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

net/publication/290035876

eMaintenance Solutions for Railway Maintenance Decisions

Article · July 2014

CITATIONS READS

4 647

3 authors:

Ravdeep Kour Ramin Karim

Luleå University of Technology Luleå University of Technology
20 PUBLICATIONS   143 CITATIONS    70 PUBLICATIONS   992 CITATIONS   

SEE PROFILE SEE PROFILE

Phillip Tretten
Luleå University of Technology
43 PUBLICATIONS   320 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Signifying the human cognition in future maintenance systems View project

Optive Volvo Cars View project

All content following this page was uploaded by Phillip Tretten on 14 June 2016.

The user has requested enhancement of the downloaded file.

 Journal of Quality in Maintenance Engineering
eMaintenance solution through online data analysis for railway maintenance decision-
making
Ravdeep Kour Phillip Tretten Ramin Karim
Article information:
To cite this document:
Ravdeep Kour Phillip Tretten Ramin Karim , (2014),"eMaintenance solution through online data analysis for
railway maintenance decision-making", Journal of Quality in Maintenance Engineering, Vol. 20 Iss 3 pp. 262
Downloaded by LULEA UNIVERSITY OF TECHNOLOGY At 06:54 24 May 2016 (PT)

- 275
Permanent link to this document:
http://dx.doi.org/10.1108/JQME-05-2014-0026
Downloaded on: 24 May 2016, At: 06:54 (PT)
References: this document contains references to 26 other documents.
To copy this document: permissions@emeraldinsight.com
The fulltext of this document has been downloaded 465 times since 2014*
Users who downloaded this article also downloaded:
(2014),"Enhancing the usability of maintenance data management systems", Journal of Quality in
Maintenance Engineering, Vol. 20 Iss 3 pp. 290-303 http://dx.doi.org/10.1108/JQME-05-2014-0032
(2014),"Condition monitoring and e-maintenance solution of railway wheels", Journal of Quality in
Maintenance Engineering, Vol. 20 Iss 3 pp. 216-232 http://dx.doi.org/10.1108/JQME-05-2014-0027
(2015),"Performance measurement and management for maintenance: a literature review", Journal of
Quality in Maintenance Engineering, Vol. 21 Iss 1 pp. 2-33 http://dx.doi.org/10.1108/JQME-10-2013-0067

Access to this document was granted through an Emerald subscription provided by emerald-srm:172900 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for
Authors service information about how to choose which publication to write for and submission guidelines
are available for all. Please visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The company
manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well as
providing an extensive range of online products and additional customer resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee
on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive
preservation.

*Related content and download information correct at time of download.

 The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1355-2511.htm

JQME
20,3
eMaintenance solution through
online data analysis for railway
maintenance decision-making
262 Ravdeep Kour, Phillip Tretten and Ramin Karim
Operation and Maintenance Engineering, Luleå University of Technology,
Luleå, Sweden
Downloaded by LULEA UNIVERSITY OF TECHNOLOGY At 06:54 24 May 2016 (PT)

Abstract
Purpose – The purpose of this paper is to demonstrate how research within the railway sector is
developing eMaintenance solutions using the cloud and web-based applications for improved
condition monitoring, better maintenance and increased uptime. This eMaintenance solution is based
on the on-line data acquisition, integration and analysis leading to effective maintenance decision
making.
Design/methodology/approach – In the proposed methodology, data are acquired from railway
measurement stations to the eMaintenance cloud, where they are filtered, fused, integrated and
analysed to assist maintenance decisions. Extensive consultation with stakeholders has resulted in
the analysis of railway data.
Findings – The paper provides a concept for a web-based eMaintenance solution for railway
maintenance stakeholders for making fact-based decisions and develops more efficient and economically
sound maintenance policies. Train wheels reaching their maintenance and safety limits are visualised in
grids and graphs to assist stakeholders in making the appropriate maintenance decisions.
Practical implications – In this paper the authors have demonstrated that the wheel profile and
force data can be remotely collected through cloud utilisation. The information generated can be used
for maintenance decision making. Similarly, other measurable data can also be utilised for maintenance
decision making.
Originality/value – This paper describes the importance of eMaintenance solution through online
data analysis to make effective and efficient railway maintenance decisions, as a case study.
Keywords Maintenance, Online, Decision, Railway, Cloud, eMaintenance
Paper type Research paper

1. Introduction
For effective and efficient maintenance decisions, accurate and timely information
must be available to all stakeholders involved in the maintenance process (Candell
and Karim, 2008). If the maintenance decision-making process is fully automated, cost
effective maintenance decisions will result, facilitating, diagnostic and prognostic
processes. This paper is based on research within the railway sector on developing
eMaintenance solutions using eMaintenance cloud and web-based applications for
improved condition monitoring, improved maintenance and increased uptime. The term
eMaintenance emerged in the early 2000s and has become an emerging area in
maintenance related literature because of ongoing technological developments. Different
definitions of eMaintenance are provided by different authors. For instance, Moore and

Journal of Quality in Maintenance
Engineering
Vol. 20 No. 3, 2014
pp. 262-275
r Emerald Group Publishing Limited
1355-2511
DOI 10.1108/JQME-05-2014-0026
The authors would like to thank Trafikverket (Swedish Transport Administration) for providing
data for this study and JVTC and LKAB/MTAB for sponsoring research work. The authors
would also like to express their sincere appreciation to eMaintenance Lab for achieving this
work. Furthermore, authors would also like to thanks Associate Professor Aditya Parida for his
valuable comments and feedback on drafts of this paper.
 Starr (2006) stated “E-maintenance is an asset information management network that eMaintenance
integrates and synchronises various maintenance and reliability applications to gather solution through
and deliver asset information where it is needed when it is needed”. In addition,
“e-maintenance emerged as a result of the integration of ICT technologies in maintenance online data
policies to deal with new expectations of innovate solutions for e-manufacturing and
e-business” (Muller et al., 2008). Therefore, industries change their manufacturing
operations from local factory integration and automation to global enterprise automation 263
with the ability to exchange information and synchronise with different e-business
systems (Zurawski, 2006). Thus, with the extensive use of internet and sensor technology
different stakeholders can use the integrated platform for effective and efficient
maintenance decisions. This integrated platform is provided by using web services and
Downloaded by LULEA UNIVERSITY OF TECHNOLOGY At 06:54 24 May 2016 (PT)

cloud-based technology. The term cloud computing was first used in a 1997 lecture by
Ramnath Chellappa to mean a “computing paradigm where the boundaries of computing
will be determined by economic rationale rather than technical limits alone” (Sourya,
2011). In 1999, Salesforce.com became the first site to deliver applications and software
over the internet; in 2007 Salesforce.com expanded its efforts, with Force.com (Salesforce,
2012). In 2002, Amazon added “Web Services” (AWS) and introduced the Elastic Compute
cloud (EC2) as a commercial web service in 2006 (Amazon, 2014). Eucalyptus arrived
next in 2008 with the first open source AWS API (application programming interface)
compatible platform for deploying private clouds, followed by OpenNebula, the first
open source software for private and hybrid clouds. Google and Microsoft finally entered
the playing field with Google Apps (application software) in 2008 and Windows Azure in
2009, respectively. Apple joined in with its development of the iCloud allowing users to
synchronise photos, apps, music and documents across a string of devices (Zhang et al.,
2010). Despite the widespread interest, however, cloud computing remains an evolving
paradigm. With cloud computing technology, data can be stored in the cloud, and these
data can be visualised in a virtual machine to meet specific customer and application
requirements (Heuwinkel et al., 2003). Rouse (2011) stated that virtual machine typically
emulates a physical computing environment, but requests for CPU, memory, hard disk,
network and other hardware resources are managed by a virtualisation layer which
translates these requests to the underlying physical hardware. According to Padhy et al.
(2011) virtualisation is a technology that combines or divides computing resources
to present one or many operating environments using methodologies like hardware
and software partitioning or aggregation, partial or complete machine simulation,
time-sharing and many others. Thus, with virtualisation multiple operating system
images can be allowed at once by using only one piece of hardware. Therefore, relevant
information services are provided in a shared view to different stakeholders via cloud
computing using virtualisation. Further, a good eMaintenance solution needs to sense the
context of the individual stakeholders to adapt the information to the stakeholder’s
current situation (Candell et al., 2011). Therefore, the aim of this research is to provide an
eMaintenance solution according to the needs and requirements of different stakeholders
using cloud computing for effective and efficient decision making. Further, it becomes
very difficult to pinpoint possible faults from large and messy data; thus, tools must be
developed to perform advanced data analysis and data mining. To this end, this paper
uses the eMaintenance cloud through online data analysis and the visualisation
of relevant information to assist railway operators in making maintenance-related
decisions. The eMaintenance solution suggested here helps stakeholders to access
information from the cloud anywhere in the world at any time according to their needs
and requirements. After introduction at Section 1, Section 2 deals with maintenance
 JQME information process and cloud computing. Section 3 discusses railway maintenance
20,3 decision making. The case study is discussed in Section 4, followed by results and
discussions and conclusions.

2. Maintenance information process and cloud computing

To support an effective maintenance decision-making process, a maintenance information
264 process is needed as shown in Figure 1. This maintenance information process consists
of data acquisition – to obtain relevant data and manage its content; data transition – to
communicate the collected data; data fusion – to compile data and information from
different sources; data mining – to analyse data to extract information and knowledge;
information extraction and visualisation – to support maintenance decision. All these
Downloaded by LULEA UNIVERSITY OF TECHNOLOGY At 06:54 24 May 2016 (PT)

steps are necessary from data acquisition to information visualisation for maintenance
decision making. In this present work, data are collected from wayside measurement
stations and send to eMaintenance cloud for data analysis and data visualisation.
Today, railways are adopting IT systems for wheel and track maintenance. Railway
organisations should plan for routine maintenance to ensure that trains are operational.
Trains infrequently suffer a complete and unexpected breakdown but if left unfixed for
too long may precipitate a breakdown. The idea of developing railway eMaintenance
solution is to make the maintenance team knew of the issue as soon as the train driver
did, or even before so that decision can be taken and the problem can be fixed.
The present work is based on the approach for establishment of eMaintenance solution
for railway using cloud-based technologies. According to the NIST (2014) cloud
computing is a model for enabling ubiquitous, convenient, on-demand network access
to a shared pool of configurable computing resources (e.g. networks, servers, storage,
applications and services) that can be rapidly provisioned and released with minimal
management effort or service provider interaction. Railway eMaintenance cloud is
implemented where real time data related to trains, axles, wheels, etc. is acquired,
integrated, analysed and visualised for maintenance decisions. Further, the term cloud
computing associates three levels of service: software (Software as a Service or SaaS), the
software platform (Platform as a Service, or PaaS) and the network infrastructure
(Infrastructure as a Service, or IaaS).
. SaaS: “Cloud providers install and operate application software in the cloud and
the cloud user can access the software over internet. This eliminates the need
to install and run the application on the cloud user’s own computers simplifying
maintenance and support. For example, Google Apps for Business provides

Data Data Data Data Information

Acquisition Transition Fusion Analysis Visualisation

Figure 1.
A generic maintenance
information process
Source: Wandt et al. (2012)
 organisations with cloud services for e-mail, calendar, document storage and eMaintenance
collaboration” (Wikipedia, 2014). solution through
. PaaS: “At this level cloud providers deliver a computing platform typically online data
including operating system, programming language execution environment,
database and web server. Application developers can develop and run their
software solutions on a cloud platform without the cost and complexity of buying
and managing the underlying hardware and software layers. For example, 265
Windows Azure AppFabric or Google App (software application) Engine provide
APIs for building customised applications, including cached datastores, URL
fetch, multi-tenant support, component cloning and automated migration for
reducing latency” (Wikipedia, 2014).
Downloaded by LULEA UNIVERSITY OF TECHNOLOGY At 06:54 24 May 2016 (PT)

. IaaS: “The user rents a remote computer. The consumer does not manage or
control the underlying cloud physical infrastructure but has control over
operating systems, storage, deployed applications and possibly limited control
of select networking components. Examples of IaaS include the Amazon Web
Services – e.g. Elastic Compute Cloud (EC2), Simple Storage Service (S3), Elastic
Block Storage (EBS) – and the Windows Azure Fabric” (Wikipedia, 2014).
Consequently, with eMaintenance cloud, proficient and operative decisions are made
for cost effective maintenance planning.

2.1 eMaintenance cloud

eMaintenance is defined at two levels of abstraction: first, “eMaintenance is maintenance
managed and performed via computing”; second, “eMaintenance is a multidisciplinary
domain based on maintenance and ICT ensuring that the eMaintenance services are
aligned with the needs and business objectives of both customers and suppliers during
the whole product lifecycle” (Kajko-Mattsson et al., 2011). This research work shows how
web-based eMaintenance solution can use the cloud to analyse and represent train track
forces and wheel profile data to the end-users/stakeholders. It draws on two data sources:
forces data and wheel profile data. The data are integrated to ensure effective and
efficient decision making. Data from both sources are collected from the wayside
measurement stations and sent to a field computer through wired connections; Radio
Frequency Identification (RFID) technology has been used for data acquisition (Kour
et al., 2014). These collected data are sent to the eMaintenance cloud in the eMaintenance
lab for data filtering, integration, analysis and, finally, visualisation to the relevant service
personal. Figure 2(a) shows how the data are acquired from the RFID Reader and send to
field computer through wired network and then accessed by eMaintenance cloud through
internet where it is filtered, fused, integrated and analysed to take decisions based on the
analysed data. Figure 2(b) is the extension of Figure 2(a) and shows the example of wheel
profile data, i.e. train identification, car number, train side, flange thickness, flange height,
rim thickness, flange slope, tread hollow, diameter, tread hollow alarm, wheel width, etc.
and force data, i.e. vertical mean force, lateral force lower left, lateral force lower right,
lateral force lower max, angle of attack, vertical transient, etc. present in the database of
eMaintenance cloud. Further, eMaintenance cloud provides an integrated platform based
on Microsoft IIS (Internet Information services), Mobile App (application software),
Messaging Server, Windows Server or any other App for the extraction and visualisation
of railway data to make decision for improved maintenance planning.
As noted above, the eMaintenance cloud can assist in online data analysis of railway
traffic for improved maintenance. The eMaintenance cloud provides web services shown
 JQME (a)
20,3 Any Other app Windows Server Messaging Server Mobile App. IIS Web Server

SQL

266
Web Services
Downloaded by LULEA UNIVERSITY OF TECHNOLOGY At 06:54 24 May 2016 (PT)

Database Administrator Database

FTP Server

eMaintenance Cloud

XML
File

XML
File

RFID Reader Field Computer

Train

(b)

• ID • ID • ID
Figure 2.
• Flange (height, • Vertical • Mileage
eMaintenance cloud with thickness, slope) • ...
• Horizontal
web services (a) and Wheel • Rim thickness Wheel • Angle-of-attack Vehicle
example of wheel profile Profile
• Tread hollow
Forces • ... Data
and force data (b) • ...
MD MD MD

in Figure 2(a) through which information can be exchanged electronically via desktop
computers or mobile devices. A Web service is a XML (eXtensible Markup Language)
software system designed to support interoperable machine-to-machine interaction
over a network (W3C, 2014; Web services, 2014). This interoperability is gained through
a set of XML-based open standards. These standards provide a common approach for
defining, locating, publishing and using web services. Data are visualised through
the process of virtualisation which provides a shared view and information services
to different stakeholders. Therefore, relevant information and knowledge services are
provided in a shared view to different stakeholders via eMaintenance cloud using
virtualisation technology.

3. Railway maintenance decision making

To provide safe and reliable railway service, the quality of the rolling stock and
infrastructure needs to be improved and maintained. Maintenance is the combination
 of any actions carried out to retain an item in, or restore it to, conditions in which it is eMaintenance
able to meet the requirements of the relevant specification and perform its required solution through
functions (IEC, 2014). Therefore, there is a need to develop effective maintenance
decision-making system to achieve the goals and requirements of railway stakeholders. online data
Decision making can be considered as problem solving activity which is terminated
with a satisfactory solution. Decision making in railway is a complex issue; which
include huge amount of data, lack of relevant data, data analysis and data integration. 267
The wayside and onboard diagnostics equipment produce huge amounts of data and it
becomes very difficult to pinpoint possible faults from these messy data. Parida (2006)
stated that the lack of relevant data and knowledge is one of the main problems for
decision making within the maintenance process. Thus, there is a need to develop
Downloaded by LULEA UNIVERSITY OF TECHNOLOGY At 06:54 24 May 2016 (PT)

effective decision-making tools to perform advanced data analysis and data mining
to get relevant knowledge and information from large and messy data. This research
work uses eMaintenance cloud through online data analysis and the visualisation of
relevant information to assist railway operators in making maintenance-related
decisions. A generic maintenance process consists of phases for management, support
planning, preparation, execution, assessment and improvement (IEC, 2014). Hence,
maintenance is multidisciplinary and involves a wide range of roles, such as managers,
process owners, maintenance technicians, maintenance planners and logistic managers
(Karim, 2008). All these actors involved in maintenance process need right information
for making right decisions at right time (Karim et al., 2009; Parida, 2006).
In the present work, extensive consultation with stakeholders has resulted in the
analysis of track forces data and wheel profile data. The railway maintenance decision
making starts from initial data collection to data presentation in the form of web
services. The collected data are integrated to compile data and information from
the two-way side measurement stations (Figure 3) and sent to the eMaintenance lab,
which provides cloud services to enable diagnostic and prognostic eMaintenance
decisions. More specifically, the eMaintenance cloud provides web services to share
online data so that the needs of various stakeholders can be met. In addition, data
analysis is a process of inspecting, cleaning, transforming and modelling data with the
goal of discovering useful information, suggesting conclusions, and supporting
decision making (Data analysis, 2014). Therefore, this analysed data can be useful in
making maintenance related decisions.

4. Case study and methodology

This case study was undertaken for Trafikverket (Swedish Transport Administration)
at northern part of Sweden. Data are collected from the railway wayside measurement
stations located in northern Sweden as depicted in Figure 3. The wayside and onboard
diagnostics equipment used by operators produce huge amounts of data. However, this

Measurement
station 1
(Force Data)
eMaintenance Data analysis tool for Maintenance
Cloud Decisions decisions
Measurement
Figure 3.
station 2
(Wheel Profile Data) Railway maintenance
decision making
Web Services
 JQME paper is limited to wheel profile data and track force data for maintenance decision
20,3 making. This collected data are then acquired by railway eMaintenance cloud and then
filtered, integrated and analysed for cost effective maintenance decision making
and adapted for visualisation of various stakeholders. The railway eMaintenance
cloud provides web services through which exchange of information can be done
electronically. One of the expected applications of Web services is that they will make it
268 easier to share real-time data about identification and tracking of trains electronically
(Kour et al., 2014). The research methodology consists of data collection, data
integration, data analysis, data visualisation and finally, decision making as depicted
in Figure 3.
Downloaded by LULEA UNIVERSITY OF TECHNOLOGY At 06:54 24 May 2016 (PT)

4.1 Data collection

Data are collected from two railway wayside measurement stations at northern part
of Sweden. The first data source is the automatic wheel profile measurement station.
This station measures wheel profile data, such as flange thickness, flange height, rim
thickness, flange slope, tread hollow, diameter, etc. (Figure 4). The second measurement
station collects track force data, including lateral and vertical forces, vertical load, vertical
transient, angle of attack, etc. An RFID Tag is implanted on vehicle of a train; it receives
a signal from the RFID reader and gives a return signal with its identification number.
The RFID reader then reads the data and send them to a nearby database server called
field computer for storage as shown in Figure 2(a). These measurement stations
are connected to the eMaintenance cloud via internet connection. The purpose of both
wayside measurement stations is to continuously retrieve data for ongoing and future
research projects, improve the maintenance of both rolling stock and infrastructure
to diagnose the condition of the wheel and rail and to optimise their maintenance (Lulea
Railway Research center, 2014).

30 qR
New
Worn
20
Length (mm)

Sh
Sd
10

0
TH

–10 Flange Tread

0 20 40 60 80 100 120
Height (mm)
Figure 4.
Original and worn
wheel profiles measured Notes: The wheel profile parameters are Sh, flange height; Sd, flange thickness; qR, flange
by the wheel profile slope/angle; and Th, tread hollow
measurement system
Source: Asplund (2014)
 4.2 Data integration and data analysis eMaintenance
In the present research work, two data sources as depicted in Figure 3 are integrated solution through
so that useful information can be extracted from a unified view of data. Data are
integrated by using integration key to inter-link data sets from these data sources. online data
For analysing data, its quality should be checked as early as possible. There are
various data analysis methods, algorithms and approaches of modern data analytics.
This paper uses web-based algorithm to analyse data. Therefore, eMaintenance cloud 269
provides web services through which the information can be disseminated electronically.
Therefore, collected data are analysed to extract information and knowledge for making
maintenance decisions more effectively and efficiently.
Downloaded by LULEA UNIVERSITY OF TECHNOLOGY At 06:54 24 May 2016 (PT)

4.3 Data visualisation

Simply stated, data visualisation supports maintenance decisions. The present
work uses .NET technology and SQL Server 2012 to develop the web application, i.e. an
eMaintenance data analysis solution to present information and knowledge services
to railway maintenance stakeholders. These services are provided in a shared view to
different stakeholders via eMaintenance cloud using virtualisation technology in the
form of grids and graphs as shown in Figures 5-10 of Section 5. Therefore, the solution
provided here is a user graphical interface for visualisation of data. Criteria for
visualisation can be user specific to find specific deviations. In this case, the
eMaintenance solution conducts online data analysis of the condition of each wheel.
In the app, if any wheel is visualised as approaching or crossing its threshold limit, the
railway operator is informed to make the appropriate maintenance-related decision.
Finally, depending upon the visualised information related to either force data or
wheel profile data, fact-based maintenance decisions are taken for developing more
efficient and economically sound maintenance policies. Thus, this concept for a
web-based eMaintenance solution for railway maintenance stakeholders helps them to
visualise data in a graphical view and leading to effective maintenance decision making.

5. Results and discussion

The eMaintenance solution is a graphical user interface which allows stakeholders
to visualise data depending on their specific needs. Figure 5 shows a web-based

Figure 5.
Screen shot of web-based
eMaintenance solution
 Downloaded by LULEA UNIVERSITY OF TECHNOLOGY At 06:54 24 May 2016 (PT)

20,3

270
JQME

Figure 6.
Charts showing axle

52 of a specific train

wheel profile data (b)

range between 33 and

(a) and forces data and

(a)

(b)
400
358.8
339.9
304.3
300 278.8
264.8
250.6 245.7
238 232 234.1
224.3 228.2
203.4 204.8
200 189.9 181.8
153.9 158.3
133.2 135.7

100

Vertical Mean force in KN

0
32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53
Axle Number

12 11.5 11.1 11 11.1 11 11.1

10.5 10.7 10.7 10.5 10.6 10.8 10.4 10.7 10.8 11.6
10.3 10.311.110.6 10.3 10.5 10.2 9.9
10.7 10.7 10.7 10.8 9.5 10.9 9.8
10 10.5 9.7 9.2 10.4 9.2 10.4
9.8 9.8 8.9
8
6
4
2

Flange Slope in mm
0
32 37 42 47 52
Axle Number
 eMaintenance
solution through
5
online data
Total Wheels Passed Threshold

4
4

3 271
2

1 1
1
Downloaded by LULEA UNIVERSITY OF TECHNOLOGY At 06:54 24 May 2016 (PT)

0
2013–03

2013–05

2013–10

Month
Figure 7.
Flange Height Year–Month Vehicle RFID train side Axle sequence number Screen shot of wheel
35.2 2013–03 732325_847447213111 R 61 profile parameter flange
35.06 2013–03 732325_847447213111 R 64
35.04 2013–03 732325_847447213111 R 93
height indicating six
35.02 2013–03 732325_847447213111 R 117 wheels are between the
35.38
35.1
2013–05
2013–10
732325_847447213038
732325_847447216370
L
L
73
109
threshold limits 35 and 36

eMaintenance solution; it provides the historical data for a number of trains passing
by the wayside measurement station. We can select any specific train (Figure 5) to
visualise track force data and wheel profile data as shown in Figure 6(b). Thus, depending
on the specific period, data on trains can be visualised for end users in the form of either
grids or graphs. For any range of axles, for example, axles 33-52 on a specific train shown
in Figure 6(a), the track force data, e.g. vertical mean force in kiloNewton (kN) and the
wheel profile data, e.g. flange slope in millimetre (mm) for left and right wheel can be
presented to end users in a graph shown in Figure 6(b). In addition, different wheel profile
parameters within the threshold limit can be displayed and data is analysed; i.e. how
many wheels have crossed a certain threshold limit is presented in Figure 7.
Table I shows the maintenance limits set by the operator and safety limits
determined by the infrastructure manager.
A screen shot of the wheel profile parameter flange height shown in Figure 7
indicates six wheels are between the threshold limits 35 and 36. In other words, these
six wheels are beyond the maintenance threshold limit (Table I) of this parameter. The
vehicle RFID and axle sequence numbers for these same six wheels are shown in
Figure 7 in the form of a grid; this tells us where these wheels are located, i.e. which
axle of which vehicle. All this information can be sent to the relevant railway
maintenance stakeholders, allowing them to take the appropriate maintenance actions.
In addition, data are analysed to determine how many wheels have crossed the
maintenance limits set by the operator and safety limits determined by the infrastructure
manager. Figure 8(a-d) shows total number of wheels that that have crossed
the maintenance limits and safety limits for the year 2013 for the four wheel profile
parameters, namely, flange height, flange thickness, flange slope and thread hollow.
The X-axis shows different months, and the Y-axis shows the total number of wheels that
have crossed the maintenance limits and safety limits in a particular month. Clearly, these
data can be useful to railway maintenance decision making.
Further, eMaintenace solution offers graph for cumulative axle count (Figure 9)
and cumulative load (Figure 10). Figure 9 shows cumulative axle count for months
 Downloaded by LULEA UNIVERSITY OF TECHNOLOGY At 06:54 24 May 2016 (PT)

20,3

272
JQME

year 2013
Figure 8.

crossing maintenance
flange height (a), flange

different months for the

(c) and thread hollow (d)

limits and safety limits in

Wheel profile parameters:

thickness (b), flange slope

(a) Flange height > 34 (Maintenance Limit) (b)
Flange height > 36 (Safety Limit) Flange thickness > 22.5 (Maintenance Limit)
Flange thickness > 22 (Safety Limit)
1,241
1,200 1,109 700 656
1,007 651
1,000 600

775 500
800

582 632 400

602 340
600 542
480 487 331
447 300
335
400

Total number of wheels

200

Total number of wheels

200
100
49 41 49 26 1 28 36 31 24 22 11 15
20 16 1 5 16 16 18 20 5 5 16 20 13
2 1 0 11 11 1 1
0 0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Jan 2013–Dec 2013 Jan 2013–Dec 2013

Flange slope < 7 (Maintenance Limit) Tread Hollow > 1.5 (Maintenance Limit)
(c) (d) Tread Hollow > 2 (Safety Limit)
Flange slope < 6.5 (Safety Limit)
350 1,200 1,225
324
286 292 1,100 1,033
300 275
1,000

250 900
213 221
800 704 701
673
200 700 641 616 633
145
600 504 519
434
150 125 500
400 301
82
Total number of wheels

100

Total number of wheels

300 218 242
41 200 120 117 123
50 94 89
19 17 15 16 67 67 59 55
5 4 4 3 3 100 35
2 2 0
2 2
0 0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Jan 2013–Dec 2013 Jan 2013–Dec 2013
 400,000 eMaintenance
384,690
solution through
online data
300,000
Cumulative Axle Count

240,156

200,000 185,617 273

100,000
Downloaded by LULEA UNIVERSITY OF TECHNOLOGY At 06:54 24 May 2016 (PT)

105,362
Figure 9.
Chart showing cumulative
0 axle count in months two
1 2 3 4 5 to four for the year 2013
Month

250
Cumulative Load GigaNewton

200

150
110.8 115.2
100.1 105.1
93.4
100 79.1 86.2
65.9 72.7
58.8
48.1 52.4
50 32.4 37.7 43.3 Figure 10.
19.9 26.2
8.6
14.4 Chart showing cumulative
2.7
0 load in weeks three to 22
2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 for the year 2013
Week number

Maintenance limit (mm) Safety limit (mm)

Flange height 34 36
Flange thickness 22.5 22
Flange slope/angle 7 6.5 Table I.
Hollow wear 1.5 2 Safety and maintenance
limits for wheel
Source: Asplund (2014) parameters

two to five for the year 2013 and represents cumulative sum of number of total
axles of different trains passed in respective months. In addition, Figure 10 shows
cumulative load, i.e. vertical mean force in GigaNewton for weeks three to 22
for the year 2013. For both cumulative loads and cumulative axle count data can
be displayed as monthly, quarterly and yearly. With the help of this analysed
data the diagnostic decisions can be taken into place. Thus, this work is done as an
approach to help efficient maintenance policies to be adopted for maintenance
decision making.
 JQME 6. Conclusions and future work
20,3 The paper shows how the eMaintenance solution through online data analysis can
assist railway maintenance stakeholders in their decision making and how relevant
information can be visualised. In this present research, the concept of web-based
eMaintenance solution led the maintenance operator and infrastructure manager to
make decisions for maintenance planning. In the proposed solution, to enhance the
274 performance of the maintenance process and to become more economically effective,
data are acquired from wayside measurement stations and sent to eMaintenance cloud
for storage and analysis; the relevant information, such as wheels reaching their
maintenance limits and safety limits, is then visualised in grids or graphs to assist
stakeholders in making the appropriate maintenance decisions. Further, cumulative
Downloaded by LULEA UNIVERSITY OF TECHNOLOGY At 06:54 24 May 2016 (PT)

axle count and cumulative load will help the railway stakeholders to predict the condition
of track. Thus, a web-based eMaintenance solution, which is a graphical user interface,
assists in railway wheel profile and force data analysis for improved maintenance
decision making. For future work if maintenance data is coupled to the data analysis tool,
then diagnosis of previous maintenance interventions can be added to the model for
better prognosis. Further, prognosis of wheel wear can help the planning of maintenance
which in turn will help prevent bottlenecks and assist in economic planning. There
may be a need to present aggregated results to different stakeholders, i.e. each is given
information relevant to their needs.

References
Amazon (2014), “Amazon elastic compute cloud (Amazon EC2)”, available at: http://aws.
amazon.com/ec2/ (accessed 20 January 2014).
Asplund, M. (2014), “Wayside condition monitoring technologies for railway systems Luleå”,
Licentiate thesis, Luleå University of Technology, Luleå, p. 108.
Candell, O. and Karim, R. (2008), “eMaintenance – Information driven maintenance and support”,
Proceedings of the 18th International Conference on Flexible Automation and Intelligent
Manufacturing, 30 June-2 July, pp. 365-372.
Candell, O., Karim, R. and Parida, A. (2011), “Development of information system for e-maintenance
solutions within the aerospace industry”, International Journal of Performability Engineering,
Vol. 7 No. 6, pp. 583-592.
Data Analysis (2014), “Data analysis”, available at: http://en.wikipedia.org/wiki/Data_analysis
(accessed 25 March 2014).
Heuwinkel, K., Deiters, W., Konigsmann, T. and Loffeler, T. (2003), “Information logistics and
wearable computing”, Proceedings of the 23rd International Conference on Distributed
Computing Systems, pp. 283-288.
IEC (2014), “International electrotechnical vocabulary (IEV)”, available at: www.electropedia.org/
(accessed 19 March 2014).
Kajko-Mattsson, M., Karim, R. and Mirjam, A. (2011), “Essential components of e-maintenance”,
Essential Components of e-Maintenance, Vol. 7 No. 6, pp. 555-571.
Karim, R. (2008), “A service-oriented approach to e-Maintenance of complex technical systems”,
doctoral thesis, Luleå University of Technology, Luleå, p. 91.
Karim, R., Candell, O. and Söderholm, P. (2009), “E-maintenance and information logistics:
aspects of content format”, Journal of Quality in Maintenance Engineering, Vol. 15 No. 3,
pp. 308-324.
Kour, R., Karim, R., Parida, A. and Kumar, U. (2014), “Applications of radio frequency identification
(RFID) technology with eMaintenance cloud for railway system”, International Journal of
Systems Assurance Engineering and Management, Vol. 5 No. 1, pp. 99-106.
 Lulea Railway Research Center (2014), “The research station”, available at: www.jvtc.ltu.se/ eMaintenance
project/research-station (accessed 10 February 2014).
solution through
Moore, W.J. and Starr, A.G. (2006), “An intelligent maintenance system for continuous cost-based
prioritisation of maintenance activities”, Computers in Industry [special issue on online data
e-maintenance], Vol. 57 No. 6, pp. 595-606.
Muller, A., Marquez, A.C. and Iung, B. (2008), “On the concept of e-maintenance: review and
current research”, Reliability Engineering and System Safety, Vol. 93 No. 8, pp. 1165-1187. 275
NIST (2014), “National institute of standards and technology”, available at: http://csrc.nist.gov/
publications/nistpubs/800-145/SP800-145.pdf (accessed 19 March 2014).
Padhy, R.P., Patra, M.R. and Satapathy, S.C. (2011), “Virtualization techniques & technologies:
state-of-the-art”, Journal of Global Research in Computer Science, Vol. 2 No. 12, pp. 29-43.
Downloaded by LULEA UNIVERSITY OF TECHNOLOGY At 06:54 24 May 2016 (PT)

Parida, A. (2006), “Maintenance performance measurement system: application of ICT and

e-Maintenance concepts”, International Journal of COMADEM, Vol. 9 No. 4, pp. 30-34.
Rouse, M. (2011), “Virtual machine (VM)”, available at: http://searchservervirtualization.
techtarget.com/definition/virtual-machine (accessed 10 February 2014).
Salesforce (2012), “A complete history of cloud computing”, available at: www.salesforce.com/uk/
socialsuccess/cloud-computing/the-complete-history-of-cloud-computing.jsp (accessed 20
January 2014).
Sourya, B. (2011), “A history of cloud computing”, available at: www.cloudtweaks.com/2011/02/
a-history-of-cloud-computing/ (accessed 20 January 2014).
Wandt, K., Karim, R. and Galar, D. (2012), “Context adapted prognostics and diagnostics”, The
Ninth International Conference on Condition Monitoring and Machinery Failure Prevention
Technologies, 12-14 January, London, p. 10.
W3C (2014), “World wide web consortium (W3C)”, available at: http://w3c.org/ (accessed 19
March 2014).
Web services (2014), “Web service”, available at: http://en.wikipedia.org/wiki/Web_service (accessed
11 February 2014).
Wikipedia (2014), “Cloud computing”, available at: http://en.wikipedia.org/w/index.php?
title¼Cloud_computing&oldid¼535823947 (accessed 2 February 2014).
Zhang, N., Liu, F., Miao, K. and He, J. (2010), “A novel cloud computing paradigm for China
railway application”, IEEE International Conference on Intelligent Computing and Intelligent
Systems, 29-31 October, Xiamen, pp. 213-217.
Zurawski, R. (2006), Integration Technologies for Industrial Automated Systems, CRC Press,
New York, NY.

Corresponding author
Ravdeep Kour can be contacted at: ravdeep.kour@ltu.se

To purchase reprints of this article please e-mail: reprints@emeraldinsight.com

Or visit our web site for further details: www.emeraldinsight.com/reprints
 This article has been cited by:

1. Jaime Campos, Shaofeng Liu. 2016. Managing the information systems in the industrial domain. Cogent
Business & Management 3:1, 1180967. [CrossRef]
Downloaded by LULEA UNIVERSITY OF TECHNOLOGY At 06:54 24 May 2016 (PT)

View publication stats