Real-Time Fault Monitoring and Predictive Maintenance

This assignment, titled 'Real-Time Fault Monitoring and Predictive Maintenance', delves into the application of predictive maintenance strategies in industrial processes. It covers methods such as real-time fault monitoring (Pouliezos & Stavrakakis, 2013), neural network-based soft sensors for control systems (Rani et al., 2013), degradation data analysis using Wiener processes (Ye et al., 2013), and predictive maintenance strategies from online resources. The assignment aims to equip students with the knowledge to implement predictive maintenance in practical scenarios, enhancing system reliability.

PREDICTIVE MAINTENANCE

Table of Contents
CHAPTER 1: INTRODUCTION....................................................................................................1
Background of the study.............................................................................................................1
Aim .............................................................................................................................................1
Objectives....................................................................................................................................1
Research questions......................................................................................................................2
Scope of the study.......................................................................................................................2
Research methods........................................................................................................................2
Data Analysis..............................................................................................................................3
Resource requirement..................................................................................................................3
CHAPTER 2: LITERATURE REVIEW.........................................................................................4
Introduction.................................................................................................................................4
Literature Review........................................................................................................................4
CHAPTER 3: RESEARCH METHODOLOGY.............................................................................8
Introduction.................................................................................................................................8
Research philosophy...................................................................................................................8
Research design...........................................................................................................................8
Research Approach.....................................................................................................................9
Methods of data collection..........................................................................................................9
Sampling framework...................................................................................................................9
Data Analysis..............................................................................................................................9
Reliability and validity..............................................................................................................10
Ethical consideration.................................................................................................................10
REFERENCES..............................................................................................................................11

CHAPTER 1: INTRODUCTION
Background of the study
Predictive maintenance (PdM) is one of the major tool that is used by organizations to
monitor condition of equipments. This technique identifies occurring problems in machines by
looking upon their symptoms. Main goal of predictive management is to keep frequency of
maintenance performed low so that unplanned breakdown can be minimized. In the modern
competitive era companies have facing huge competition so for sustaining in the market for
longer duration they have to be prepared all time. Poor maintenance strategies can harm the
overall production capacity of entity (Ye and et.al, 2013). PdM is the great tool that can support
entities in maintaining standard and reducing failure risk of equipments and machines. This
technique help in maximizing useful life of machines and avoiding risk of unplanned downtime.
It is one of the beneficial tool for manufacturing firm because by this way they can save their
cost and can enhance their profitability. It analyses the data related to equipment sand predict
when it will require repairing or maintenance. Once problems has been analysed then manager
repair he parts before they needed repairing so that unnecessary issues can be minimized (Fowler
and et.al, 2013).
Present dissertation is based on Finning company which is world's largest caterpillar
dealer. It offers parts, engines, equipments to industries like mining, construction etc (Defining
Preventive & Predictive Maintenance, 2017). It is proudly providing exceptional services to
service users in vast market. Current study will develop understanding about implementation of
predictive maintenance in current maintenance system (Danyluk and Provost, 2014). It will
describe the concept of predictive maintenance in the Finning. Impact of PdM on the Finning
will be illustrated in this dissertation. Furthermore, It will analysis current maintenance system of
organization.
Aim
“ To understand implementation of predictive maintenance in current maintenance system: A
case study of Finning”.
Objectives
 To understand the concept of predictive maintenance (PdM).
 To identify impact of PdM on the Finning.
 To analysis current maintenance system of Finning.
1

 To determine the ways of implementation a predictive maintenance strategy for
improving current maintenance system of the organization.
Research questions
 Explain the concept of predictive maintenance (PdM)?
 How implementation of PdM impact on the Finning.
 What is the current maintenance level of machines and parts in the Finning.
 What are the ways of implementing predictive maintenance strategy for improving
current maintenance system of the organization.
Scope of the study
Present study on implementation of predictive maintenance in current maintenance
system very effective. Now-s-days companies are facing the problems of poor maintenance that
causes unplanned downtimes and harm production capacity of the machines (Del Rincon and
et.al, 2013). This critical situation has become problematic for the big organizations and they are
required to maximize useful life of parts which are at high risk. By addressing their problems
before time companies can enhance life of machines and can make effective control over cost.
PdM can resolve this problem of entities and can support in maximizing useful life of
equipments (Defining Preventive & Predictive Maintenance, 2017).
It will be beneficial for the entities in reducing sudden failure of machines and by this
way life of equipments will be maximized (Nurjono and Lee, 2013). By identifying impact of
predictive maintenance entire industry will be able to enhance life their machines and they will
be able to get high profit. It will support in addressing problem and reduce these issues before
they need repairing so that production or manufacturing do not get stopped at any movement
(Defining Preventive & Predictive Maintenance, 2017).
Research methods
Research methodology is one of the essential part of dissertation that supports in
gathering in-depth information about the subject matter and getting optimistic results related to
subject matter. In the present study individual will take support of descriptive research design.
Through questionnaire individual will be able to describe predictive maintenance system
effectively (Pouliezos and Stavrakakis, 2013). Furthermore, researcher will use interpretivism
philosophy that would support in gathering deep details about research topic. It is qualitative
research thus, scholar will use inductive approach that would be the best suitable for the present
2

investigation. Random sampling will be used in the current study and 10 managers of Finning
company will be chosen randomly as respondents in order to get information about PdM
maintenance system of equilibrium and its impact on the organization (Dennis, 2016).
Through questionnaire individual will gather detail about the topic and will determine the
advantage and disadvantages of this system in the corporation and how implementation of
predictive maintenance in the current maintenance system can be beneficial for the Finning
company (Guerrero and et.al, 2013). Researcher will also take support of secondary data
collection sources and through books, journals, internet articles individual will develop better
understanding about the research topic (Predictive maintenance and the smart factory?, 2017).
Data Analysis
Data Analysis can be considered as effective tool that support the scholar in getting
optimistic results. Use of appropriate data analysis tools are depended upon the type of
investigation. If its is quantitative then individual has to include facts and figures and can use
SPSS, T-test etc. in order to get valid results (Kylili and et.al, 2014). On other hand in the
qualitative study researcher uses thematic analysis tool for getting outcome. Present investigation
is qualitative thus investigator will use thematic analysis tool and through graphs individual will
present the results. It will develop better understanding about implementation of PdM and
effectiveness of predictive maintenance system in the organization (Defining Preventive &
Predictive Maintenance, 2017).
Resource requirement
For completing the dissertation successfully and accomplishing objective of the
investigation, researcher will require several sources. One of the major source that is essential for
the study is availability of sufficient financial resources. Individual will require monitory support
so that person can use advanced technologies and can test benefits of PdM technique in the
organization (Sachs and Clevers, 2014). Furthermore, scholar will also require human resource
so that relevant information can be gathered about the subject matter. Apart from this, researcher
will also require physical resources so that individual can test the effectiveness of predictive
maintenance system. Time is one of the most important resource without it individual can not
complete its dissertation. Investigator will need sufficient time so that person can investigate
about PdM and can know its impact of business units (Guillot and Rousset, 2013).
3

CHAPTER 2: LITERATURE REVIEW
Introduction
It is second chapter of dissertation in which scholar has to review literatures of other
authors. It is one of the essential part of the study through which individual can develop better
understanding about the subject matter (Rani, Singh and Gupta, 2013). Critical review supports
in gathering relevant information about topic and getting in-depth knowledge about the study. In
this chapter researcher will look upon the literature of other researchers on PdM system
(Predictive maintenance and technologies, 2017).
Literature Review
Concept of predictive maintenance
According to Li and et.al, (2015) predictive maintenance is the effective tool in which
companies predict the failure that might occur in any equipment and then it prevents the machine
from occurrence of such failure so that it can perform with consistency (Defining Preventive &
Predictive Maintenance, 2017). . Author has stated that every tool or equipment has specific time
duration after that period machine requires repairing. If companies do not repair them on time
then it will cause problem suddenly. By this, entities will have to stop their manufacturing for
certain time duration and that will cause financial harm to the corporation. As per the view of
Parida and et.al, (2015) mining industry, agriculture field, plants regularly use many tools in their
workplace in order to produce quality products for consumers. If these equipments get failed in
middle of production then organization will not be able to produce quality goods on time. That
would decrease satisfaction level of consumers and will harm its brand image as well. As per the
view of researcher for keeping machines working all the time entities are required to use
predictive maintenance strategy so that they can maintain the parts and can keep them working
every time. It is an effective way of producing quality products to its consumers and enhance
their satisfaction level.
Jasiulewicz-Kaczmarek, (2013) has stated that traditionally there were not that much
advanced technologies that can identify problem in the equipments but recently there are many
advanced technologies that are capable and inexpensive enough in order to maintain machines
effectively. As per the view of Sliva and et.al., (2015) In the modern business world computer
power, bandwidth has helped the organizations in making PdM as viable option for the entities
through which they can maintain their company parts and can run the operations smoothly.
4

According to Vogl, Weiss and Helu, (2016) predictive maintenance is completely differed from
the preventive maintenance strategy, in this method company can relies actual condition of the
available equipments and analysis the problem that can cause in coming duration (Defining
Preventive & Predictive Maintenance, 2017). By this way upcoming problems of issues of
failure can be come out in front of organization and authorities can take immediate action to
resolve these occurring problems. Maintaining the equipments before it requires maintenance is
the cost effective way that helps in increasing life of machines and by this way firms will be able
to produce quality products. As per the view of scholar predictive maintenance is suitable for big
or large sized organizations because they can maintain and afford such type of expenses. But if
any small size company use this method to reduce errors in the machines then it would be
negative and will harm its financial position to great extent. Researcher has said that it is costly
process so only those firms which has strong financial resources can use this technique.
Li and et.al, (2015) has argued that analysis of data is one of the crucial function for the
management team of the corporation. They need capable employees, highly advanced
machineries those which can identify problems in the parts and can resolve them on time. It can
support in minimizing the future problems due to this company's brand image can get affected.
PdM is the method in which company identifies the problems and through networks relevant
information get circulated to related persons. As per the view of Benkedjouh and et.al,
(2013Analytics tools predict which part of machine is going to fail in near future. This
information then circulated to workers. After that they take immediate action to repair the
equipments so that sudden failure situation can be avoided in the company. It helps the
organization in maintaining parts, machines, equipments fit all the time so that it can help in
producing quality products for the consumers (Defining Preventive & Predictive Maintenance,
2017).
Impact of PdM
Finning is the largest organization and caterpillar dealer that delivers unrivalled services
to the mining, construction, forestry industries. According to Kylili and et.al, (2014) digital and
physical technologies impact on the business units to great extent. It mainly affects two major
objective of entities; these are business operations and business growth. Author has stated that
PdM has become one of the best options for the organizations those which are operating in the
same industry. Predictive maintenance impact positivity on the business growth of the firm. With
5

the help of this tool firms can avoid sudden failure and can examine the problems that can occur
in coming future (Defining Preventive & Predictive Maintenance, 2017). By this way they will
be able to take immediate action to maintain such machines so that operations can not be stopped
in middle of the period and they can deliver quality products to consumers. By maintaining the
equipments on time companies can resolve unnecessary expenses on machines. On other hand
researcher has viewed that for delivering quality products to consumers company can do more
research and can use marketing strategies. Maintenance of machines only help entity when
sudden problems in operations get arisen otherwise it is not that much effective in improving
quality of products.
On other hand Ye and et.al, (2013) has argued that PdM helps in running business
operations in smooth manner. It is the tool that can help in increasing life of these equipments
and by this way operations will not get stopped at any movements (Defining Preventive &
Predictive Maintenance, 2017). It repairs the machines before it needed repairing that helps in
keeping them alive and fit so that through these parts firms can produce quality standards
products and can satisfy needs of target consumers. As per the view of Danyluk and Provost,
(2014) corporations spend huge time in machine inspections so that trouble shooting unplanned
downtime issues can be minimized. But traditionally companies inspect the equipments once in a
year. Sometimes some parts stop working before inspection gets finished. To avoid such type of
condition in the organization, now-a-days firms are using PdM methods in which they identify
cause that can occur in near future and before it requires maintenance, firm repair machines so
that company can save its time and cost (Defining Preventive & Predictive Maintenance, 2017).
By this way entities are able to minimize such type of unplanned downtime that enhances
operational efficiency of the firms. It is positive impact of predictive maintenance system.
On other hand as per the view of researcher predictive maintenance system is not that
much effective because it consumes too much time and due to this production gets delay. It is
essential to monitor the machines time to time so that their issues can be identified on time.
Scholar has argued that predictive maintenance requires lots of money to invest that may harm
the financial position of the firm to great extent. As per the view of Parida and et.al, (2015)
connected technologies of predictive maintenance system can pull data and legacy helps in
providing insights information about tools and equipments. By this way managers can deploy
6

their resources and can utilize them more effectively (Defining Preventive & Predictive
Maintenance, 2017).
Current maintenance system
According to Guerrero and et.al, (2013) manufacturing industry maintains first level and
then go to reactive, it tries to reach to predictive level as soon as possible. In this industry there is
separate technical department presents which keep monitor activities and performance of all
machines so that unplanned failure can be minimized. It is the main function of the firm that to
provide working and fit equipments or products to the end users. This separate department has to
schedule necessary services in the organizations so that it can reduce chances of failure in the
business units. As per the view of Li and et.al, (2015) firms that operate in the manufacturing
industry has fluid laboratory in which skilled people check condition of oil and particles. It helps
in identifying problems in the machines so that its maintenance can be done before critical
failure occur. On other hand researcher has argued that predictive maintenance system helps in
improving life of equipments but time to time checking is too much time consuming process.
Apart from this due to this, company fails to manufacture the products on time that cause delay
in operations. That may affect overall performance of the organization to great extent.
According to Vogl, Weiss and Helu, (2016) manufacturing are working efficiently and its
maintenance system is too effective. Caterpillar products link system, service data, valuable
information helps in running the operations without any failure. By this way firms are able to
maintain its machines all the time. As per the view of Martinelli, (2016) Predictive maintenance
system in helping these entities in getting overview of the equipments and managers take actions
to repair it so that further issues can be minimized. Li and et.al, (2015) has argued that
companies are giving training to its staff members so that they can measure working of each
parts and equipments and can communicate to management before timing if any complications
are going to arise in near future. According to Sorigue and et.al., (2015) Most of the
manufacturing companies are taking support of finsight system that provides detail about the
latest condition of the machines and parts or related equipments with manufacturing (Defining
Preventive & Predictive Maintenance, 2017). They monitor the performance of these parts time
to time and allow in developing better understanding about how these assets of the organization
are currently working, whether there is any fault or any issues that can be taken place in coming
period.
7

Predictive maintenance strategy
According to Kylili and et.al, (2014) implementation of predictive maintenance system is
one of the effective through which entities are able to generate high revenues and getting success
in this industry (Predictive maintenance and the smart factory?, 2017). Researcher has stated
that designing PdM program is the effective strategy that can help in minimizing failure of the
equipments. But they need to use implementation in effective manner so that cost of the
operations do not get increased and company can run its operations smoothly. By this way
companies can look failure history and can find out root cause of issues in particular machines of
part. The equipment which is failing has potential for reliable improvement (Defining Preventive
& Predictive Maintenance, 2017). It can help in reducing cost of the entities and can help in
increasing profitability of the organizations.
As per the view of Guillot and Rousset, (2013) selection of appropriate technology is the
only effective strategy that can make the predictive maintenance system more successful. Use of
advanced technologies can support in monitoring performance of each machine. By this way
management team will get to know which part is required repairing after certain time duration.
Accordingly firms can take necessary action to repair it before time duration so that operations
do not get harmed. Researcher has viewed that use of advanced technology is beneficial but it
increases cost of the entities too. Due to this they face trouble in running their business smoothly.
In predictive maintenance they have to change the equipments before it requires repairing that is
very costly process. Apart from this if company fails to get revenues then this maintenance will
give loss to the organization.
8

CHAPTER 3: RESEARCH METHODOLOGY
Introduction
Research methodology is the effective way through which scholar can find out optimistic
results of specific issues. There are several methods that are included in the research
methodologies that support the investigator in gathering in-depth information about the subject
matter so that objective of the investigation can be accomplished. In this chapter scholar will
described several methods that have been used by the researcher in order to find out the results
about predictive maintenance system (About, 2017).
Research philosophy
It is one of the essential part of research that defines the scholar the way of gathering the
information about the subject matter. It is the way through which scholar can develop
individual's knowledge and can conduct the investigation in effective manner. There are several
techniques that are used in dissertation and in research project such as positivism, interpretivism.
In the present study on predictive maintenance system scholar has used the interpretiovism
philosophy (What is predictive maintenance?, 2017). Individual has focused on subjective part
and has taken support of theories and models in order to develop understanding about the subject
matter. The main reason of using this tool is conduct the investigation on the bases of research
questions. Findings are much more valid and reliable that has helped the investigator in finding
the solutions of research problems.
Research design
It can be defined as blue print of data, it is the detailed outline and guide individual
regarding how investigation is required to be completed. It defines the general plan that how the
investigation is going to be done. There are several types of tools that are used in this section;
descriptive, exploratory etc (Defining Preventive & Predictive Maintenance, 2017). In the
present study scholar has taken support of descriptive research design. It was the best suitable
tool for the current investigation. With the help of this method scholar has defined the
characteristics of predictive maintenance system effectively and has gathered information about
the subject matter. This technique has helped in highlighting the issues in respect to maintenance
of machines and equipment sand finding proper solutions of these problems (Li and et.al, 2015).
9

Research Approach
It is another part of research methodology in which researcher carry out the investigation
in specific manner. It identified the tool of data collection so that in-depth information regarding
subject matter can be gathered and valid results can come out. Inductive and deductive are two
best approaches that are used in the dissertation. In the present investigation scholar has used the
inductive research approach (Sachs and Clevers, 2014). Through this study individual has
developed theory and has found results. That was the best approach through which researcher
has developed link between predictive maintenance system and company growth. With the help
of this method individual has gathered relevant information and has find out the optimistic
results (Vogl, Weiss and Helu, 2016).
Methods of data collection
Data collection is one of the most important part of research methodology, that supports
in developing knowledge about the research topic. There are primary and secondary sources
through which data can be collected regarding the subject matter. In the present research scholar
has taken support of questionnaire and has gathered first hand information about predictive
maintenance system (Guillot and Rousset, 2013). Apart from this, scholar has also taken support
of secondary sources. It has helped in gathering the already available information. That has
supported in getting optimist results about impact of PdM on companies performances.
Sampling framework
Sampling framework is another method that supports in selecting the correct sample out
of large population. In the present investigation scholar has used random sampling technique and
has selected 10 managers of Finning company randomly )Sachs and Clevers, 2014(. Individual
has asked questions with them regarding predictive maintenance system and current maintenance
system of the organization. Managers have sufficient information about the company and nits
operational performances. So they can give appropriate information about the subject matter.
That is why researcher has chosen them as respondents in order to develop knowledge about the
PdM.
Data Analysis
It is the most important part of research methodology. It is essential to analysis all
collected information in order to reach to the final co0nclusion. It depends upon the type of
investigation whether it is qualitative or quantitation (Jasiulewicz-Kaczmarek, 2013). In the
10

quantitative study individual uses SPSS, T-tests. Whereas, in the qualitative persons goes with
thematic analysis. Present dissertation is on the PdM which is qualitative study thus, scholar has
used thematic analysis tool for data analysis.
Reliability and validity
Data that has been collected by the scholar in the present investigation are reliable and
valid. Individual has not manipulated any information. This thesis can help other authors in
conducting the further investigation on the same subject (Sachs and Clevers, 2014).
Ethical consideration
Ethical consideration can be defined as effective approach that helps in identifying
difference between right and wrong. For preparing the valid dissertation it is essential to ensure
ethics. Data that are used by researcher are taken from authorized website, there is not any single
information that has been taken from unauthorized site. Individual has not commercially used
this information, all details and results are used only in this dissertation (About, 2017). Data has
been keep protected. In addition, researcher has copied any detail from anywhere, individual has
read the articles that are presented on websites and has developed understanding about the
subject matter. Individual has written all the detail in his own words so there would not be any
issue of plagiarism (Defining Preventive & Predictive Maintenance, 2017).
11

REFERENCES
Books and Journals
Benkedjouh, T. and et.al., 2013. Remaining useful life estimation based on nonlinear feature
reduction and support vector regression. Engineering Applications of Artificial Intelligence.
26(7). pp.1751-1760.
Danyluk, A. and Provost, F., 2014, May. Small disjuncts in action: learning to diagnose errors in
the local loop of the telephone network. In Proc. of Tenth International Conference on
Machine Learning (pp. 81-88).
Del Rincon, A. F. and et.al., 2013. A model for the study of meshing stiffness in spur gear
transmissions. Mechanism and Machine Theory. 61. pp.30-58.
Dennis, P., 2016. Lean Production simplified: A plain-language guide to the world's most
powerful production system. Crc press.
Fowler, V.R. and et.al., 2013, October. Energy requirements for the growing pig. In Energy
metabolism. Proceedings of the 8th symposium. European association of animal
production, publication (No. 26, pp. 151-156).
Guerrero, J. M. and et.al., 2013. Advanced control architectures for intelligent microgrids—Part
I: Decentralized and hierarchical control. IEEE Transactions on Industrial
Electronics. 60(4). pp.1254-1262.
Guillot, G. and Rousset, F., 2013. Dismantling the Mantel tests. Methods in Ecology and
Evolution. 4(4). pp.336-344.
Jasiulewicz-Kaczmarek, M., 2013. Sustainability: orientation in maintenance management—
theoretical background. In EcoProduction and Logistics (pp. 117-134). Springer Berlin
Heidelberg.
Kylili, A. and et.al., 2014. Infrared thermography (IRT) applications for building diagnostics: A
review. Applied Energy. 134. pp.531-549.
Li, J. and et.al., 2015. Big Data in product lifecycle management. International Journal of
Advanced Manufacturing Technology. 81.
12

Martinelli, S., 2016, November. Smart Actuators: A Predictive Maintenance Strategy to Achieve
Cost-Saving Targets. In Abu Dhabi International Petroleum Exhibition & Conference.
Society of Petroleum Engineers.
Nurjono, M. and Lee, J., 2013. Predictive utility of blood pressure, waist circumference and body
mass index for metabolic syndrome in patients with schizophrenia in Singapore. Early
intervention in psychiatry. 7(2). pp.205-209.
Parida, A. and et.al., 2015. Performance measurement and management for maintenance: a
literature review. Journal of Quality in Maintenance Engineering. 21(1). pp.2-33.
Pouliezos, A. and Stavrakakis, G. S., 2013. Real time fault monitoring of industrial
processes (Vol. 12). Springer Science & Business Media.
Rani, A., Singh, V. and Gupta, J. R. P., 2013. Development of soft sensor for neural network
based control of distillation column. ISA transactions. 52(3). pp.438-449.
Sachs, N. and Clevers, H., 2014. Organoid cultures for the analysis of cancer phenotypes. Current
opinion in genetics & development. 24. pp.68-73.
Sliva, A. and et.al., 2015. Tools for validating causal and predictive claims in social science
models. Procedia Manufacturing. 3. pp.3925-3932.
Sorigue, M. and et.al., 2015. Prevalence, Predictive Factors Therapy and Outcome of Patients
with Follicular Lymphoma Refractory to First Line Immunochemotherapy.
Vogl, G. W., Weiss, B. A. and Helu, M., 2016. A review of diagnostic and prognostic capabilities
and best practices for manufacturing. Journal of Intelligent Manufacturing. pp.1-17.
Ye, Z. S. and et.al., 2013. Degradation data analysis using Wiener processes with measurement
errors. IEEE Transactions on Reliability. 62(4). pp.772-780.
Online
About, 2017. [Online] Available through:
<http://www.finning.com/en_GB/company/about.html>. [Accessed on 4th July 2017].
Defining Preventive & Predictive Maintenance, 2017. [Online] Available through:
<http://www.danielpenn.com/insights-resources/case-studies/preventive-predictive-
maintenance/>. [Accessed on 4th July 2017].
13

Predictive maintenance and technologies, 2017. [PDF] Available through:
<https://energy.gov/sites/prod/files/2013/10/f4/OM_6.pdf>. [Accessed on 4th July 2017].
Predictive maintenance and the smart factory?, 2017. [PDF] Available through:
<https://www2.deloitte.com/content/dam/Deloitte/us/Documents/process-and-operations/
us-cons-predictive-maintenance.pdf>. [Accessed on 4th July 2017].
What is predictive maintenance?, 2017. [Online] Available through:
<https://www.fiixsoftware.com/maintenance-strategies/predictive-maintenance/>.
[Accessed on 4th July 2017].
14