Heriot-Watt University
ROS: The Research Output Service. Heriot-Watt University EdinburghNot a member yet
4689 research outputs found
Sort by
Proposed cost data management framework to accelerate whole life costing in Tanzania building construction industry
Get PDFCapital cost has been the key criterion for decision making when it comes to building
construction, until the realisation of whole life cost. Whole life costing plays an essential
role in ensuring that value for money is attained throughout the building's life. The
Tanzania building construction faces the challenge of not incorporating whole life costing
in decision making. One of the key barriers to whole life costing is the lack of reliable
cost data to facilitate its undertaking.
This study aims to develop a proposal cost data management framework to help accelerate
whole life costing in the Tanzania building construction industry. The literature review
was used to collect secondary data which involved gaining in-depth knowledge on whole
life costing and cost data management. Sequential explanatory mixed method research
was used to collect primary data in which questionnaire surveys followed by interviews
were conducted to explore whole life costing and cost data management from Tanzania
building construction professionals. A total of 77 and 63 questionnaires were completed
on whole life costing and cost data management respectively, and 20 interviews were
conducted to gain a deeper understanding.
Findings from the literature review, the questionnaire survey and the interview, lead to
the development of a new proposed cost data management framework for the Tanzania
building construction industry. The proposal framework sets out a path on how cost data
should be collected, analysed, stored and disseminated in the Tanzania building
construction industry to ensure there is availability of reliable cost data to facilitate whole
life costing undertaking.
The proposed framework for cost data management was evaluated by Tanzania building
construction professionals through a mixture of questionnaire surveys followed by
interviews. A total of 34 questionnaires were completed, and 11 interviews were
conducted. The proposed framework was considered to be clear, comprehensive, relevant
to Tanzania's building construction industry and relevant in promoting whole life costing
by providing reliable cost data
Semi-discrete optimal transport methods for the semi-geostrophic equations
Get PDFWe develop a framework for using semi-discrete optimal transport theory to rigorously study the semi-geostrophic equations, which model large scale atmospheric
flows and frontogenesis. Our framework is based on the geometric method of Cullen
and Purser (1984) – an energy-conserving Lagrangian discretisation of the semi-geostrophic equations. Within this framework, we give a new and constructive
proof of the existence of global-in-time weak solutions of the 3-dimensional incompressible semi-geostrophic equations in geostrophic coordinates, obtaining improved
time-regularity for a large class of discrete initial measures. Our proof is advantageous in its simplicity and its explicit relation to Eulerian coordinates through the
use of Laguerre tessellations. We give explicit examples of solutions to the discrete
system and we demonstrate how they can be used to approximate Eulerian solutions of the semi-geostrophic equations. Our proof naturally gives rise to a new
and efficient implementation of the geometric method, which we use to solve the
semi-geostrophic Eady slice equations – a formal low Rossby number approximation
of the Eady-Boussinesq vertical slice equations. Our implementation combines the
latest results from numerical optimal transport theory with a novel adaptive time-stepping scheme. Since the geometric method is an energy-conserving discretisation,
it is desirable to initialise this scheme with discrete approximations of a given initial
condition that have a specified energy. We prove that this is possible for a wide
class of initial conditions. Our numerical results support the conjecture that weak
solutions of the Eady-Boussinesq vertical slice equations converge to weak solutions
of the semi-geostrophic Eady slice equations as the Rossby number tends to zero.UK Engineering and Physical Sciences Research Council (EPSRC) Grant -
EP/L016508/0
Deep learning applied to the assessment of online student programming exercises
Get PDFMassive online open courses (MOOCs) teaching coding are increasing in number and popularity. They commonly include homework assignments in which the students must write code that is evaluated by
functional tests. Functional testing may to some extent be automated
however provision of more qualitative evaluation and feedback may
be prohibitively labor-intensive. Provision of qualitative evaluation at
scale, automatically, is the subject of much research effort.
In this thesis, deep learning is applied to the task of performing
automatic assessment of source code, with a focus on provision of
qualitative feedback. Four tasks: language modeling, detecting idiomatic code, semantic code search, and predicting variable names are
considered in detail.
First, deep learning models are applied to the task of language modeling source code. A comparison is made between the performance of
different deep learning language models, and it is shown how language
models can be used for source code auto-completion. It is also demonstrated how language models trained on source code can be used for
transfer learning, providing improved performance on other tasks.
Next, an analysis is made on how the language models from the
previous task can be used to detect idiomatic code. It is shown that
these language models are able to locate where a student has deviated
from correct code idioms. These locations can be highlighted to the
student in order to provide qualitative feedback.
Then, results are shown on semantic code search, again comparing
the performance across a variety of deep learning models. It is demonstrated how semantic code search can be used to reduce the time taken
for qualitative evaluation, by automatically pairing a student submission with an instructor’s hand-written feedback.
Finally, it is examined how deep learning can be used to predict
variable names within source code. These models can be used in a
qualitative evaluation setting where the deep learning models can be
used to suggest more appropriate variable names. It is also shown that
these models can even be used to predict the presence of functional
errors.
Novel experimental results show that: fine-tuning a pre-trained
language model is an effective way to improve performance across a
variety of tasks on source code, improving performance by 5% on average; pre-trained language models can be used as zero-shot learners across a variety of tasks, with the zero-shot performance of some architectures outperforming the fine-tuned performance of others; and
that language models can be used to detect both semantic and syntactic errors. Other novel findings include: removing the non-variable
tokens within source code has negligible impact on the performance of
models, and that these remaining tokens can be shuffled with only a
minimal decrease in performance.Engineering and Physical Sciences Research Council (EPSRC) fundin
Behaviour-driven motion synthesis
Get PDFHeightened demand for alternatives to human exposure to strenuous and repetitive labour, as
well as to hazardous environments, has led to an increased interest in real-world deployment of
robotic agents. Targeted applications require robots to be adept at synthesising complex
motions rapidly across a wide range of tasks and environments. To this end, this thesis
proposes leveraging abstractions of the problem at hand to ease and speed up the solving. We
formalise abstractions to hint relevant robotic behaviour to a family of planning problems, and
integrate them tightly into the motion synthesis process to make real-world deployment in
complex environments practical. We investigate three principal challenges of this proposition.
Firstly, we argue that behavioural samples in form of trajectories are of particular interest to
guide robotic motion synthesis. We formalise a framework with behavioural semantic annotation
that enables the storage and bootstrap of sets of problem-relevant trajectories.
Secondly, in the core of this thesis, we study strategies to exploit behavioural samples in task
instantiations that differ significantly from those stored in the framework. We present two
novel strategies to efficiently leverage offline-computed problem behavioural samples: (i) online
modulation based on geometry-tuned potential fields, and (ii) experience-guided exploration
based on trajectory segmentation and malleability.
Thirdly, we demonstrate that behavioural hints can be extracted on-the-fly to tackle highlyconstrained, ever-changing complex problems, from which there is no prior knowledge. We
propose a multi-layer planner that first solves a simplified version of the problem at hand, to
then inform the search for a solution in the constrained space.
Our contributions on efficient motion synthesis via behaviour guidance augment the robots’
capabilities to deal with more complex planning problems, and do so more effectively than
related approaches in the literature by computing better quality paths in lower response time.
We demonstrate our contributions, in both laboratory experiments and field trials, on a
spectrum of planning problems and robotic platforms ranging from high-dimensional
humanoids and robotic arms with a focus on autonomous manipulation in resembling
environments, to high-dimensional kinematic motion planning with a focus on autonomous safe navigation in unknown environments. While this thesis was motivated by challenges on motion
synthesis, we have explored the applicability of our findings on disparate robotic fields, such as
grasp and task planning. We have made some of our contributions open-source hoping they
will be of use to the robotics community at large.The CDT in Robotics and Autonomous Systems at Heriot-Watt University and The University of EdinburghThe ORCA Hub EPSRC project (EP/R026173/1)The Scottish Informatics and Computer Science
Alliance (SICSA
An exploration of safeguarding cultural heritage textiles in Thailand : the perspectives from expected and unanticipated preservers
Get PDFThis thesis aims to explore safeguarding culture in both tangible and intangible heritage
relating to textiles in Thailand. The thesis contributes to understanding the foundations of
local cultural heritage practices relating to textiles, contemporary practices of preserving
those heritage textiles, and perspectives on ‘preserving heritage’ from different fields of
participants who have been involved with conserving heritage textiles.
The research combined participant observation with practice-based auto-ethnography and
gathered further data through in-depth interviews and a focus group. The research focuses
on key actors involved in safeguarding the cultural weaving heritage in Thailand through
their contribution to textiles practices. The key actors are divided into three main groups:
local practitioners, unanticipated preservers, and expected preservers, by doing so, they
present how people from different fields can be part of safeguarding cultural heritage
activities. With key actors from various fields, including local textiles makers, Buddhist
monks, fashion designers, businesswomen, authorities, academics, and foreigners based
in Thailand, the research explores their various modes of involvement in safeguarding
cultural heritage textiles, identifies their common problems and considers their
suggestions for designing a potential framework (to model ways) to safeguard heritage
textiles.
A key theme emerging from the fieldwork was the importance of educating producers
and customers wishing to support local craft as part of a fashion business, especially in
developing countries. Findings from this case study in Thailand, a country rich in craft
but experiencing the familiar stresses of local products being undercut by imported
fashion, offer lessons for other countries and regions suffering from decreased
consumption of local makers. The case considers the continuation of direct craft
consumption and local craft products into local fashion against current economic trends
of global fashion production turning to local crafts.
The fieldwork mainly collected data from local communities in Northern Thailand,
especially in the so-called ‘Lan Na Kingdom’, where local culture and Buddhist practices
are heavily intertwined with heritage textiles. The research uncovered many challenges
to forging successful collaborations between local craft makers and global design players;
amongst other things, it found arguments and misunderstandings on the purpose of specific design approaches, issues with lack of recording practice, the willingness to
educate and thus enable to continue heritage skills, and the interest in being part of present
global fashion demands. After careful reflection and analysis, the thesis can suggest
solutions to such issues, with particular emphasis on how to link sustainability themes to
safeguarding heritage textiles concerns, with conclusions reaching beyond the specific
field studied in this instance
Victim moves or survivor stays? Domestic abuse safe housing in England and Scotland
Get PDFThis doctoral research focuses on a key question arising from the interconnection between housing
problems and domestic abuse: How can we prevent survivors of domestic abuse from becoming
homeless while ensuring their safety and aiding their recovery? Since the 1970s, responses to this
question in both England and Scotland have largely relied on the statutory homelessness system and
domestic abuse refuges to relocate survivors to safer housing, in what can be described broadly as
‘victim moves’ approaches. However, since the late 1990s, sanctuary schemes aim to provide an
alternative form of protection for survivors within their own homes by excluding the perpetrator and
offering enhanced security features and support. This can be described as a ‘survivor stays’ approach.
There, have, however, been differences of emphasis between England and Scotland. For example,
England has had a sharper focus on homelessness prevention, with the promotion of sanctuary
schemes (i.e., enhanced security measures in the home for survivors of domestic abuse) within this,
while Scotland has focused on widening the statutory homelessness safety net for all households.
Within domestic abuse policy, England has adopted a more criminal justice focused lens, while
Scotland has approached domestic abuse as an issue of wider gender-based inequality.
In this comparative study of these two countries, I first examine the role of the statutory homelessness
system for domestic abuse survivors facing housing challenges and assess its relative strengths and
weaknesses. Secondly, I ask what role refuges currently play and what the strengths and weaknesses
are of this model. Thirdly, I trace the distinctive evolution of sanctuary schemes in the two countries
and consider whether they are a valuable addition to the service network assisting domestic abuse
survivors. To answer these questions, I draw on evidence from in-depth qualitative semi-structured
interviews with 33 key stakeholders in homelessness, housing, and domestic abuse services in
England and Scotland, and two service user focus groups with women who had experienced domestic
abuse and lived in refuge or had personal experience with sanctuary schemes.
The availability of rights under the statutory homelessness system was found to be a significant
positive and vital safety net in both countries, though it was viewed as superior in Scotland relative
to England because 1) priority need was given to all survivors, not just those with children, and 2) the
better quality of available TA. However, key informants also highlighted practical difficulties in both
countries: for example, staff being inadequately versed in assisting survivors, as well as more in-principle objections to the homelessness system being the default route out of domestic abuse.
TA in a specialist refuge was generally seen as better than generic TA, with key informants
emphasising increased safety and professional expertise as the crucial benefits of this setting. Mutual support in shared experiences was also lauded, though this element has diminished in importance
with time and some key informants actually understood it to be potentially detrimental to recovery.
Traditional shared refuges where households were expected to share facilities such as bathrooms and
kitchens were found to be a highly problematic and out-dated model, with both cluster refuges (which
offer self-contained living accommodation and, in some cases, shared social spaces) and dispersed
refuges (in ordinary housing) viewed as more appropriate.
Sanctuary schemes were promoted by central Government in England and positively evaluated in
national research, whereas Scottish sanctuary schemes appear to have developed ‘bottom up’,
partially due to resistance from an influential stakeholder. Ideological concerns saw sanctuary
schemes within a gender-neutral criminal justice framing, one which did not adequately consider the
complexity of domestic abuse and placed responsibility for protection with the survivor, rather than
society. While critics in both countries articulated concerns that sanctuary schemes could be
inappropriate if offered without proper follow-up and tie-in to the wider service network, the evidence
presented in this study indicates that sanctuary schemes can provide physical and psychological
benefits for survivors of abuse in addition to important societal benefits such as aiding homelessness
prevention and promoting a survivor stays agenda
Class-incremental lifelong object learning for domestic robots
Get PDFTraditionally, robots have been confined to settings where they operate in isolation and in highly
controlled and structured environments to execute well-defined non-varying tasks. As a result,
they usually operate without the need to perceive their surroundings or to adapt to changing
stimuli. However, as robots start to move towards human-centred environments and share the
physical space with people, there is an urgent need to endow them with the flexibility to learn
and adapt given the changing nature of the stimuli they receive and the evolving requirements
of their users. Standard machine learning is not suitable for these types of applications because
it operates under the assumption that data samples are independent and identically distributed,
and requires access to all the data in advance. If any of these assumptions is broken, the model
fails catastrophically, i.e., either it does not learn or it forgets all that was previously learned.
Therefore, different strategies are required to address this problem.
The focus of this thesis is on lifelong object learning, whereby a model is able to learn
from data that becomes available over time. In particular we address the problem of classincremental learning with an emphasis on algorithms that can enable interactive learning with
a user. In class-incremental learning, models learn from sequential data batches where each
batch can contain samples coming from ideally a single class. The emphasis on interactive
learning capabilities poses additional requirements in terms of the speed with which model
updates are performed as well as how the interaction is handled.
The work presented in this thesis can be divided into two main lines of work. First,
we propose two versions of a lifelong learning algorithm composed of a feature extractor
based on pre-trained residual networks, an array of growing self-organising networks and a
classifier. Self-organising networks are able to adapt their structure based on the input data
distribution, and learn representative prototypes of the data. These prototypes can then be
used to train a classifier. The proposed approaches are evaluated on various benchmarks under
several conditions and the results show that they outperform competing approaches in each
case. Second, we propose a robot architecture to address lifelong object learning through
interactions with a human partner using natural language. The architecture consists of an
object segmentation, tracking and preprocessing pipeline, a dialogue system, and a learning
module based on the algorithm developed in the first part of the thesis. Finally, the thesis also
includes an exploration into the contributions that different preprocessing operations have on
performance when learning from both RGB and Depth images.James Watt Scholarshi
A critical study of the interrelationship of management control systems and workplace friendships in SMEs : a Greek case study
Get PDFManagement Control (MC) and Management Control Systems (MCS) have been
examined mostly in large organisations. This study attempts to understand more about
them in the SME context. Additionally, the lens is turned to workplace friendships
when looking at the human factor from a critical perspective. The SME environment
allows for more human interactions, providing the participants with more chances for
strong organisational culture, high trust, and the development of friendships, which
can also occur between manager-owner and employee. The interest of this study is to
identify if these friendships are influenced by MC and MCS, or if they influence the
nature of MC and MCS in SMEs. An additional step is taken to identify ways forward
in which workplaces can use friendships in their environments to be emancipated
from oppressive accounting practices by shaping accounting practices with an aim to
realise their emancipatory potential.
This research project was based on two case studies, in Agriculture and Clothing. The
researcher observed the two organisations for approximately 100 hours, and
interviewed 19 of their organisational participants, between 12-45 minutes. The
findings show that the SME context provides grounds for workplace friendships to
develop and through them the shape of MC and MCS from formal and impersonal to
informal and engaging, to even providing chances for a collective management and
participation ignoring the structures of the status quo. Where MCS seem to have an
impact on workplace friendships is when management, in the name of rational
decision-making, chooses to detach itself and force formal controls. Then, not only
the employees and the management are distant from each other, but internal trust
building also becomes difficult. The theoretical contribution of this research lies in
understanding the importance of critical research in SMEs and their fruitful context,
as well as reimagining SMEs and their internal structures from a critical standpoint,
by understanding their pressures in the capitalist world
Data-driven prognostics for critical electronic assemblies and electromechanical components
Get PDFThe industrial digitalisation enables the adoption of robust, data-driven maintenance
strategies that increase safety and reliability of critical assets such as electronics.
And yet, an implementation of data-driven methods which primarily address the
industrialisation of diagnostic and prognostic strategies is opposed by various, application specific challenges. This thesis collates such restricting factors encountered
within the oil and gas industry, in particular for the critical electrical systems and
components in upstream deep drilling tools. A fleet-level, tuned machine learning
approach is presented that classifies the operational state (no-failure/ failure) of
downhole tool printed circuit board assemblies. It supports maintenance decision
making under varying levels of failure costs and fleet reliability scenarios. Applied
within a maintenance scheme it has the potential to minimise non-productive time
while increasing operational reliability. Likewise, a tailored and efficient deep learning data pipeline is proposed for a component-level forecast of the end of life of
electromagnetic relays. It is evaluated using high resolution life-cycle data which
has been collected as a part of this thesis. In combination with a failure analysis,
the proposed method improves the prognostics capabilities compared to traditional
methods which have been proposed so far in order to assess the operational health of
electromagnetic relays. Two case studies underpin the need for tailored prognostic
methods in order to provide viable solutions that can de-risk deep drilling operations.
In consequence, the proposed approaches alleviate the pressure on current maintenance strategies which can no longer meet the stringent reliability requirements of
upstream assets
Photoactive materials enabled by and for emerging synthetic technologies
Get PDFIn recent years there has been a significant increase in the development and
commercialisation of new synthetic tools and technologies, which offer significant
advantages compared to traditional round-bottomed flask chemistry. This work explores
the use of some of these emerging technologies with the goal of developing new
photocatalytic processes, which would not otherwise be easily feasible with batch
techniques. Specifically, we have used continuous flow chemistry, mechanochemistry,
and 3D printing in four distinct research projects. This thesis is split into five chapters in
total.
Chapter 1 aims to act as a global introduction to the different synthetic technologies that
were used and compare their utility and drawbacks against traditional batch synthetic
methodologies. The remaining chapters 2-5 each represent a separate research project
that utilises these new technologies. Each chapter contains its own introduction to the
topic of research along with conclusions and proposed future work.
Specifically, in Chapter 2, a rapid, high yielding, and work-up free synthesis of an
unusual organic luminophore is developed. Its twisted, propeller-like geometry gives rise
to much sought aggregation induced emission properties. Moreover, we were able to use
this material as a reusable heterogeneous photosensitiser to produce singlet oxygen under
continuous flow conditions.
In Chapter 3, we report the unprecedented ring contraction of 1,2,6-thiadiazines to 1,2,5-
thiadiazole 1-oxides. The transformation is fast, work-up free, offers quantitative yields,
and is mediated by auto-photosensitised singlet oxygen. We exploited continuous flow
processes to further improve the reaction scope and efficiency.
Then, in Chapter 4 we describe the batch and mechanochemical syntheses of optically
active dihydroxamic acids ligands, and their subsequent use for the synthesis of metalloorganic assemblies.
Finally, Chapter 5 demonstrates how mechanochemical and 3D printing technologies
were used to access a series of N-aryl amides from O-protected hydroxamic acids.
The broad scope of this work aims to demonstrate the usefulness of alternative reactor
designs in chemical synthesis and encourage their implementation by others
Access Repository Dashboard
Do you manage ROS: The Research Output Service. Heriot-Watt University Edinburgh ? Access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard!