1

Introduction to engineering research

1.1 Why engineering research?

The disciplines of engineering are all described as the application of science to realistic systems
which benefit humankind [1]. Engineering research is therefore based on the principles of scientific
research which, in turn, are based on the scientific method, in which observations (experiments),
theories, calculations and models are derived from the existing body of scientific knowledge and
verified independently by others who are experts in the field [2–4]. This latter process is called ‘peer
review’. While this formal review by peers is not foolproof, it constitutes the best method of
validation and verification of research results. Engineering research is based on precisely the same
scientific method; however, the research is directed toward the practical application of science to
products, services and infrastructure.
Most research starts with a hypothesis; that is, a statement which can be either proved or disproved.
In most cases it is easier to disprove a hypothesis because only one counter example is required to
discredit the idea. To prove a hypothesis, it is necessary to exhaustively examine every possible case
and make sure the hypothesis applies. Often this results in the creation of limiting conditions. The
conclusion becomes slightly modified in that the hypothesis is valid providing certain conditions are
met. A full evaluation of a hypothesis may take many years without a conclusive resolution.

Example 1.1 Hypothesis statements

‘All mechanical systems can be described by damped simple harmonic motion equations.’ You
could test many mechanical systems and find that this is true. However, if you find one example
where this is not true, then the hypothesis fails. In this case, it is necessary to apply some limits
to the statement.
‘The maximum efficiency of a solar cell is 28%.’ If you find one example of a solar cell
which has a higher efficiency, then the hypothesis fails.
‘The laws of physics apply throughout the universe.’ Physicists and astronomers continue to
assume this is true when describing the formation of stars.

The history and philosophy of science encapsulates the scientific method and the creation of new
knowledge [2–4] based on a new theory which has been subsequently verified by observation,
experimentation and the logical development from previously accepted theories, but this is not the
subject of this book. In some cases theoretical concepts are deduced long before experimental
verification. In other cases, theoretical concepts are deduced from experimental observations. The
history of science is full of examples of both.
There are many books which discuss scientific research and its methods [5–8]: so how does
engineering research differ from research in science? A preliminary answer is to require engineering
research activities to fulfill all of the following objectives:

The research must be applied to human systems;

The research must yield practical outcomes;
 The research must yield outcomes which benefit humanity;
The research must be ethically based;
The research should consider environmental outcomes;
The research must be based on standard industry based testing.

A more detailed explanation of these issues is provided in the following chapters.

Example 1.2 Research opportunities

A new material has been proved scientifically to be a reliable replacement for asphalt and
concrete for road building. The material has excellent physical and chemical properties. An
engineering research study of this material might also verify that the material is in abundant
supply from renewable resources, the material itself can be recycled at the end of its usefulness
in road material, and the toxicity of the material does not adversely impact the environment.
A new transistor technology is based on a rare-earth metal which has extremely low
abundance, is difficult to obtain and difficult to recover from e-waste. The research in this field
constitutes esoteric science rather than engineering research because the outcomes are unlikely
to be adopted widely unless improved environmental outcomes can be assured.
Air transport using hydrogen balloons requires very little energy to raise the load and
return the load to the ground. It was found to be not practical because of the flammability of the
gas, its confinement/storage is difficult and the speed of movement is highly limited.

Clearly scientific research and engineering research are not mutually exclusive. All medical
science is directly related to improving the human condition through medical practice. Other human
related fields, for example physiology, pharmacy, dentistry, psychology, education, etc also have
some outcomes relevant to improving the human condition. Similarly many engineers engage in
purely scientific research to test ideas with the long term aim of finding solutions to the practical
implementation problems associated with the research outcomes. As there is no clear dividing line
between these fields, many conferences and scientific journals report both scientific and engineering
based research. This book is confined to engineering based research strategies, but the concepts are
also very applicable to purely scientific research. Thus, further reading is readily available from
scientific research methods books and papers [7–9].
Engineers, and hence engineering research, are constrained by quite formal codes of ethics. Every
discipline of engineering has a code of ethics covering engineering practice in one or more different
countries. The codes should influence how the research is conducted and ensure that the outcomes are
improvements to humankind through improved environmental outcomes and minimal risk to users of
the technology. In particular the concept of economical engineering design must be balanced with
aspects of fail-safe and an understanding of materials and product reliability. In many cases these
aspects are inherent in the research design, but good engineering research outcomes will require
independent verification of not only the research itself, but also the applicability of systems designed
from these research outcomes. Codes of ethics and their importance are discussed in Section 1.4.
A research project is not complete until the results have been presented publicly for other experts
in the field to comment and review. Thus publication of findings in the open, refereed, international
literature and/or presentation at a meeting of research peers is an essential requirement of any
research project. Only when the research outcomes have been reviewed by suitably qualified peers
can the researchers declare that new knowledge has been created. This means that work conducted in
secret (for example in a military research facility, in a high security research laboratory, or in other
private venues), does not contribute to the world-wide body of knowledge, and therefore cannot be
described as research.

Example 1.3 Unsubstantiated claims

Claims that top secret research by the US Government had revealed the existence of
unidentified flying objects and the landing of extraterrestrial creatures have never been
subjected to international scrutiny and so must not be regarded as contributing to new
knowledge.
Some claims of aromatherapists, chiropractors, water diviners, etc have never been
substantiated by rigorous scientific examination and so do not contribute to new scientific or
engineering knowledge.

Similarly, a search of previous publications and patents does not constitute research. Thus, when a
primary school child conducts ‘research’ on the Great Wall of China by copying the outcomes
located using a computer search engine, this does not constitute rigorous scientific or engineering
research. This student is gathering well established and previously reported information. This is an
important distinction: research outcomes which are new to the researcher but are well known to
others does not constitute original, publishable research. As a logical consequence of this argument,
any original research must clearly identify all relevant prior work before the authors can claim to
have developed new knowledge. This can be a significant challenge as the volume of published works
continues to grow at an accelerating rate.

1.2 Next step research

All research is built on the background and understanding of science developed over the centuries.
When a person plans to engage in a research project to create new knowledge, it is vital that a recent
and thorough understanding of the field is gained before designing the research project. A new
research project will be built on the work of others, from Newton and Maxwell to Mohr and Edison.
In addition, the research strategy and methods applied should be well regarded by the world-wide
community of scholars.
In order to emphasise this concept, it is possible to describe two approaches to research: (a) A new
fundamental innovation that changes the way we think about the world in scientific terms. This can be
described as a paradigm shift. (b) A step forward in our understanding of the engineering world
based on one or more of the following ideas:

The application of techniques commonly used in one field to another field;

The modification of an existing concept or technique with improved outcomes;
The modification of current technologies for improved efficiency, miniaturization, sustainability
or environmental outcomes.
 Example 1.4 Translational research opportunities
Mechanical engineers used the finite element method for many years before the
electromagnetic equations were solved numerically using the method. This resulted in a new
field of computational electromagnetics in the 1980s.
Image analysis techniques used for face recognition and satellite based vegetation
categorization can be applied to two and three-dimensional data sets in any field of
engineering.
Inertial sensors used in the automotive industry as air-bag triggering devices are now used
in sports engineering for movement analysis and in mechanical engineering for vibration
analysis.

1.3 Research questions

A common method of focusing on a research project is to phrase a research question. The design of a
single, succinct question is a challenge for all researchers and the research team may consider several
iterations before it is accepted. The research question will directly lead to one or more methods of
investigation, and these can be divided into a number of research aims.
The research question can be phrased using one of the following questioning words:
Why?

Example 1.5 Research question ‘why?’

Why did the wind turbine fail in 100 kph winds?
This question suggests a number of avenues of investigation. For example the researchers
might:

– review the literature for previous failure reports,

– assess the wind conditions at the time of failure,
– undertake numerical modelling experiments,
– review fatigue and possible points of weakness,
– conduct inspections of other wind turbines located in the area.

What?

Example 1.6 Research question ‘what?’

What is the effect on the strength of concrete when recycled concrete is used in the mix?
This question suggests a number of avenues of investigation. For example the researchers
might:
 – review the literature seeking results from previous trials,
– conduct compression and shear experiments using different mixtures of concrete,
– conduct strength calculations based on aggregate strength theory.

How?

Example 1.7 Research question ‘how?’

How can the braking system of a railway carriage be self-activating when its velocity exceeds a
threshold value?
This question suggests a number of avenues of investigation. For example the researchers
might:

– review the literature and patents for automatic braking systems,

– calculate the braking power required,
– conduct model-based experiments on braking systems.

When?

Example 1.8 Research question ‘when?’

When will the roof bolts in an underground tunnel fail through environmental degradation?
This question suggests a number of avenues of investigation. For example the researchers
might:

– review the literature for previous studies in different rock types and environmental
conditions,
– review the types of roof bolts in common use,
– conduct a survey of rock bolts in different tunnel environments to assess degradation,
– conduct experiments to measure the degradation of the roof bolts under accelerated
environmental conditions.

These examples might suggest some of the work that has been previously reported. A review of
the literature will mean that the research team does not have to ‘re-invent the wheel’, and can build
their research on the published reports of others. It will also suggest that even if the same problem has
not been solved previously, the methods used to solve similar problems might be appropriate to solve
their specific problem. A good literature review can impact positively on the research methods that
the research team might use. This is of significant value as the use of previously reported and peer
reviewed methods adds confidence about the reliability of the research method and the subsequent
independent review of the journal and conference papers arising from the research.
Novice researchers should also note that a number of different methods of approach are suggested
for each research question. It is mandatory that more than one method of investigation is used in all
research projects in the hope that the results from a number of different approaches can be used to
substantiate the conclusions from the project. This adds confidence in the research outcomes.

1.4 Engineering ethics

As engineers are involved in all major infrastructure projects (dams, bridges, roads, railway lines,
electricity distribution, telecommunications, vehicles, etc), it is not surprising that when one of these
facilities fails, caused either by natural events (earthquakes, adverse weather, landslides, material
failure, etc) or by human intervention (terrorist attack, stadium overloading, land contamination,
water contamination, air pollution, lack of maintenance of facilities, land subsidence due to mining,
etc), inevitably some of the blame and responsibility is levelled at the engineers who undertook the
design, construction, maintenance and control of the facilities. Over the past 200 years, there have
been engineering failures that have resulted in loss of human life, medical problems, the extinction of
species, damage to the environment, and damage to the economic wellbeing of towns and entire
countries. As a consequence of these disasters, the professionals engaged in engineering activities
have created and joined professional societies. These societies are designed to minimize the
likelihood of repeat occurrences through member registration and mandated self-improvement. This
is achieved through two methods; firstly by the free exchange of information between practitioners,
and secondly, through an adherence to codes of ethics which are designed to eliminate poor practice
and, in extreme cases, to prevent negligent and incompetent people from placing the community at
risk by working on such projects. These professional engineering societies therefore engage in
processes of accreditation of individuals and university degree programmes, and the publication of
research, engineering standards (best practice) and failure analyses.

Example 1.9 Engineering disasters

The mining of phosphate on small islands has resulted in untenable farmland, local climate
change and water shortage for the inhabitants.
The draining of a lake in Centre Asia resulted in the destruction of the environment and
subsequently village life.
The incorrect fitting of the fuel tank seals in the Challenger space shuttle resulted in the
vehicle exploding during launch.
The meltdown of the nuclear reactor at Chernobyl killed many people, caused significant
radiation damage to those who survived, and caused the local area to be contaminated for the
next several hundred years.

Engineering societies require that practising engineers take a holistic view of projects which
deliver a complete solution providing maximum benefit to all stakeholders (e.g. the community, the
users of the product or service, their company or organization commissioned to undertake the work,
and the environment both local and international).
In most countries professional engineers are registered. It may be mandated in law that only a
registered engineer is allowed to build a bridge or sign off on a control circuit used in a theme park
ride. Without the signature of a registered engineer, the work is not permitted to start, or not
permitted to operate.
The registration process requires the person to be suitably trained (a university degree in
engineering is most common) and must pledge to work within the guidelines of a Code of Ethics.
Should engineers not conduct themselves appropriately, they can be removed from the register of
professional engineers and barred from continuing to practise. Most engineering codes of ethics also
prohibit engineers from practising outside their engineering discipline.
While most engineering codes of ethics differ in the fine detail, there are some tenets (or
statements) which are common to almost all of the engineering codes [8].
Importantly, professional engineers are required to act in the best interests of humanity and the
community. This requirement stands above all else and overrides engineers' responsibilities to:

Their employers;
Their family, friends and relatives;
Their town, region and country.

From this basic tenet, other tenets are derived which relate to:

The environment (including sustainability);

Public safety (including fail-safe design);
The economic viability of their employer (cost-effective, reliable outcomes).

The adherence to a code of ethics maintains the reputation of the engineering profession. For
example the Institution of Engineers Australia code of ethics begins with the words:

‘As engineering practitioners, we use our knowledge and skills for the benefit of the community
to create engineering solutions for a sustainable future. In doing so, we strive to serve the
community ahead of other personal or sectional interests.’ [9]

The American Society of Civil Engineers (ASCE) use the following fundamental principles [10]:

‘Engineers uphold and advance the integrity, honor and dignity of human welfare and the
environment by:

1 Using their best knowledge and skill for the enhancement of human welfare and the
environment;
2 Being honest and impartial and serving with fidelity the public, their employers and clients;
3 Striving to increase the competence and prestige of the engineering profession; and
4 Supporting the professional and technical societies of their discipline.’

And the first ASCE fundamental cannon is:

‘Engineers shall hold paramount the safety, health and welfare of the public and shall strive to
comply with the principles of sustainable development in the performance of their professional
duties.’
Many professional engineering organizations are now international, and this allows engineers trained
in one country to operate without impediment in another country if both subscribe to a common code
of ethics. The largest international engineering agreement is the Washington Accord [11], where
signatory nations must allow regular inspection of the academic qualifications of their member
nations and adherence to a common code of ethics.
The codes of ethics commonly restrict the unauthorised copying of designs and other intellectual
property. For this reason, engineering researchers must acknowledge the work of others in the
development of their findings.
When undertaking engineering research, the research team must understand the implications and
restrictions which apply to their research projects based on their code of ethics. There is also an
obligation that researchers must undertake their work to the best of their ability and in line with their
training, within the standards defined by their discipline, to use appropriate terms to describe their
work, and to provide unbiased reports on the success of their work [8, 12]. It is most important that
the conclusions of a research project not only summarise the outcomes in a positive light, but probe
the uncertainties and problems which might occur if the research results are applied to products and
services. These matters will be discussed in some more detail in later chapters of this book.
When the research involves the use of humans or animals, research funding bodies and the
publishers of research outputs (journals and conference technical committees) require that the project
plan be assessed for impacts on the human and animal subjects before the project begins. Such
research projects include surveys (the participants are asked to respond to a set of questions), physical
activities (the participants are asked to perform manual tasks) and mental activities (participants are
asked to solve puzzles). Asking individuals to participate in a research project either voluntarily or
for a reward or other benefit, can have significant negative consequences on the volunteers and the
research outcomes.

Example 1.10 A biased survey

A company decides to ask its employees for feedback on their latest product, which was
designed in-house. Some employees feel obliged to support the product concept because they
do not wish to offend the design team or the general manager. The net result is a biased
outcome of the survey and potential damage to the careers of those employees who did not
respond favourably.

In order to avoid the problems of biased research outcomes and perceived or real threats to the
participants, an ethical approval procedure must be available to ensure that negative outcomes to the
research and the research participants are avoided.
This requires the following information to be presented to the target group in an understandable
way:

The task is voluntary;

The participants are provided with information about the tasks and risks before commencing;
The participants are provided with contact details if they have additional questions or feel
aggrieved;
The participants can withdraw at any time during the survey without recrimination;
 The survey responses are anonymous and the data are stored without the identification of
individual participants;
The accumulated results are provided to the participants when the final analysis has been
completed.

These requirements can be achieved in surveys using a number of different methods of approach:

A common method is to have the work (for example surveys can be administered by an
independent organization) conducted by an independent third party. This party will solicit the
volunteers, collect the responses and ensure that the company seeking the data will not be
provided with identification tags for individuals.
There are now options for online surveys which are effectively anonymous.
The individual profiles used in the survey are not sufficiently specific so that each individual
cannot be easily identified.

The design and conduct of surveys are discussed in more detail in Chapter 6.

Example 1.11 Individual identification in surveys

A participant profile which sought precise values for age, height and weight could easily result
in the identification of an individual. A participant profile which sought responses to broad
ranges of age, height and weight would reduce the chances of individual identification.

1.5 What constitutes conclusive proof?

Research is designed to create new knowledge. This new knowledge needs to be substantiated
appropriately, initially by the research team, and subsequently through the peer review process by the
world-wide community of scholars who are experts in this field. Thus novice researchers must
convince themselves that their results and conclusions are valid and supported by strong evidence.
This is commonly done using more than one of the following techniques:

Experimental measurement (particularly using standard tests);

Theoretical development;
Logic and mathematics;
Numerical simulation;
Statistical analysis;
Comparison with previously published research outcomes;
The use of multi-parameter optimization methods to obtain the best outcome of the design.

Ideally a research team would engage in many, if not all, of these techniques to verify its results.
This gives the independent reviewers of the work the best possible proof that the new knowledge has
been validated and the conclusions are correct.
Often, however, there may be problems in using this approach, for instance:

There are no standard experimental methods applicable to a particular research investigation;

 Repeated experimental measurements are not possible so that statistical support is difficult;
The theoretical frameworks available are too simplistic to describe the outcomes;
There is no prior work which closely resembles this investigation; and
Multi-parameter optimization is not possible as computational models are ineffective.

These difficulties can all result in a probability of failure or project risk assessment. This is
usually expressed as a probability.
The publication of research results mandates that some supporting evidence is provided. A failure
to do this will result in a failure to have the research outcomes accepted. It is therefore very important
that the research team design a project in a manner which ensures that sufficient supporting evidence
is available at its completion, either from the work itself or from other published works.

Example 1.12 Questions for probability

A glass window in a high rise building falls to the street below. A structural engineer is asked to
calculate the probably that this event will occur again. How can the engineer verify her results
given that only one event was recorded?
The flight control system in an aircraft is rated as 99% reliable. Would you fly in this
aircraft?
A civil engineer quotes a dam wall as being able to with-stand a 1 in 100 year weather event.
Would you be happy to live beneath the dam wall?
In all cases, conclusive proof is required for the safety of the population.

In mathematics, the concept of upper and lower limits can be applied to gain information about the
solution to seemingly intractable problems. For example, if a function is not integrable (i.e. cannot be
integrated using analytic means), one can choose two integrable functions – one of which is always
greater than the function, and the other is always less than the function. By completing the two simpler
integrals one can deduce the range within which the unknown function lies. In the same manner,
simple models which provide upper and lower bounds to a more complex problem can be solved
theoretically and/or computationally to deduce the likely range of outcomes from the more complex
model. This method can provide additional support to the research results, even when a complete
model has not been solved.

1.6 Why take on a research project?

There are many reasons for an engineer to become involved in a research project. There are benefits
to society as part of an engineer's charter as well as personal rewards.
There is significant excitement in new discoveries. The possibility of developing something
completely new, something of benefit to humanity and something to add to the world-wide body of
knowledge is a strong motivation for undertaking research. While practising engineers may leave
their mark on society through buildings, dams, rail-lines, aircraft, electronics products, medical
devices, etc, usually the team of engineers or their company might only be known to the general
public through a temporary sign on the work-site or a slim column in a newspaper. Sanitation and
pipeline engineers, software engineers and many others might never be recognised publicly. It is
often the case that the names of engineers become known when there are catastrophic failures. The
publication of research outputs in the archival literature means that names of the members of the
research team will never be lost in time. The research team will be recorded permanently.
A successful research project can enhance your career. The peer review process and open
publication means that the team is capable of work at the highest possible standard. This recognition
is world-wide and can be used to advantage in developing tenders for international projects.
The process of research training is part of every engineering undergraduate degree programme.
Every laboratory experiment and every calculation is part of the development of an engineer, and the
skills learnt are part of the expertise that can be used in research projects. For this reason many
undergraduate engineering degree programmes either include or plan to include a research project
or a research methods component. Undergraduate research training can provide a thorough
grounding in many of the techniques outlined in this book.
In a complementary way, engineering faculty staff should mandate that laboratory reports,
assignments and other deliverables employ the typical research requirements outlined in this book –
the research writing format, statistical analysis, appropriate checks on the validity of results and
appropriate referencing. In addition, the development of lecture material and other resources should
follow the norms of research methods, referencing, etc. This solid reinforcement of research
techniques can greatly improve the quality of graduates and improve the student experience in
undergraduate engineering programmes.

1.7 Chapter summary

The objective of all engineering research is the creation of new knowledge which can be of benefit to
humanity. New knowledge is based on well-known and well-accepted principles of scientific thought
and measurement, and should be initiated through the development of a research question.
In order that a research contribution can be claimed, novice researchers must have a good
understanding of the basic principles of their discipline as commonly found in the well-known
textbooks in the field.
All engineering research is based on the scientific method of validation using the peer review
process. Peer review of the research outcomes is an essential process in research.
All engineering researchers must be familiar with their code of ethics and must behave ethically.
In particular the research team must seek out and acknowledge the work of others.

Exercises
Many of the exercises in this book require the reader to choose a refereed, published paper in their
discipline, to analyse the paper according to the instructions given in the exercises, and to extend the
work reported as a new line of research.

1.1 Use an academic web search to locate a journal paper which describes a design outcome in
your field of interest (i.e. your engineering discipline). You must enter several keywords
which relate to your topic. Read the paper and, using your own words, demonstrate your
 understanding of the paper by:

• Writing out the major conclusions of the paper;

• Outlining the verification method(s) used to support these conclusions
• Describing the authors’ reflective comments on the quality of the design (positive
and negative).

List your comments on the following:

• The positive and negative environmental impacts of the new design;

• The fail-safe quality of the new design;
• The cost of manufacture or implementation of the new design compared to
previous designs.

1.2 Find the code of ethics which covers your engineering discipline (the code might be
specific to one particular country or geographical region). Rank the following aspects in
terms of importance (usually indicated by the position in the code of ethics). Assume that
the first tenet is the most important:

• Responsibility to the environment;

• Responsibility to sustainable outcomes;
• Responsibility to an employer;
• Responsibility to the general public;
• Responsibility to the nation.

Following the strategy in Exercise 1.1, choose a published paper from your engineering
discipline and comment on any of these priorities identified in this paper. Compare your
results with the code of ethics list. What is your conclusion about the ethical approach
taken in this paper?
1.3 After reading a published research paper, write down the research question you think the
authors have addressed in undertaking this research. Do you think the paper adequately
supports the conclusions reached in addressing this question?
1.4 In your undergraduate education or master's degree, choose three experiments (practical
laboratory work) and identify the following:

• The aims of and methods used in the experiment;

• The conclusions from the experimental work;
• The methods of verification used to support your conclusions;
• The theory applied to support the experimental outcomes;
• The statistical techniques used to verify your outcomes.

Do you consider that the post experimental analysis supported the experiment aims and
conclusions? Outline how this research method might be improved to verify the outcomes
through conclusive proof.
References
Keywords: engineering research, engineering ethics, code of ethics, research question, engineering
sustainability, engineering disasters
[1] Davis, M., ‘Defining engineering: how to do it and why it matters’, in M. Davis (ed.), Engineering
Ethics, Aldershot, England: Ashgate, 2005.

[2] Popper, K.P., The Logic of Scientific Discovery. New York: Basic Books, 1959.
[3] Gattei, S., Karl Popper's Philosophy of Science: Rationality without Foundations, New York:
Routledge 2009.

[4] Kuhn, T.S., The Structure of Scientific Revolutions, (2nd edition), Chicago: University of Chicago
Press, 1970.
[5] Marder, M.P., Research Methods for Science, Cambridge, UK: Cambridge University Press, 2011.

[6] O’Donoghue, P., Research Methods for Sports Performance Analysis, Oxford, UK: Routledge,
2010.

[7] Walliman, N., Research Methods, the Basics, Oxford, UK: Routledge, 2011.
[8] Fleddermann, C.B., Engineering Ethics, (4th edition), Upper Saddle River, NJ: Prentice Hall,
2011.

[9] Engineers Australia, ‘Our code of ethics’, adopted 28 July 2010,

www.engineersaustralia.org.au/ethics, accessed 4 December 2012.

[10] American Society of Civil Engineers, ‘Code of Ethics, Fundamental Principles’,

http://www.asce.org/Leadership-and-Management/Ethics/Code-of-Ethics/, accessed 4 December
2012.

[11] Hanrahan, H., ‘The Washington Accord: past present and future’, IEET accreditation training,
International Engineering Alliance, 2011. http://www.washingtonaccord.org/washington-
accord/Washington-Accord-Overview.pdf
[12] Whitbeck, C., Ethics in Engineering Practice and Research, Cambridge, UK: Cambridge
University Press, 1998.