Study of Aspects of Data Quality in eMaintenance

Mustafa Aljumaili1, Phillip Tretten1, Ramin Karim1 and Uday Kumar1

1

Division of Operation & Maintenance engineering,

Lule University of Technology, Sweden

Mr. Mustafa Aljumaili obtained his Bachelor and M.Sc. degrees in Computer Science from
Nahrain University, Baghdad. He is a lecturer at the college of Information Engineering in Nahrain
University. He is currently a doctoral student at LTU in the area of eMaintenance.

Dr. Phillip Tretten is a researcher at the Lule University of Techonology. His academic experience
has been in the area of Industrial Design and Engineering Psychology, where he is published and
recognized in the fields of Traffic Psychology and Human Factors. He has also supervised, taught
and advised undergraduate and graduate students.

Assoc. Prof. Ramin Karim is Associate Professor at Lule University of Technology (LTU). He has
a Ph.D. in Operation & Maintenance Engineering. He worked within the Information &
Communication Technology (ICT) area for 18 years, as architect, project manager, software
designer, product owner, and developer. Dr. Karim is responsible for the research area eMaintenance
at LTU, and has published more than 30 papers related to eMaintenance.
Prof. Uday Kumar obtained his B. Tech. in India during the year 1979. He joined the postgraduate
program of Lule University of Technology, Sweden, and obtained his PhD in the field of Reliability
and Maintenance during 1990. Afterwards, he worked as a senior lecturer and an associate professor
at Lule University of Technology. His research interests are equipment maintenance, equipment
selection, reliability and maintainability analysis, system analysis, etc. He has published more than
170 papers in international journals and conferences.
ABSTRACT
Effective and efficient maintenance requires proper information logistics, which can be delivered through
eMaintenance solutions. An eMaintenance solution provides services for data acquisition, data processing, data
aggregation, data analysis, data visualization, context-sensing, etc. The development of eMaintenance solutions
faces challenges and one of these challenges is how to ensure a high quality of data. Data Quality (DQ) concerns
all the phases of the maintenance process.
Hence, the purpose of this paper was to explore and identify the aspects of DQ in eMaintenance. To achieve this,
seven case studies, within three industries, were conducted. The empirical data was collected through interviews,
observations, archival records and workshops. In this paper, the different aspects related to DQ were presented
with their relationship to the phases of a generic maintenance process. Furthermore, the DQ issues have been
related to how data is generated, either by manual input (Human-generated) and/or automatic input (Machinegenerated). One of the main findings of this paper shows that the organizations do not often implement
maintenance solutions whole-heartedly, thus, the application of the systems never are used to their full potential.
Keywords: eMaintenance, maintenance process, information logistics, Data Quality (DQ), Information and Communication
Technology (ICT), Information Quality (IQ), Information System (IS).

1. INTRODUCTION
Development of Internet Communication
Technologies (ICT) has contributed to the emergence of
the eMaintenance concept. Since its conception in the
early 2000s, eMaintenance has become a common term in
maintenance-related literature [1]. The emergence of
eMaintenance, e.g. ICT solutions implemented in
maintenance, has led to a revolutionary change in the
maintenance of industrial assets [2]. There are several
industrial and academic contributions that describe
eMaintenance. It can be defined as a maintenance strategy
where tasks are managed electronically using real-time
equipment data obtained through digital technologies (i.e.
devices that are mobile, remote sensing, condition
monitoring, knowledge engineering, tele-communication
and internet technology based) [3]. It is also explained as
a maintenance plan to meet productivity through condition
monitoring, proactive maintenance and remote
maintenance through real-time information for decisionmaking [3]. To sum it up eMaintenance can be termed as
maintenance support for e-operation through remote
diagnostics and asset management, simulation for
optimisation and decision making under an e-business
scenario for an organisation. In a provider and consumer
relationship scenario, there are, a number of views on
what eMaintenance actually is and the most common one
is the application of ICT to achieve effective information
logistics within the maintenance area [4]. In conclusion,
eMaintenance solutions aim to facilitate maintenance
decision-making by providing effective and efficient
maintenance information logistics.
In a decision-making process, the quality of a
decision is strongly linked to the quality of the data used
during the data analysis [4]. When dealing with
maintenance information logistics, content management
(i.e. what to deliver) is highly important, since it
addresses issues related to the content; this includes both
data and information [5]. Hence, aspects related to Data
Quality (DQ) need to be emphasised in solutions aimed to
support the decision-making process. DQ in maintenance
forms the integral part of information logistics, which is
the backbone of a maintenance system derived from a
clear maintenance strategy that, in turn, should be derived
from, and linked to, the corporate strategy. Hence, the
purpose of this paper was to explore and identify the
aspects of DQ in present eMaintenance solutions.

2. THEORITICAL FRAMEWORK
2.1 eMaintenance
Maintenance is the combination of all technical
and administrative actions, and managerial including

supervision, during the life cycle of an item intended to

retain it in, or restore it to, a state in which it can perform
a required function [6]. A generic maintenance process
consists of phases for management, support planning,
preparation, execution, assessment and improvement [7].
The emerging applications of ICT are to support
companies to shift manufacturing operations from the
traditional factory integration philosophy to an e-factory
and e-supply chain philosophy [8]. From a general
perspective eMaintenance concerns the use of new ICT
solutions in the maintenance area [9].
The management of maintenance consists of the
following activities: developing and updating the
maintenance policy, providing finances for maintenance,
coordinating and supervision of maintenance, as shown in
Figure 1 [7]. The elements of maintenance support
planning
are:
maintenance
support
definition,
maintenance task identification, maintenance task analysis
and maintenance support resources [7]. Maintenance
preparation concerns the planning for maintenance tasks,
which includes planning of specific tasks, scheduling
activities and assigning and obtaining resources [7]. The
maintenance execution phase includes the actual
performance of maintenance, recording of results and
special safety and environmental procedures. Maintenance
assessment includes measurement of maintenance
performance, analysis of results and assessment of actions
to be taken. Finally, maintenance improvement is
achieved by improving the maintenance concept,
improving the resources, improving the procedures and
modifying the equipment that is maintained [7].

Figure 1: Phases of an overall maintenance process, adopted from [7]

eMaintenance is a multidisciplinary domain based

on maintenance and ICT, which has a focus upon ensuring
that eMaintenance services are aligned with the needs and
business objectives of both customers and suppliers
during the whole product lifecycle [10]. eMaintenance
also can be defined as a new concept, a part of
maintenance support, which provides
information
resources and information services, which can be used to
enable the development of and establishment of a
proactive decision making process through the enhanced

use of ICT [5]. From a generic perspective, eMaintenance

is maintenance managed and performed via computing
[11].
eMaintenance can be considered as a maintenance
management system where assets are monitored and
proactive maintenance decisions are transported over the
internet and with other ICT tools [12]. eMaintenance
provides companies with predictive intelligence tools to
monitor their assets (equipment, products, process, etc.)
through the Internet and wireless communication systems
to prevent them from unexpected breakdowns. Also, these
systems can compare a product's performance through
globally networked monitoring systems, thus allowing
companies to focus on degradation monitoring and
prognostics rather than fault detection and diagnostics
[13]. Therefore, eMaintenance technologies increase the
possibility to: 1) utilize data from multiple origins, 2)
process large volumes of data and make more advanced
reasoning and decision-making and 3) implement
collaborative activities [14].
Although the eMaintenance concept is a common
term in maintenance-related literature [11] there is no
clear consensus of what eMaintenance is. Some consider
eMaintenance as a maintenance strategy, where tasks are
managed electronically by the use of real-time item data
obtained through digital technologies [3]. Others view
eMaintenance as a support to execute a proactive
maintenance decision process [11]. Furthermore,
eMaintenance also can be considered as the integration of
all necessary ICT-based tools for optimization of costs
and improvement of productivity through utilization of
Web services [15]. In this paper we adopt a broad view on
eMaintenance in line with Kajko et al. (2010) [11], who
defines it as maintenance processes managed and
performed via computing. This definition includes
maintenance activities in all phases of the maintenance
process and it includes a variety of ICT solutions ranging
from computerised maintenance systems to sensor
technologies.
Due to the improvements eMaintenance has
contributed to this has led to an increased attention in
research. Muller et al. (2008) identifies four general issues
addressed in eMaintenance literature: 1) standards, 2)
platform development, 3) process formalization and 4)
system development and implementation [16]. Besides
those issues, the literature also addresses eMaintenance
from an information logistics perspective [3] [17] [18].
Maintenance processes are supported by heterogeneous
resources, such as, documentation, personnel, support
equipment, materials, spare parts, facilities, information
and information systems [19]. Hence, the provision of the
right information to the right user with the right quality
and in the right time is essential [20] [21] [18]. This

desirable situation can be achieved through appropriate

information logistics, which aims to provide just-in-time
information to targeted users and to optimize the
information supply process. While the provision of justin-time information to the right users is an essential issue
in maintenance work we propose in addition to that the
issue of correct information at the correct time, i.e.
information based upon high quality data.

2.2 Data Quality

Quality plays an important role as one of the
powerful competition advantages for those companies that
run businesses in the information society [22]. Quality can
be defined as fitness of an artefacts (e.g. product, service,
and function) performance to its specifications and
intended use, from the customers perspective [23]. For
most organizations, quality management is well known
and is integrated into the daily regime of improving the
product or services quality.
Significant effort has been done in order to define
what is meant by Data Quality. DQ can be defined as
data that is fit for use by data consumers [24]. Commonly
it was thought that DQ mainly focused on attributes like
accuracy, precision and timeliness. Levitin et al., (1998)
considered two important aspects in trying to ensure a
high DQ by: 1) data models that are clearly defined and
2) data values that are accurate [25]. In the Computerized
Maintenance Management Systems (CMMSs), three
important parties may affect DQ: data producers, data
custodians and data consumers. Data producers are people
or systems that generate data. Data custodians are people
who provide and manage computing resources for storing
and processing data. Finally, data consumers are people or
systems that use data. Therefore, data users are critical in
defining data quality [26].
The consequences of poor DQ are significant to
businesses, governments and society in general [27]. High
quality information is dependent on the quality of the raw
data and the way it is processed. Since data processing has
shifted from a role of operations support to become a
major aspect in operations itself, therefore, the need for
quality management of data has become more important
[28]. Many similarities exist between the manufacture of
quality data and the manufacture of quality products, such
as, conformity to specifications, lowered defect rates and
improved
customer
satisfaction
[29].
Data of
poor quality will result in customer dissatisfaction, lost
revenue and higher costs, which are associated to the
spending of additional time to reconcile data. On a larger
plan this can lead to customers losing credibility in a
system, as well as, an increased risk of noncompliance to

regulations. Failure in high DQ also increases consumer

costs, increases taxes, decreases shareholder value and can
even lead to mission failure [27].
DQ principles have become a core business
practice in fields, such as, business, medicine, geographic
information systems, remote sensing and many others. It
is often the case that data which is used uncritically,
without consideration to the potential errors contained
within, can lead to erroneous results and misleading
information [30]. A short survey of DQ literature has been
conducted to identify how DQ has been considered in
some areas of management, engineering and medicine, to
provide a sample of DQ aspects:
Wang et al., (1995) developed a framework that can
be used to analyse existing research on DQ and
identify important future research directions. In
selecting the appropriate body of research for
examination, two primary criteria were used: 1) The
authors of the articles specifically recognize a DQ
problem, which they attempt to address in their work,
that is, the research is motivated by a DQ issue. 2)
The researchers addressed a problem that, although
not specifically described as a DQ issue (e.g., user
satisfaction with information systems), is comprised
of components that are related to DQ management.
Many of these authors papers are referenced by
research that falls into the first category [31].
Ballou and Tayi (1999) offered conceptual
framework for enhancing DQ in data warehouse
environments. They have explored the factors that
should be considered, such as, the current level of
DQ, the levels of quality needed by the relevant
decision processes, and the potential benefits of
projects designed to enhance DQ [32].
Leitheiser (2001) examined the issues health care
organizations face in trying to deliver high quality
information to clinical and financial end-users in an
environment with many diverse source systems and
organizational units with different business rules. A
model for understanding DQ issues in this
environment is developed and applied to a mid-sized
hospital health care organization. The suggested
model
called
(architected
data
warehouse
environment) which includes the following
components:
Data
source
systems,
Data
warehouse(s), Datamarts, End-user analysis tools,
and Transformation/Translation/Transportation tools
[33].
Xu et al. (2002) described an empirical study of DQ
issues during the implementation of ERP (Enterprise
Resource Planning) systems within organizations. A
model was designed to illustrate DQ issues in

implementing ERP (SAP) in two companies. Case

studies in two large Australian organizations were
used to provide useful insights into the nature of DQ
issues in implementing ERP [34].
Lee (2003) has suggested that professionals solve
problems by crafting rules to integrate business
process and data processing. They approach this
integration by explicating DQ contexts embedded in
data or re-establishing the missing contexts in data.
Reflecting on and explicating contexts dictates how a
DQ problem is framed, analysed and solved. He has
also suggested that context-reflective knowledge
about solutions must be recorded and shared.
Reflective context explicates knowledge about why a
company collects particular data, how the data was
stored, what constraints were imposed and how the
information was used [35].
Li et al. (2011) analysed knowledge maintenance
logs from the control flow perspective to find a good
characterization of knowledge maintenance tasks and
dependencies. In addition, the logs are analysed from
the organizational perspective to cluster the
performers who are qualified to do the same kinds of
tasks and to find relations among the clusters. The
proposed approach has previously been applied in the
knowledge management system. The results showed
that the approach was feasible and efficient [36].
We need to mention here Metadata is an
important part of every database. DQ is mainly concerned
with the data as content but Metadata is also an important
aspect that affects DQ. Metadata is a description of data
that can be described as data about data. It is also known
as a system catalogue [37]. Metadata represents a set of
concepts that describe the semantic content of a piece of
information. Therefore, good designed Metadata will help
in the production of high quality data.

2.3. Data Quality in eMaintenance

DQ is an important aspect to consider, since
without control of the DQ there is no control of the
accuracy of the output [38]. The reason is the amount of
information available continues to increase almost
exponentially in maintenance decision-making, the real
concern with data quality is to ensure that the DQ in our
information systems are accurate enough, timely enough
and consistent enough for the organization to survive and
make reasonable decisions [39].
However, maintaining a high quality data level
involves significant costs. These costs are associated with
efforts to detect and correct deficiencies, set governance
policies, redesign processes and investments in
monitoring tools. Thus, from an economic perspective one

would attempt to find an acceptable level of quality at a

minimum cost [40].
There are a number of contributions dealing with
DQ such as 1) ISO (2009) [27]; 2) Price and Shanks
(2005) [41]; 3) Wand and Wang (1996) [42], and Wang
(1996). In ISO standards, there is a description of DQ by
exploring data architecture, explaining the principles of
DQ, describing the characteristics of data that determine
its quality and finally specifying the processes to ensure
DQ [27]. Price and Shanks (2005) have suggested a
semiotic framework that classifies data into levels: form,
meaning and application according to its use and quality.
Different techniques to improve DQ have also been
suggested, such as, data profiling, data standardization,
linking and data cleaning. The need for high DQ is
important both for designing an eMaintenance solution
and operation of eMaintenance solutions [41]. In addition,
Wand and Wang (1996) have defined DQ as a theoretical
approach. In this approach, they used a set of assumptions
and definitions to derive DQ dimensions using a designbased perspective.
They tried to assume some
deficiencies of the data in the information system that can
be incomplete, ambiguous, meaningless or incorrect [42].
Finally, Wang (1996) takes an empirical approach to
define DQ. He has divided DQ into categories and related
them with some DQ dimensions. This empirical approach
can be summarized in Table 1 below [28].
DQ Category
Intrinsic
Accessibility
Contextual

Representational

DQ Dimensions
Accuracy, Objectivity, Believability,
Reputation
Access, Security
Relevancy, Value-Added, Timeliness,
Completeness, Amount of Data
Interpretability, Ease of understanding,
Concise Representation, Consistent
Representation

Table 1 DQ Dimensions [28]

2.4. Data Generation Categories

In maintenance, data used for decision making is
mainly generated automatically by use of technical
equipment, such as, a sensor or provided though human
intervention. Therefore, data can be divided mainly into
two categories based upon how it is generated; Humangenerated data, such as, data generated through interviews
and case studies, and Machine-generated data, such as,
condition based monitoring (CBM) data. However, there
are different issues related to DQ concerning these two
data generation categories.
1) Human-generated Data: Good quality data from

sensors depends on, e.g. sensor location, mounting,

measurement interval and timing of the measurement.
The DQ issue of this concerns when the measurements
are conducted and how the different individuals in the
organization interpret this. All these factors affect the
DQ. Good DQ is dependent on, e.g. professional skills,
experience and knowledge of the person observing the
target. Also, instructions, education and the common
practices of the organization are an important factor in
DQ. Human error is also another aspect affecting DQ.
Therefore, usability issues of information systems and
the different tools used for data collection are important
in reducing the possibility for human error.
2) Machine-generated data: There is an increasing
requirement to have maintenance data collected
automatically in real-time for decision-making. Data
can be collected using different types of sensors, such
as, temperature, pressure etc. This information is then
stored directly to a computerized database. In addition,
there are different Enterprise Resource Planning (ERP)
systems or Computerized Maintenance Management
System (CMMS). A CMMS is a software package that
maintains a computer database of information about an
organizations maintenance operations. All available
CMMS systems offer data collection facilities, although
more expensive systems offer formalised modules for
the analysis of maintenance data [43]. Some examples
of CMMSs are MAXIMO, Movex, IFS and SAP.
MAXIMO is asset management software that provides
insight for all of enterprises assets, their conditions and
work processes [44]. Movex, Enterprise Resource
Planning (ERP) software, is described as a basic
requirement in integrating between the head office and
sales units all over the world. The Movex Maintenance
part enables maintenance truly to be integrated with all
business areas [45]. SAP which stands for Systems,
Applications and Products in Data Processing is also
an ERP system that includes many tools for enterprise
management solutions, one of these solutions is Product
Lifecycle Management (PLM) and it is used for product
management, product data management and life cycle
management [46]. However; issues related to DQ could
be resulted from i) multiple internal and external data
sources, ii) free-form text containing multiple hidden or
inaccessible business iii) unexpected data values in
some fields [43].
3)

3. METHODOLOGY
In order to analyse and identify DQ aspects in
maintenance, several case studies in different industries
have been conducted. A summary of these case studies are
provided in next section. In the description of the case
studies specific information collected from the companies
and even the industries they concerned were to be held
confidential due to security concerns. The methods used
to collect and analyse information concerning DQ issues
were included in the case studies section below.

the responses. To gain a more complete picture about

issues related to DQ, peoples thoughts, feelings and
motives were also sought after.
Title

Task

Section
Manager

Automation and plant maintenance

Foreman

Mechanical maintenance

Mechanic

Mechanical maintenance

3.1. The Case Studies

Foreman

Electrical/automation maintenance

Case studies were conducted to learn how DQ

was maintained in eight large companies, of which,
represented three industries. Case studies were conducted,
in Scandinavia, at two mining companies, five pulp and
paper companies and an airplane manufacturer. The goal
of the case studies was to 1) compile a list of possible DQ
issues, and to 2) present it to the eMaintenance
community. Based upon the needs of the study, which was
to gain a more complete and unaltered understanding of
DQ issues in these industries were interviews used.
Before
the
interview
process
began,
documentation was requested for and provided by the
companies involved so that the organizational structure
and the operational processes could be studied more
thoroughly. Based upon that knowledge, individuals
holding specific positions in the organizations were
chosen according to their knowledge of the work
environment. The interviews were conducted using a
prepared battery of questions so that results could be
compared between the different actors and companies.
People from different positions and organization levels
were interviewed to get as many perspectives on the
maintenance process as possible, which included both
maintenance and production organizations. The focus of
the interviews was on data and information flow within
the maintenance process. This included the systems
supporting these flows and problems related to these
areas.
The interviews were conducted in two rounds; the
first round was conducted with maintenance and
production personnel and in the second round was
conducted with maintenance technicians. This was done
to understand the information flows and IT support for
maintenance execution. Table 2 shows titles and tasks of
the personnel interviewed. As is true with many interview
studies are the results are highly dependent on peoples
perspectives and experiences and, thus, if the interviewees
and interviews are not carefully prepared then
incongruences may arise resulting in unreliable data from

Mechanic

Electrical/automation maintenance

Engineer
Technician

Development of mechanical
maintenance, Condition monitoring
Preventive maintenance, KaTTi upkeep

Mechanic

Preventive maintenance

Operator

Production

Designer

Automation systems

Table 2 Titles and tasks of interviewed personnel

The goal of this work is to identify problems
related to DQ in all phases of the maintenance process.
The interviews were semi-structured with open-ended
questions and each lasted about one hour. The interview
structure was based on the sub phases of the maintenance
process, see Figure 1. These phases guided the semistructured interviews, followed by questions and
additional issues raised by the respondents. During the
interviews, the aim was to get an understanding of how
the different maintenance phases were supported by ICT
and the information flows throughout the phases. After
the interviews, problems related to information flow in
maintenance were compiled and structured according to
their relevance to DQ and impact on the different phases
in the maintenance process. All the interviews were
recorded and reviewed during the data analysis.

4. RESULTS
The results from the data analysis of the case
study data show that there are a number of fundamental
aspects of DQ in maintenance that need to be considered
when developing and establishing eMaintenance solutions
as support to maintenance decision making. After doing
the study in the case studies above, eMaintenance

solutions used in maintenance at the studied companies

are, with few exceptions, a number of stand alone
CMMSs with poor integration and manual data exchange.
In the study, a number of problems were identified related
to the use of ICT in the maintenance work.

4.1. Identified Data Quality related aspects

During this study, a number of problems have
been identified which related to the use of ICT in the
maintenance phases.
1. Stand-alone IT-systems: The CMMS was not used
comprehensively nor did it contain all information
required for operative and strategic decision making
in the different phases of the maintenance process.
Along with that, there were different sections in the
maintenance information systems, which were not
interconnected. Hence, that leads to data integration
problems, like maintenance management, and
information search difficulties. Data from production,
material management, parts storage and purchasing
related IT-systems were not connected to data in the
maintenance information system since they were
stand-alone IT-systems. The data was then
incompatible and difficult to integrate, which
complicated operative and strategic maintenance
decisions.
2. Data multiplicity: This problem concerns the
documentation and storage of duplicate information.
Certain
information
regarding
technical
documentation and drawings were stored both in the
maintenance system and in other media (i.e. computer
hard-drives, CD-ROMs, paper), although, in many
cases, there was no technical information, drawings or
work instructions available in the maintenance
system. This resulted in stored information that was
not regularly updated or maintained, thus making it
difficult for the users to know if it was reliable or not.
Moreover, maintenance information was stored in
different and non-correlating subparts of the system.
Therefore, conducting a historical overview of
previous maintenance activities was difficult. A
previous and more successful solution was paper
device cards, which, contained technical information,
drawings, work instructions and historical information
about previous maintenance actions. This was
successful since one person was always responsible
for updating these cards, resulting in much more
reliable information.

3. Manual input and transfer of data: The first problem

relates to the practice of manual data input. The
practice of manual data entry has resulted in data of
varying quality, due to the fact that, some of the
maintenance personnel unknowingly omit compulsory
data while others may enter very detailed and
descriptive failure reports. This results in the
maintenance foreman, more often than not, consulting
the operational staff for help and, additionally,
conducting a physical inspection of the previous
work. Secondly, since the data was manually
transferred between systems there was a risk of data
loss or corruption each time data was transferred and
re-entered. This made it increasingly difficult for the
personnel to look for additional information.
4. System usability: This problem points to the interface
and functionality of the systems used in the
maintenance processes. One interviewee stated that
the usability of the maintenance system was low
because it was difficult to find relevant information
and even guidance on how to use the system properly.
The search function in the systems were not well
designed and making it difficult to find relevant
information,
i.e.
historical
information
on
maintenance actions.
5. Accessibility to IT tools: Maintenance related ITsolutions are developed to improve different parts of
the maintenance process. However, IT-solutions were
often not available at the workplace. The most
common situation was that maintenance and operative
staff needed to access the IT-solutions through an
office PC. However, the most significant part of the
data from the maintenance process was collected from
the work place during maintenance execution part
(see figure 1). Since IT-solutions were separated from
the actual maintenance work, they could not support
maintenance actions correctly.
6. Performance indicators missing or difficult to obtain:
Maintenance assessment was based on information
from a software vender, where material costs, salaries
for the maintenance staff and costs for subcontractors
were followed up. However, material costs were only
followed on a sub-process level and not on equipment
level. Along with that were salaries only followed on
a plant level and there was no information available
for the efficiency of the maintenance organization
(due to inaccurate and invalid data of work time
execution and work order status). Thus, it was

7.

8.

9.

10.

11.

difficult to make assessments and give improvements

to the organization based on its actual performance.
No instructions or guiding: A lack of instructions and
guide available to the operation and maintenance staff
resulted in the lack of a plan or common practice
concerning what information, by whom and in what
form it was to be collected in, during the different
phases of the maintenance process. Since there were
different ways to collect and report data the
information found in the database was not consistent.
The lack of system helps made things even more
complicated for the users, thus, it was difficult to do
systematic data-analysis and to produce key
performance indicators.
Poor data upkeep: There was no systematic
maintenance/upkeep of the data, therefore the quality
of the data entered into information system decreased.
To be able to give reliable information from a large
amount of data, the data has to be both entered
correctly and in the correct form. To achieve that,
regular controls and correction of the data are
required, since there was no one responsible for these
things fell behind.
Problems with Metadata: Metadata is the description
of data and is mainly developed and added to the
system database during the implementation phase of
the information system, e.g. equipment hierarchy,
position numbers, devices and spare parts numbers.
Therefore, the gaps and mistakes in the Master-data
significantly affected the quality of the aggregated
data.
Data not used for strategic decisions: In this study,
the CMMS was mainly used for information transfer
and communication between people in different
positions and organizations, i.e., the sending and
receiving of work orders or accessing information
from the system. Thus, data in the CMMS was not
collected for operative and strategic decision-making.
That resulted in employees who did not see any
reason to collect nor enter good quality data into the
system.
No Knowledge recycling: It was difficult to share and
recycle knowledge from previously conducted
maintenance. In general, all unplanned shutdowns
lasting for more than one hour were followed up in a
root cause analysis. The section manager was
responsible to read and store everything in a Word
template. Since the organization did not allow for

easy access to historical data, saved files were rarely

consulted nor were previous work consulted. Another
point was that knowledge was not reused in assisting
the completion of work orders, thus, little information
was reused from previously recorded work orders
making it difficult to use the work orders as a source
of knowledge when similar errors occured in the
equipment.
12. Poor team cooperation: The internal structure did not
encourage cooperation between the maintenance
teams. In some of the studied cases, were some
failures were not specifically a mechanical or an
electrical problem, they could only be solved by both
the electrical and mechanical technicians working
together. The work orders were even designed for
either the mechanical or electrical groups. If one team
could not find the solution, they directed it back to the
other team and vice versa. The teams also had
separate break areas and foremen, thus, the
information sharing between the groups was not
facilitated by the system or the organizational
structure.
13. Poor IT competence: CMMSs are usually developed
to assist mainly maintenance organisations in their
tasks. In addition, some companies use the CMMS to
manage their whole operative and strategic actions. In
contrast, the CMMS was used only by the
maintenance organisations in the companies studied.
Therefore, many actions, made by the operative staff
and service providers, were not reported in the
CMMS. In several instances, they did not even have
any access to the CMMS. On top of that, another
problem was that not everyone in the maintenance
organization was able to use the CMMS, which
resulted in many situations where mechanics were not
collecting, reporting or seeking information from the
system. Instead, information was, at a later point in
time, entered by the foreman who did not conduct the
task nor did they observe it being completed. Due to
this, data in the CMMS could easily be incorrect
and/or incomplete. Thus, when the operative and
strategic decisions were made, they were based on
partial data, and the resulting decisions were often
built upon limited knowledge.

4.2. Rooting DQ issues to maintenance

The perceived problems, which were presented in
the previous section, are all rooted in as different aspects

of DQ. Table 3 gives an overview of the different

problems identified during the studies, and also a linkage
between the problems and the phases in the maintenance
process. This shows the relation between each DQ aspect
and each phase of the maintenance process that may affect
it. It also shows if the DQ issues are resulted from
Machine and/or Human sources during the maintenance
process.

Table 3: Identified DQ issues and linking them to

maintenance process.

5. DISCUSSION
Overall, the results show that the greatest question
concerning DQ pertains to the usefulness of the ITsystem. In many of the cases the system has
incompatibility issues, accessibility issues and/or usability
issues. From those cases several aspects arose and almost
all of them concern the user instead of the performance of
the computer system itself.
One of the main concerns was the complexity
and/or incompatibility of the users to effectively exploit
the given IT solutions. This inability has led to severe DQ
issues. Unfortunately, the phases of the maintenance
process, which produce the greatest DQ issues, deal
directly or indirectly with the personnel on the floor. This
would include those who plan, conduct and assess the
undertaken maintenance. This leads to probably the
greatest question facing each organization, that is, what
goal do they have concerning DQ? Since low DQ can
severely affect long and short term strategic planning it
may be worthwhile to seriously consider revamping some
areas of the maintenance structure. While other items
concerning Master data may require a greater effort to
repair, which may even requiring a system wide revamp
to remove the critical weaknesses.

There are also noticeable differences in the

problems associated to the Machine-generated and
Human-generated data. The main problem with the
Automatic Sources is that software compatibility issues
often arise from the system. This is most likely caused by
the organizations choice to implement standalone systems
and not use an integrated approach. Without deep insight
on the compatibility or the cost things may seem to be an
easy choice. Although to implement several different
systems in a plant wide network, the resulting costs often
are both economically costly, as well as, a system wide
cost resulting in systems with poor or no communication
with each other. Concerning Human generated data
sources, it has been shown to be great challenge for (11)
of the stated (13) areas of DQ issues as shown in Figure 2.
This shows that the level of human responsibility
in the maintenance system is quite large. IT solutions have
been known to reduce the level of human dependency in
the system and reduce the risk for human errors. The
success of these two factors should also, in turn, result in
a diminished need for specifically skilled workers. A
downsizing of the workforce did occur in tact with the
introduction of these systems and along with that the roles
of the service and support personnel have changed. The
result of this is IT solutions that now need application
specific upkeep. A new type of employee is now needed
in the organization and often the importance of this type
of personnel is overlooked. Without the proper personnel
the required upkeep is not undertaken resulting in
deficiencies in other areas. What has happened is that the
organizations initial attitude towards newly implemented
systems as being a cure all or an implement and
forget solution has not been true. After haven applied
this strategy, new and unforeseen needs have arisen
resulting, in some cases, an even less understanding and a
poorer general overview of the whole process. Overall,
the level of understanding of the production system is less
than it initially was before the new IT solutions were
installed, resulting in new and unforeseen production
problems. To combat these types of problems it is
necessary to understand the issues involved in
management decisions and systematically work towards
complete solutions.

Figure 2: Human and Machine related DQ issues.

Generally
speaking
people
within
the
organizations had differing understandings of how the IT
systems could be applied. From the level of
leadership/management were strategic decisions made
based upon unrealistic expectations, while those who used
the system saw the IT solutions as a threat to their
capabilities and, therefore, did not seek to fully
understand them. Due to these incorrect expectations the
resulting systems were not used to their full potential,
which indirectly resulted in DQ issues.
Another area of concern was the personnels
attitude towards IT solutions. Up until today, the push was
to use more IT solutions to improve efficiency and save
money. On the other hand, several cases showed that the
company culture did not assist this. The implementation
of a true or complete system was never possible. It
almost seemed that the leadership was not interested in
paying for the finished product, which involved a complex
and time consuming process. Without support from the
company leadership many of the improvements fell on
the wayside. A conclusion is that if the whole
organization is not completely convinced in the systems
usefulness it will most likely also result in a similar fate.
One way to prevent or reduce the negative
attitudes towards new implementation and the usage of IT
technology is to increase the IT competence of the whole
organization. With a better understanding of how IT
solutions can be used within the organization, personnel
will they be more positive towards its implementation.
Along with that, when they are part of the implementation
process, they are more likely to feel like system owners,
having responsibility in the IT system. This will help lead
a cultural change in the organization. Just as the results
have shown, there is a need to reduce manual input of
maintenance information and increase the IT competence
within the organization.
To maintain a high level of performance in an IT
solution it is necessary to produce and maintain high
quality Master data. Master data is often created and

implemented at the point of installation. In several of the

organizations studied resulted the lack of motivation from
the organization have to incomplete or incorrect Master
data. If the motivation level of the whole organization is
high then a united leadership and workforce would also
help implement high quality systems with high quality
Master data. This is also true for data upkeep. With a
through data upkeep it is even possible to identify and
correct problems in data collection. As in this case, it is
necessary to have a motivated organization, since if one of
the groups in the organization does not understand the
need for good DQ then they could easily become a
bottleneck in the systems implementation.
Metadata has also an important role in the
management and sharing of knowledge. In this case a
good description of data makes it possible to reduce errors
during the data aggregation process, which will lead to an
increase Metadata quality. Even though, there will always
be some aspects of the Metadata that are inaccurate,
inconsistent or out of date, therefore, it is essential that
quality assurance is built into the system during the
Metadata creation process. Thus, its scope extends beyond
the local context and the resulting Metadata will
positively impact the DQ.
Finally, to be able to keep the organization
updated and motivated, clear goals need to be established.
Without the goals the organization will tend fall back to
their previous habits not support the implementation
properly. It is necessary that all in the organization
understand why things should be implemented, e.g. data
entry, upkeep, analysis, etc. An overarching final goal
should be that the whole organization works together to
solve these issues. As with that an important issue,
cooperation between work teams is really important. To
facilitate this, everyone must understand the overall goals
of what needs to be done and why it must be completed.
In this way maybe even work teams will look for
cooperative solutions.

6. CONCLUSIONS
The purpose of this paper was to explore and
identify the aspects of DQ in eMaintenance. However,
based on the results from the conducted case studies it can
be concluded that effective and efficient maintenance
requires proper information logistics, of which could be
provided through eMaintenance solutions. Although, the
development of eMaintenance solutions faces several
challenges. Some of these challenges can be related to
services aimed for acquisition, processing, analysis and
delivery of content aimed for a target context. Others can
be related to the content, which is provided though the
services. In both cases, ensuring the quality of data is

highly important. Hence, eMaintenance solutions should

provide mechanisms that measure, analyse, manage and
visualise Quality of Service and DQ. The conducted case
studies indicate that DQ can be related to different
features (e.g. usability, accuracy, and relevancy). These
DQ features should to be clearly visualized in order to be
measured and managed by the eMaintenance solution.
This includes user specific helps.
Good quality Master data and Metadata are
necessary for the DQ output, therefore, time and resources
need to be administered by the organization leadership.
Overall, the whole organization needs to be motivated to
maintain high DQ. By stating clear goals and following
them the organizational leadership must take the lead and,
in turn, the whole organization should follow after. DQ in
data upkeep would most likely also improve in suit.
Finally, the result of this work would most likely lead to a
better cooperation between the work teams.
It can be concluded that some generic
contributions, dealing with DQ exist, but there is a need to
develop
quality
assurance
mechanisms
within
eMaintenance. These mechanisms should encompass all
phases of the maintenance process; management of DQ
from different data sources and at various aggregation
levels; the visualisation of DQ for improved usability;
adaptation to the users context and decision processes;
and for criticality management. Furthermore, it is also
important that there is an overarching plan for information
management that covers a systems whole lifecycle.

5.

6.

7.

8.

9.

10.

11.

12.

7. ACKNOWLEDGEMENTS
The authors would like to thank Ville Rauhala,
Kemi-Tornio University of Applied Science, Technology
Kemi, Finland, and Katrin Jonsson, Department of
Informatics, Ume University, Ume, Sweden, for their
support in accomplishing this study.

13.

8. REFERENCES

14.

1. Muller, A., Suhner, M. C. and Iung, B.,

Maintenance alternative integration to prognosis
process engineering Journal of Quality in
Maintenance Engineering, 13(2): 198-211, 2007.
2. Jonsson, K., Holmstrm, J. and Levn, P.,
Organizational dimensions of e-maintenance: a
multi-contextual perspective
International
Journal of Systems Assurance Engineering and
Management, 1(3): 210-218, 2010.
3. Tsang, A. H. C., Strategic dimensions of
maintenance management Journal of Quality in
Maintenance Engineering, 8(1): 7-39, 2002.
4. Sderholm, P., Karim, R., and Candell, O.,
Design of experiment and simulation for

15.

16.

17.

identification of significant e-maintenance

services International Journal of Performability
Engineering, 7(1): 77-90, 2011.
Karim, R., A Service-Oriented Approach to
eMaintenance of Complex Technical Systems
Dissertation, Lule University of Technology,
2008.
EN 13306,
EN 13306:2010, Maintenancemaintenance
terminology,
CEN/TC
319,
European standard, Bruxelles, 2010.
IEC,
IEC
60300-3-14:
Dependability
management - Part 3-14: Application guide Maintenance
and
maintenance
support
International
Electrotechnical
Commission,
Geneva, Switzerland, 2004.
Zurawki, R., The Industrial Information
Technology, CRC Press LLC, Boca Raton, FL,
USA, 2005.
Levrat, E., Iung, B., and Crespo Marquez, A., Emaintenance: review and conceptual framework,
Journal of Production Planning & Control,
19(4):1-22, 2008.
Kajko-Mattsson,
M.,
Karim,
R.,
and
Mirijamdotter, A., Essential Components of eMaintenance,
International Journal of
Performability Engineering 7(6): 555-571, 2011.
Kajko-Mattsson, M., Karim, R., Mirijamdotter, A.,
Fundamentals of the eMaintenance Concept
Paper presented at the eMaintenance 2010, Lule,
Sweden, 2010.
Verma, A. K., Srividya, A, and Ramesh, P. G., A
systems approach to integrated E-maintenance of
large engineering plants International Journal of
Systems Assurance Engineering and Management,
1(3): 239-245, 2010.
Lee ,J., A framework for Web-enabled emaintenance systems Environmentally Conscious
Design and Inverse Manufacturing, Proceedings
EcoDesign
2001:
Second
International
Symposium, 450-459, 2001.
Iung, B., Levrat, E., Marquez, A.C., and Erbe, H.,
Conceptual framework for e-maintenance:
illustrtaion by e-maintenance technologies and
platforms, Annu Rev Cont 33:220-229, 2009.
Bangemann, T., Rebeuf, X., and Reboul, D., et al
PROTEUS - Creating distributed maintenance
systems through an integration platform Journal
of Computers in Industry, 57(6): 539-551, 2006.
Muller, A., Crespo Marquez, A., and Iung, B., On
the concept of eMaintenance: Review and current
research Reliability Engineering and System
Safety, 93(8): 1165-1187, 2008.
Koc, M. and Lee, J., A system framework for
next generation EMaintenance Systems The 2nd
international Symposium on Environmentally

18.

19.

20.

21.

22.

23.
24.

25.

26.

27.

28.

29.

30.

31.

Conscious Design and Inverse Manufacturing,

Tokyo, 11-15 December, 2001.
Parida, A., Phanse, K., and Kumar, U., An
integrated approach to design and development of
Maintenance
system
Proceedings
of
VETOMAC-3 and ACSIM, New Delhi, India, 6-9
December, 2004.
ISO/IEC, ISO/IEC 15288: System engineering System life cycle processes. International
Organization for Standardization, Geneva,
International
Electrotechnical
Commission,
Geneva, Switzerland, 2008.
Parida, A., and Kumar, U., Maintenance
Performance Measurement (MPM): Issues and
Challenges Journal of Quality in Maintenance
Engineering, 12(3): 239-251, 2006.
Lee, J., Ni, J., Djurdjanovic, D., Qiu, H., and Liao,
H.,
Intelligent
prognostics
tools
and
eMaintenance Journal of Computers in Industry,
57(6): 476-489, 2006.
Ryu, K. S., Park, J. S., Park, J. H., A Data
Quality Management Maturity Model ETRI
Journal, 28(2): 191-204, 2006.
Juran, J. M., and Blanton, G., A., Jurans Quality
Handbook. McGraw-Hill, New York, 1999.
Strong, D. M., Lee, Y. W., and Wang, R. Y.,
Data quality in context Communications of the
ACM, 40(5): 103-110, 1997.
Levitin, A., V., and Redman, T., C., Data as a
resource:
Properties,
implications,
and
prescriptions, MIT
Sloan
Management
Review, 40(1): 89-101, 1998.
Leitheiser, R., L., Data Quality in Health Care
Data Warehouse Environments, System Sciences,
Proceedings of the 34th Hawaii International
Conference, USA, 2001.
ISO, ISO/TS 8000-100:2009, Data quality - Part
100: Master data: Exchange of characteristic data:
Overview. International Organization for
Standardization, Geneva, 2009.
Wang, R. Y., Beyond Accuracy: What Data
Quality Means to Data Consumers Journal of
Management Information Systems, 12(4): 5-34,
1996.
Wang, R. Y., Henry, B. K., and Stuart, E. M.,
Data Quality Requirements Analysis and
Modeling , Ninth International Conference of
Data Engineering, April, Vienna, Austria, 1993.
Chapman, A. D., Principles of Data Quality,
Version 1.0. Report for the Globalb Biodiversity
Information
Facility,
Copenhagen.
http://niobioinformatics.in/pdf/workshop/Data%20
Quality.pdf. Accessed 29 May 2012.
Wang, R. Y., Storey, V. C., and Firth, C. P., A
Framework for Analysis of Data Quality

32.

33.

34.

35.

36.

37.

38.

39.
40.

41.

42.

43.

44.

45.

Research, IEEE Transactions on knowledge and

data engineering, 7(4): 623-640, 1995.
Ballou, D. P. and Tayi, G. K., Enhancing Data
Quality in Data Warehouse Environments,
Communications of the ACM, 42(1): 73-78, 1999.
Leitheiser, R. L., Data Quality in Health Care
Data Warehouse Environments, System Sciences,
Proceedings of the 34th Hawaii International
Conference, USA, 2001.
Xu, H., Nord, J. H., Brown, N., and Nord, G. D.
Data quality issues in implementing an ERP,
Industrial Management and Data Systems, 102(1):
47-58, 2002.
Lee, Y. W., Crafting Rules: Context-Reflective
Data Quality Problem Solving, Journal of
Management Information Systems, 20(3): 93-119,
2003.
Li, M., Liu, L., Yin, L., and Zhu, Y., A process
mining
based
approach
to
knowledge
maintenance, Information Systems Frontiers,
13(3): 371-380, 2011.
Connolly, T., Begg, C., and Holowczak, R.,
Business database systems, Pearson Education,
London, 2008.
Karim, R., Kajko-Mattsson, M., and Sderholm,
P., Exploiting SOA within eMaintenance
Proceedings of the 30th International Conference
on Software Engineering (ICSE), the 2nd
international workshop on Systems development in
SOA environments, Liepzig, Germany, 10 May ,
2008a.
Orr, K., Data quality and systems theory,
Communications of the ACM, 41(2): 66-71, 1998.
Even, A., Shankaranarayanan, G., and Berger, P.
D., Evaluating a model for cost-effective data
quality management in a real-world CRM setting
Decision Support Systems, 50(1): 152-163, 2010.
Price, R. and Shanks, G., A semiotic information
quality framework: Development and comparative
analysis, Journal of Information Technology,
20(2): 88-102, 2005.
Wand, Y. and Wang, R. Y., Anchoring data
quality dimensions in ontological foundations,
Communications of the ACM, 39(11): 86-95,
1996.
Labib, A. W., A decision analysis model for
maintenance policy selection using a CMMS
Journal of Quality in Maintenance Engineering,
10(3): 191-202, 2004.
IBM ,
http://www.01.ibm.com/software/tivoli/products/m
aximo-asset-mgmt/. Accessed 29 May 2012.
Melin, U., The ERP System as a Part of an
Organizations Administrative Paradox, 11th
European conference on information systems,

http://60.88.dyn.lse.ac.uk/asp/aspecis/20030095.pd
f. Accessed 29 May 2012.

46. SAP, http://www.sap.com/solutions/businesssuite/plm/index.epx. Accessed 29 May 2012.