugal Program, University of Porto,
Rua Dr. Roberto Frias, s/n 4200-465 Porto, Portugal
2
Department of Civil Engineering, Faculty of Engineering, University of Porto,
Rua Dr. Roberto Frias, s/n 4200-465 Porto, Portugal

(Received: June 2010 / Revised: June 2010 / Accepted: July 2010)

ABSTRACT
The implementation of High Speed Railway (HSR) networks involves a large amount of
financial support imposing, not only at the conception and design level, but also during the line
operation, a demanding, a complete, and a rigorous estimation of the total cost involved in the
life cycle of the system. By using appropriate tools for estimating HSR life cycle costs (LCC), it
is possible to minimize the final cost and, at the same time, to identify the most important
aspects and parameters influencing the cost evaluation. Research, therefore, is not only required
on the LCC modeling, but also on the estimation of major degradation factors and in the
assessment of its impact on the maintenance needs. This paper deals with this former aspect.
The various methodologies for evaluating the geometrical track quality are presented and
compared to each other, namely the J Synthetic Coefficient, the Indian TGI and also the
approach presented in the European Standard EN 13848-5. In order to compare these three
methodologies, they are applied to a railway stretch of the Portuguese Northern Railway Line.
By doing so, the prediction of track degradation rate within the period of research can be
determined, which possibly is used in the future for defining cost-effective maintenance
strategies.

Keywords: Maintenance optimization; Regression analysis; Track degradation model; Track

geometry; Track quality index

1. INTRODUCTION
In recent years, studies on railway track degradation have attracted a great deal of attention.
Intensive research activities have been conducted by many organizations targeting not only to
secure a high level of safety and reliability of infrastructure systems, but also to diminish the
problems associated with the degradation of performance in terms of ride quality, comfort, etc.
For this reason, many railway Infrastructure Managers (IMs) spend a substantial proportion of
their budget on the Maintenance and Renewal (M&R), which makes up a considerable part of
total railway operating cost and accounts for up to 70% from total life cycle cost of track
infrastructure (Jianmin, 2007). With this huge amount of financial expenditure, undoubtedly, a
small reduction in the maintenance cost will bring a significant impact, particularly on the
overall life cycle cost.

*
Corresponding author’s email: boy_berawi@yahoo.com, Tel. +351‐225082178
Berawi et al. 39

Several approaches and methodologies to evaluate track degradation for track maintenance
optimization have, therefore, been developed during the last few years, from simple models that
are just concentrated on one individual track component to the most comprehensive ones which
embrace all major factors in the track degradation. According to the available literature, these
predictive models may be considered based on two aspects (Sadeghi & Asgarinejad, 2007):
 Track Degradation considered from structural aspect
 Track Degradation considered from geometrical aspect.
In the first aspect, track degradation model is based on the growth of physical structure
conditions. Parameters influencing track degradation, including passage tonnage, train speed,
ballast characteristics, rail types, etc., are investigated and the correlations among them are
analyzed to derive a general equation that quantifies the rate of degradation. The good reviews
of some of these are given in (Sato, 1995; Zhang et al., 1999). Conversely, track degradation
models in the second aspect use geometrical parameters as the main degradation criteria. In
order to measure the track conditions by using this model, typically the track is divided into
several shorter sections and geometry statistics are performed to each of them. The geometry
statistics are then summed up to give a measure of overall segment quality, which is commonly
called Track Quality Indices (TQIs). Use of TQIs provides the possibility to assess railway
track performance indicators, to design interventions, and to compare track performances before
and after the interventions. (Fortunato et al., 2007).
The present research aims to improve the understanding on the mechanism of degradation, their
likelihood to occur in the railway track, and their evolution over the entire lifetime. For these
interests, the Track Quality Indices (TQIs) has, therefore, been chosen in the analysis. All the
aspects related to TQIs, starting from their reflection in the assessment of railway quality, the
role of each geometrical variable to form the index value as well as their implementation, will
be discussed.
The specimen used in this paper comes from the 8 years collection of historical data of one rail
track stretch in Portugal, subjected to mixed traffic. The maximum train speed in this rail track
in study is 220 km/h. Although the rail track in study presents approximately 1 km long, this
paper is intended to show moreover how the method of TQI’s is put into practice in the quality
measurements of the railway. With this approach, we are able to predict the likely rate of track
degradation within the period of research, which may be used in the future for defining
maintenance models.

2. GEOMETRICAL TRACK DEGRADATION CONCEPT

Generally, the major contributors to the track degradation can be categorized in three different
areas, i.e. the initial value of Track Quality Indices (TQIs), traffic, and maintenance (IMRT,
2005). Track Quality Index is defined as a numerical value that represents the relative
condition of the track surface geometries (El-Sibaie & Zhang, 2004). Sadeghi and Askarinejad
(2008) distinguished TQI into two dependent variables. The first variable is the Track
Geometry Index (TGI), which is defined as a function of one or more of the main geometry
parameters such as profile, alignment, gauge, cant, and twist (Figure 1). The specification for
each parameter is detailed as follows. Profile and alignment are delineated with the track
geometry of each rail projected longitudinally against the vertical and horizontal plane,
respectively. Any changes in the elevation of the two rails relative to a designated level is
called profile deviation, while the lateral variation of the rails to a given centerline of the track
contribute to alignment irregularities. Gauge specifies the inner distance between two rails
measured at 16 mm below the top surface of the railhead. In Portugal, the gauge for primary
Railway Lines is 1,668 mm although a wide variety of gauges are used around the world. The
40 Evaluating Track Geometrical Quality through Different Methodologies

term of gauge irregularities, therefore, will refer to the deviation of the track from this
specified value. Cant irregularities measure the amount of vertical deviation between two flat
rails from their designed value. This designed value, commonly known as super-elevation,
helps to compensate the centrifugal force of the vehicle on a given curve. Consequently, cant is
not considered as defect unless it deviates from the predetermined super-elevation. The last
parameter, twist, is also associated with super-elevation. It measures the difference in the
super-elevation between two points taken at a separate fixed distance along the track.

Figure 1 Track Geometry Parameters (Bing & Gross, 1983)

Still according to Sadeghi and Akbari (2008), the second variable of TQI is defined as the Track
Structure Index (TSI), which expresses the condition of the track structure, including the
condition of rail, sleeper, ballast and drainage systems. In this paper, only the TGI will be
analyzed.
Traffic is another major parameter influencing track geometry condition. Ferreira and Murray
(1997) divide traffic related deterioration factors into three groups; dynamic effects, speeds, and
loads. The dynamic effects vary with the type of vehicle on the track, from heavy haul freight
traffic to passenger trains and from fast passenger unit to lower speed mixed traffic. As a result,
the track bed is subject to a wide range of bearing and bending stresses that may come not only
because of the static mass of vehicles, their wheel-sets and their cargo, but also from the
dynamic actions such as lateral forces in curves, acceleration, vibration and imperfection on the
rail surfaces. As the speed increases, the dynamic forces will influence the deterioration of track
geometry significantly and lessen at low speed.
The last parameter, maintenance action, which consist of activities of tamping, grinding, ballast
cleaning, lubrication, replacement etc., is also affecting the ratio of track degradation. For
instance, when the tamping action is performed, the ballast under the ties is re-compacted to
provide the proper load bearing. The ties thus distribute the weight of the rail and rolling stock
and keep the track properly aligned, that in turn, impede the acceleration of rail degradation
rate.

3. TRACK QUALITY INDICES (TQI’S)

As discussed earlier, in order to evaluate the condition of a railway track, an appropriate
assessment technique namely Track Quality Indices (TQI’s) is used. Some methods by which
track quality indices can be obtained are discussed below.

3.1. J Synthetic Coefficient

J synthetic coefficient is used as an indicator of the track quality based on the standard
deviation evolved by Polish Railways (Madejski & Grabczyk, 2002). Four track geometry
parameters are considered in this index: vertical irregularities, horizontal irregularities, twist,
and gauge. The equation for calculating J synthetic coefficient is:
Berawi et al. 41

S z  S y  S w  0.5* Se
J (1)
3.5

where Sz, Sy, Sw and Se are the standard deviation of vertical irregularities, horizontal
irregularities, twist, and gauge, respectively. The standard deviation for each measured
parameter is calculated by the following equation:

1 n
S 
n i 1
( xi 1  x) 2 (2)

Based on the above equation, n is identified as the number of signals registered on the track
being analyzed, xi represents the value of geometry parameters at point i and x is the average
value of the measured signals. The J synthetic track quality coefficient also specifies the
allowable deviation of J, determining the track condition with respect to the state defined by the
track operating appropriately on one side and the track requiring maintenance on the other.

Table 1 Allowable deviations of J coefficient

based on line speed (Madejski & Grabczyk, 2002)
Speed J Coeff. Speed J Coeff.
[km/h] [mm] [km/h] [mm]
80 7.0 150 2.3
90 6.2 160 2.0
100 5.5 170 1.7
110 4.9 180 1.6
120 4.0 190 1.5
130 3.5 200 1.4
140 2.8 220*) 1.1
*)
Calculated through extrapolation

3.2. Track Geometry Index (TGI)

Indian Railways has developed a formula to represent the quality of track called TGI. This
model is based on the standard deviation of different geometry parameters over a stretch of 200
m segment. TGI is calculated for each segment and the average value of such segments in every
km gives the general TGI value (Talukdar et al., 2006). With respect to the effect of each
geometry parameter on the ride quality, TGI has given different value for various geometry
parameters as shown in the following formula:
2UI  TI  GI  6 AI
TGI  (3)
10
where UI, TI, GI, and AI are the index for unevenness, twist, gauge, and alignment
respectively. For each measured track parameters, the index is calculated from the relation:
SDmes SDn
( )
GI , TI , AI ,UI  100 x e SDma int  SDn (4)

where SDme s is the standard deviation of measured geometry parameters, SDn represents the
standard deviation prescribed for newly laid track and SDmaint is the prescribed standard
deviation for maintenance. The standard deviation values used in Equation 4 are specified in
42 Evaluating Track Geometrical Quality through Different Methodologies

Table 2. For the classification of track condition according to the required maintenance is given
in Table 3.
Table 2 Standard deviation (SD) values (Sadeghi& Asgarinejad, 2008)
SD for SD for
SD for
Chord maintenance with maintenance with
Parameters newly laid
Length max. speed ≥ 105 max. speed < 105
track
km/h km/h
Unevenness 9.60 2.50 6.2 7.2
Twist 3.60 1.75 3.8 4.2
Gauge 1.00 1.00 3.6 3.6
Alignment 7.20 1.50 3.0 3.0

Table 3 TGI Classification for maintenance (Talukdar et al., 2006)

No TGI Value Maintenance requirement
1 TGI > 80 No maintenance required
2 50 < TGI < 80 Need basic maintenance
3 36 < TGI < 50 Planned Maintenance
4 TGI < 36 Urgent Maintenance

3.3. European Standard EN 13848-5

The rail track geometry on the sample segment has also been evaluated in accordance with
European Standard EN 13848-5 (CEN, 2005). With respect to the Standard, three track
geometry parameters are considered: longitudinal level, alignment, and gauge. The
measurement of longitudinal level and alignment are conducted based on the standard deviation
of irregularities on a 200 m long segment, while the irregularities on the Gauge is measured
based on a mean value of 100 m long segment. Apart from that, specification of geometry
irregularities with wavelength domain in the range of 3 m < λ ≤ 25 m is another required
parameters to be calculated in the standard deviation.
The allowable thresholds for geometry parameters based on the European Standards are given
in Tables 4-5.

Table 4 SD Threshold values for profile and alignment (CEN, 2005)

Wavelength domain [mm]
Speed
Profile Alignment
80 < V ≤ 120 1.8 – 2.7 1.2 – 1.5
120 < V ≤160 1.4 - 2.4 1.0 – 1.3
160< V ≤ 220 1.2 – 1.9 0.8 – 1.1
220 < V ≤ 300 1.0 – 1.5 0.7 – 1.0

Table 5 Distance limit between specified gauge and mean over 100 m segment (CEN, 2005)
Difference between specified gauge and mean gauge over 100 m segment (mm)
Speed
Safety Limit (SL) Intervention Limit (IL) Alert Limit (AL)
[km/h]
Min Max Min Max Min Max
80< V ≤ 120 -7 +27 -6 +25 -5 +22
120< V ≤160 -5 +20 -4 +18 -3 +16
160< V ≤ 220 -5 +20 -4 +18 -3 +16
220< V ≤ 300 -5 +20 -4 +18 -3 +16
Berawi et al. 43

4. CASE STUDY
For the purpose of this paper, a case study on 1 km straight segment located in Portuguese
Northern Railway Line has been done. The data were obtained from Track Recording Car
(TRC), which provides information about track geometry parameters in two wavelength ranges.
The use of this vehicle makes it possible to record any variations on the track in every 0.25 m,
while still keeping on running with max. speed of 120 km/h. In order to evaluate the track in
terms of its quality and to compare its behavior, two different time periods were considered,
which are the periods before and after renewal actions.
4.1. Track shift adjustment
In order to synchronize the individual measurement data, the researcher uses the data from two
track geometry measurement surveys; one as a reference, while the other is treated as the
dataset to be shifted. Both data are then plotted in MATLAB and by performing a cross
correlation algorithm at specified intervals, the shifted data is matched to the reference track.
The coefficient correlation is expressed as follows:
N

 ( x  x)( x
t t k  x)
rk  t 1
N (5)
 ( xt  x)2
t 1

where x t is data value at time t, k is the lag, and the overall mean is given by:
N
x
x t (6)
t 1 N

After the synchronization, the start and the end points of each track are identified and grouped
for the analysis.
4.2. Evaluation results
In order to make the analysis comparable between the European Standard EN 13848-5 and the
two universal quality index (J synthetic coefficient and TGI), we have divided the track into
200 m-long segment and we observe the evolution of the track condition. The regression
analysis is imposed on the resulted indices for each period in each data measurement. The
advantage of using this method is that it can show the accuracy of our prediction to determine
value from regression squared (R2), with the magnitude range between 0 and 1 (Sadeghi &
Askarinejad, 2009). The higher R2 means the more significant correlation among the data points
and the more accurate the prediction of degradation rate. Some results of the quality
computation are given in the Figures 3 to 7.
J Synthetic Coefficient Track Geometry Index (TGI)

3.00 v = 140 km/h 200.00

2.50 Increase of 160.00

v = 160 km/h degradation
2.00 Increase of
120.00
J Coeff.

Degradation
TGI

1.50
v = 220 km/h 80.00
1.00
40.00
0.50 140 km/h 160 km/h 220 km/h

0.00 0.00
0 500 1000 1500 2000 2500 0 500 1000 1500 2000 2500
Days
Days
Track Geometry Indices TGI with no Maintenance required
J Synthetic Coefficient Threshold Value TGI for need based maintenance TGI for urgent maintenance

Figure 3 J Synthetic coefficient evolution for Figure 4 Track Geometry Index (TGI)
sample segment of Block 1 evolution for sample segment of Block 1
44 Evaluating Track Geometrical Quality through Different Methodologies

Standard Deviation Profile Standard Deviation Alignment

3.00 1.40

Increase of 1.20
2.50
degradation 1.00
2.00 Increase of
0.80
degradation

SD
SD

1.50 0.60
0.40
1.00
0.20
0.50
0.00
0.00 0 500 1000 1500 2000 2500
0 500 1000 Days 1500 2000 2500 Days

Left Profile D1 Right Profile D1 Threshold Value D1 Left Alignment D1 Right Alignment D1 Threshold Value D1

Figure 5 Profile standard deviation of Block 1 Figure 6 Alignment SD of Block 4

Mean Block 1

19

14
Mean [mm]

-1
0 500 1000 1500 2000 2500

-6
Days

Block 1 Safety Limit Intervention Limit Alert Limit

Figure 7 The evolution of mean value of Block 1

Figure 3 to 7 shows the evolution of track quality computed from November 2001 to January
2009, as in this analysis will be represented starting from day 0 to 2632. For each individual
figure, two periods of time are distinguished; one is the time period before renewal consisting
of 8 inspected measurements (day 0-706), and the other is the time period after renewal
consisting of 13 inspected measurements (day 1205-2632). This renewal strategy is also
followed by the policy to increase the line speed from 140 km/h (ordinary track) to 220 km/h
(high speed track).
The evolution of track geometry index calculated by using J synthetic coefficient and TGI are
shown in Figures 3 and 4 respectively. As demonstrated, the quality measurement on the track
segment is in the range under the threshold value for maintenance action, which indicates that
the track is suitable for train operation. Comparing the two periods of time, the renewal
constructed line has revealed a significant quality improvement, a smooth degradation trend,
and lower degradation rate, than the lines before renewal actions. These differences allow us to
justify the feasibility of the implemented renewal strategy in the effort to obtain the established
objective.
In Figures 5 and 6, the evolution of the standard deviation of the track profile and alignment for
track block of 200 m length are presented respectively. Finally, in Figure 7, the evolution of the
mean value of gauge is also shown. The trend of degradation, particularly in profile and
alignment, were quite similar with what is shown in the previous of TQI, indicating that
irregularities in the geometry parameters are still within the acceptable range of value. On the
other hand, the gauge measurement has shown a slightly different result. In the initial
Berawi et al. 45

measurement, the gauge mean value is outside the service tolerance, thus making the mean
value negative. Please also be advised that the negative in gauge mean value specifies the
narrower distance between two rails than it should be. However, as the time passes by, the size
of the track gauge is increasing and when the renewal action is conducted, the track is achieving
a tremendous improvement in quality.
Furthermore, according to the analysis, the rate at which a railway track has degraded is not the
same between period of time before and after renewal. The replacement of the broken parts and
removal of irregularities in the renewal segments indicate a considerable improvement in
quality and show bigger resistance to the nature of degradation of the track. The computation of
degradation rate of J synthetic coefficient in Tables 6 and 7 are included in the samples which
strengthen this argument.

Table 6 Degradation rate before renewal Table 7 Degradation rate after renewal
(Day 0-706) (Day 1205 – 2632)
Days 0-706 Days 1205-2632
Block Linear ( y = α + βx ) Exponential (y=α.eβx ) Block Linear ( y = α + βx ) Exponential (y=α.eβx )
α β R2 α β R2 α β R2 α β R2
1 1.68 0.0003 0.83 1.68 0.0002 0.82 1 0.27 0.00007 0.79 0.28 0.0002 0.79
2 2.16 0.0005 0.91 2.17 0.0002 0.91 2 0.26 0.00007 0.82 0.28 0.0002 0.82
3 2.01 0.0006 0.80 2.02 0.0003 0.79 3 0.11 0.0002 0.92 0.21 0.0004 0.89
4 1.60 0.0005 0.99 1.61 0.0003 0.99 4 0.28 0.00008 0.80 0.31 0.0002 0.79
5 1.73 0.0007 0.99 1.74 0.0003 0.99 5 0.26 0.0001 0.88 0.29 0.0002 0.86
Av. = 0.00052 Av=0.905 Av.= 0.00026 Av.=0.9 Av.= .000104 Av.=0.84 Av. = 0.00024 Av. =0.83

Tables 6 and 7 present the average values of degradation rate of J synthetic coefficient obtained
for every 200 m long segment. The results show the rapid development of track degradation in
the segments before renewal actions (av. β= 0.052 mm/100 days with linear regression), while
in the renewal constructed line, the track indicates its resistance towards to the degradation
process (av. β= 0.0104 mm/100 days with linear regression). Regarding the accuracy of
prediction, the values of regression squared (R2) for both intervals are more than 0.80. Although
there is no absolute standard for what is a “good” R2 value, the application of the regression
square may optionally be used to evaluate the quality of prediction of track degradation. As the
same logic applies, the rate of degradation is also computed for other methods of TQIs, which is
summarized in Table 8.
It can be seen in the above table that the TGI degradation rate value is significantly higher than
those obtained with other methods. However, please be reminded that this does not mean that
the track quality calculated with TGI is much slower to arrive at a degraded state, since the way
of measurements and the range of upper and lower values of the quality index are different for
each one of the methods.
Furthermore, as a comparison for the three geometry parameters measured using European
standards, gauge is demonstrated as having the highest variables of geometrical defect with
degradation rate before renewal (β) approximately 0.319 and after renewal (β) approximately
0.084. This rate is considerably higher than the value of profile (before renewal, β = 0.102; after
renewal, β=0.015), and alignment (before renewal, β = 0.024; after renewal, β=0.0083).
However, it might be interesting to know that globally, the percentage of quality index value of
Gauge from the overall index value is sufficiently small (J synthetic coeff.: gauge (14,3%),
profile (28,3%), alignment (28,5%); TGI: gauge (10%), profile (20%), alignment (60%))
(Talukdar et al., 2006).
46 Evaluating Track Geometrical Quality through Different Methodologies

Table 8 Summarize of degradation rate

Track before Renewal Track after Renewal
Type of Measurements Linear Exponential Linear Exponential
β [mm/ 100 β [mm/ 100 β [mm/ 100 β [mm/ 100
days] days] days] days]
J synthetic Coefficient 0.052 0.026 0.0104 0.0240
Track Geometry Index
1.560 0.016 0.9700 0.0054
(TGI)
Profile 0.102 0.500 0.0150 0.0580
Alignment 0.024 0.025 0.0083 0.0330
Gauge 0.319 n.a. 0.0840 0.0250

5. CONCLUSION
In this paper, several methods to evaluate the track conditions have been introduced and the
applications of each method in the assessment of railway quality were studied. From the results
obtained in this research, the following conclusions are drawn.
The analyses of results in two different periods have indicated a better quality levels obtained
by the track after renewal. This improvement of performance indicators, thus, can be used as a
parameter to assess the appropriateness of the implemented works to obtain the established
objective of renewal strategy.
As discussed before, the rate of degradation of the track segment will be different according to
the use of various measurement methods. TGI, for instance, has been accounted as the highest
degradation rates with β = 1.56 mm/100days. However, the result was not surprising since the
range of interval quality between upper and lower value in TGI is wider than others. The use of
TGI also has imposed the necessity to adjust the tolerance value for maintenance in the case for
high speed implementation. Concerning with European Standard EN 13848-5, the three
geometrical parameters have shown different rate values leading to degradation as well as the
amount of time needed to reach the critical conditions for maintenance. Since the maintenance
will be perceived ineffective if it is just solely based on the irregularity of one particular
parameter, therefore, it might be necessary to have more comprehensive analysis by combining
all the different parameters to construct a uniform index that could facilitate the assessment of
the overall fitness of the track segments. The applications of TGI and J synthetic have allowed
this objective by using various geometry indicators in the track evaluation with different
weighted values for each of them.

6. ACKNOWLEDGEMENTS
The authors also wish to acknowledge the support of the project “HSR-LIFE Development of
tools for HSR lifecycle costs estimation for track design and maintenance management system”
of the MIT-Portugal Program - Transportation Systems Area.
 Berawi et al. 47

7. REFERENCES
Bing, A. J. & Gross, A., 1983. Development of railway track degradation models.
Transportation Research Record 939, pp.27-31.
CEN, 2005. Railway applications -Track- Track geometry quality - Part 5: Geometric quality
assessment; German version prEN 13848-5:2005
El-Sibaie, M. & Zhang, Y., 2004. Objective Track Quality Indices. Proceedings of 83rd TRB
Annual Meeting, Washington, D.C.
Ferreira, L. & Murray, M., 1997. Modelling rail track deterioration and maintenance current
practices and future needs. Transport Reviews, 17, pp.207-221.
Fortunato, E., Pinelo, A., Lobo Costa, J., Gonçalves, D. & Pratas, A., 2007. Some results of
performance indicators of a railway track obtained during its renewal process. 4th
International SIIV Congress-Palermo, Italy.
Iran Ministry of Roads and Transportation (IMRT), 2005. Railway track superstructure general
technical specification Standard no. 301. Iran: IMRT Ministry Publication Service.
Jianmin, Z., 2007. Modelling and optimizing track maintenance and renewal. Phd Thesis, Civil
Engineering Department, University of Birmingham.
Madejski, J. & Grabczyk, J., 2002. Continuous geometry measurement for diagnostics of tracks
and switches. Proceedings of the International Conference on Switches Delft University of
Technology, Delft, The Netherlands.
Sadeghi, J. & Askarinejad, H., 2007. Influences of track structure, geometry and traffic
parameters on railway deterioration. International Journal of Engineering, Volume 20, No.
3.
Sadeghi, J. & Askarinejad, H., 2008. Development of Improved Railway Track Degradation
Models. International Journal of Structure and infrastructure Engineering, Vol.3, Issue 4.
Sadeghi, J. & Askarinejad, H., 2009. Investigation on effect of track structural conditions on
railway track geometry deviations. Proceeding of the institution of mechanical engineers,
Part F, Journal of rail and rapid transit.
Sato, Y., 1995. Japanese studies on deterioration of ballasted track. Vehicle System Dynamic,
vol.24, pp.197-208.
Talukdar, K., Arulmozhi, U., Prabhakar, K. & Satyanarayan, 2006. Project Presentation
Improvement of TGI Value by Computer Analysis. Ministry of Railways Indian Railways
Institute of Civil Engineering, India.
Zhang, J., Murray, H. & Ferreira, L., 1999. An Integrated Model for Track Degradation
Prediction. World Transport Research: Selected Proceedings of the 8th World Conference
on Transport Research, vol.3, pp.527-539.

View publication stats