Heriot-Watt University
ROS: The Research Output Service. Heriot-Watt University EdinburghNot a member yet
4689 research outputs found
Sort by
Exploring the adoption of Kraljic’s Portfolio Model in the upstream oil and gas company : a case study of an upstream multinational oil and gas company operating in Southeast Asia
Get PDFOil and gas companies are involved in complex and challenging operations due to the
circumstances faced in the environment where oil and gas companies operate. In this industry,
companies rely heavily on their suppliers to execute their operational activities. However,
many researchers have criticised the sourcing and supplier relationship management approach
adopted in the industry as being weak, inefficient, and lacking a structured methodology to
derive value-add for the companies. Therefore, for oil and gas companies to be competitive in
the marketplace, it is imperative to adopt appropriate Procurement strategies required to deal
with suppliers in the entire supply chain system to reduce cost and improve resilience to
weather the constant market volatility of the oil price fluctuation. These strategies should be
tailored to match the required criticality of the goods and services.
This study aimed primarily at exploring the adoption and/or adaptation of Kraljic’s (1983)
Purchasing Portfolio Model (KPM) in the procurement process to determine how this tool can
be leveraged to devise procurement strategies to support the Oil Company within the oil and
gas industry context. A case study methodology was adopted for this research by considering
an oil and gas company operating in Southeast Asia (which opted to remain anonymous and is
named in this study as an Oil Company). Primary data was collected via semi-structured
interviews with six participants involved in the decision-making for the supplier selection and
relationship management process. A qualitative thematic analysis technique was adopted to
analyse the data.
The analysis outcome strongly evidenced that although Kraljic’s model has been prevalent in
other industries and also adopted in the oil and gas sector, however, to be effective, this tool
will require some adaptation. Furthermore, the study identified that the lack of detailed
sourcing guidelines was the most predominant from the weaknesses identified with KPM.
Therefore, this gap led to the development of a Guideline with a simplified step-by-step
Sourcing Guide based on KPM. The Guideline should aid practitioners of the case study
company and other oil and gas companies in developing strategies that can reduce cost and
supply chain risks
Millimetre-wave radar development for high resolution detection
Get PDFAutomotive technology today is focusing on autonomous vehicle development. The
sensors for these systems include radars due to their robustness against adverse
weather conditions such as rain, fog, ash or snow. In this constant search for advancement, high resolution systems play a central role in target detection and avoidance. In this PhD project, these methods have been researched and engineered to
leverage the best radar resolution for collision avoidance systems.
The first part of this thesis will focus on the existing systems consisting of the
state-of-the-art at the time of writing and explain what makes a high resolution
radar and how it can cover the whole field of view. The second part will focus on
how a non-uniform sparse radar system was simulated, developed and benchmarked
for improved radar performance up to 40% better than conventional designs. The
third part will focus on signal processing techniques and how these methods have
achieved high resolution and detection: large virtual aperture array using Multiple
Input Multiple Output (MIMO) systems, beampattern multiplication to improve
side-lobe levels and compressive sensing. Also, the substrate-integrated waveguide
(SIW) antennas which have been fabricated provide a bandwidth of 1.5GHz for the
transmitter and 2GHz at the receiver. This has resulted in a range resolution of 10
cm. The four part of this thesis presents the measurements which have been carried
out at the facilities within Heriot-Watt University and also at Netherlands Organisation for Applied Scientific Research (TNO). The results were better than expected
since a two transmitter four receiver system was able to detect targets which have
been separated at 2.2◦
in angle in the horizontal plane. Also, compressive sensing was used as a high resolution method for obtaining fine target detection and
in combination with the multiplication method showed improved detection performance with a 20 dB side-lobe level suppression. The measurement results from the
6-months placements are presented and compared with the state-of the art, revealing that the developed radar is comparable in performance to high-grade automotive
radars developed in the industry
Meta-interpretive learning of proof strategies
Get PDFIn modern mathematics, mechanised theorem proving software is playing an ever-increasing role. By enlisting the help of computers mathematicians are able to
formally prove more complex results than they perhaps otherwise could, however
those computers are still incapable of drawing many of the conclusions which would
be obvious to a human user and so human intervention is still required.
In this thesis we consider the use of an adapted machine learning technique to
begin addressing this issue. We consider the use of proof strategies to provide a
high-level view of how a proof is structured, including information about why a
particular step was taken. We extend the Metagol meta-interpretive learning tool
to facilitate learning these strategies. We begin with a small set of examples and
refine our approach, demonstrating the improvements experimentally. We go on to
discuss the learning of more complicated strategies, some of the issues faced in doing
so and how we could address them. We conclude by evaluating the experiments as
a whole, identifying the weak points in our approach and suggesting ways in which
they can be addressed in future work
Relational knowledge and representation for reinforcement learning
Get PDFIn reinforcement learning, an agent interacts with the environment, learns from feedback about the quality of its actions, and improves its behaviour or policy in order
to maximise its expected utility. Learning efficiently in large scale problems is a major challenge. State aggregation is possible in problems with a first-order structure,
allowing the agent to learn in an abstraction of the original problem which is of
considerably smaller scale. One approach is to learn the Q-values of actions which
are approximated by a relational function approximator. This is the basis for relational reinforcement learning (RRL). We abstract the state with first-order features
which consist of only variables, thereby aggregating similar states from all problems
of the same domain to abstract states. We study the limitations of RRL due to
this abstraction and introduce the concepts of consistent abstraction, subsumption
of problems, and abstract-equivalent problems. We propose three methods to overcome the limitations, extending the types of problems our RRL method can solve.
Next, to further improve the learning efficiency, we propose to learn different types
of generalised knowledge. The policy is influenced by directed exploration based on
multiple types of intrinsic rewards and avoids previously encountered dead ends. In
addition, we incorporate model-based techniques to provide better quality estimates
of the Q-values. Transfer learning is possible by directly leveraging the generalised
knowledge to accelerate learning in a new problem. Lastly, we introduce a new class
of problems which considers dynamic objects and time-bounded goals. We discuss
the complications these bring to RRL and present some solutions. We also propose a framework for multi-agent coordination to achieve joint goals represented by
time-bounded goals by decomposing a multi-agent problem into single-agent problems. We evaluate our work empirically in six domains to demonstrate its efficacy
in solving large scale problems and transfer learning
Energy-based approach to develop soft robots
Get PDFSoft robotic systems offer advantages against rigid robot systems in applications that
involve physical robot-human interactions, unstructured or extreme environments, and
manipulating delicate objects. Soft robots can offer inherently safe operation and adapt
to unknown geometry of the environment or object. The current soft robot development
approach is an empirical approach starting from a type of soft actuation technology,
whereas the development of rigid robots can start from a top-level task in a System Engineering framework. The rigid robot developer can select from well-defined components
to construct the task-orientated system. Soft robots are relatively novel systems compared with rigid robots and do not have well-defined components due to a wide range
of soft actuation technologies. The initial choice of soft actuation technology places
constraints on the system to perform the task. Soft robotic systems are not widely used
despite the advantages compared to rigid robots.
In this thesis, I study an abstraction approach to enable a System Engineering framework
to develop soft robotic systems. My research focus is on an energy-based approach that
encompasses the multi-domain nature of soft robotic systems. The impact on the final
system from the energy transfer characteristics of the initial choice of the soft actuator
has not been fully explored in the literature. I study how energy, and rate of energy
transfer (power), can describe different components of each type of soft actuation and
how the total energy can model the top-level system. This thesis includes (i) a literature
review of soft robots; (ii) an abstraction approach based on bond-graph theory applied
to soft actuation technologies; (iii) a port-Hamiltonian theory to describe the top-level
soft robotic system, and (iv) an experimental application of the approach on a type of
soft actuation technology.
In summary, I explore how energy and rate of energy transfer can provide the abstraction
approach and in time provide the well-defined components necessary for task-orientated
design approaches in a System Engineering framework. In particular, I applied the
approach to soft pneumatic systems for additional insights relevant to the development
of future task-orientated soft robotic systems.EPSRC Doctoral Training Centre in Robotics and
Autonomous Systems funding
Buyer control over outsourced service delivery in public service triads
Get PDFThe aim of this research inquiry is to understand how, in public contracts, service delivery is
controlled in relation to the separation between the buyer, service provider and service delivery.
Moreover, to explore the influence of buyer separation, task programmability, and outcome
uncertainty variables on a buyer’s control mechanisms during pre-tendering, during tendering,
and post- tendering stages in public service contracting. Next, buyer risk mitigating strategies
that facilitate maintaining control over service delivery in public sector service triads is
investigated by comparing different public service triad types that have varying buyer
separation, task programmability and outcome uncertainty levels. The research inquiry
embraces a multiple embedded case study research design and is conducted in cooperation with
Scotland Excel and seven Scottish Councils in four different service sectors – care homes,
fostering, waste management, and engineering and technical consultancy. A unique form of
thematic analysis that is suggested by Stake (2006), specifically for a structured analysis of
qualitative multiple case studies is embraced by this research inquiry. The analysis method is
conducted in two-levels, for embedded and for main cases of the research. Consequently, the
research contributes to the gap in our knowledge in three fundamental area. First, a literature
contribution by proposing a taxonomy of public sector buyer’s control difficulties over service
delivery in service triad setting. Second, a literature contribution by identifying and elaborating
an impacting variable for service triads buyer’s separation. Third, a theoretical contribution
to the principal-agency theory by identifying new principal and agent relationship types that a
specific for public sector service triads
Development of single-frequency and compact diode-pumped solid-state lasers for industrial and scientific applications
No full textAbstract and full text unavailable. Please refer to PDF. Restricted access until 01/01/2027
Modelling the contact line dynamics of two-phase flows across surfaces of longitudinal multiple wettability sections in capillary channels for the characterisation of capillary filling processes
Get PDFIn natural porous media, the wall of each induvial pore is commonly formed by
chemically different materials to give various surface energies and wettability to fluids.
In a two-phase capillary system with contrasting wet-surface sections along the capillary,
when the interface between the invading and receding fluid moves cross each wet-surface
interface, the dynamics of fluid-fluid-solid contact line is subject to a transition stage, in
an accelerating (slip) or decelerating (stick) manner, due mainly to the transformation of
surface, kinetic energy and viscous dissipation. Time and length durations of the slip-stick process are influenced by fluid properties, capillary geometry, and wettability as
well as forces driving flow through imposed boundary conditions. The slip-stick process
can have a strong impact on the dynamics of capillary filling process in individual
capillaries and therefore subsequent selection of plausible pathways for a fluid to migrate
through a pore system which is filled by another fluid. This impact is likely to be more
severe in less-well connected pore systems with strong contrasts in surface wettability
such as organic-rich shales and coals. However, the transitional dynamics of the contact
line, the slip-stick process, is poorly understood in a parameter space of practical interest.
This research fills this knowledge gap through a combined study of numerical modelling
and analytical analysis. A two-phase multi-relaxation-time colour-gradient lattice-Boltzmann (LB) model is implemented for simulating contact line movement around
wettability transitions in 2D. A 1D analytical model for homogenous-wet capillary filling,
which considers the inertial effects of contrasting fluids, is employed to derive a
dimensionless model for parameter groups and to perform dimensional analysis on energy
and forces involved. The LB code is validated using this analytical model and is used to
simulate capillary filling in 2D channels which contain two sections of different wetting
surfaces along each channel, for a set of parameters, including respective velocity and
pressure flow driven boundary conditions. The variation of the viscous dissipation
generated by the slip-stick processes is analysed and it is found that the transient dynamics
of the contact line coincides with the temporal variation of the viscous dissipation and
that duration reflects the time the contact line requires to balance the change of surface
energy arisen from the wettability change. The time and distance in which the transitional
dynamics take place are evaluated under low capillary numbers over a wide range of fluid
and capillary geometric parameters. Using the data obtained from this study, an empirical
effective capillary filling model is developed. It is shown this model can be used to
establish a simple rule-based criterion for determining whether an individual pore throat in a pore network is to be invaded or not under given boundary conditions during drainage
and imbibition processes. Finally, using this criterion, the significant impact of multi-wet
pore-throats on phase saturation in a simple pore system is demonstrated under quasi-static displacements.James Watt Scholarshi
Implication of mineralogical and petrophysical property changes to mechanical integrity of reservoir and caprock for safe CO2 storage
Get PDFThis thesis is studying the re-injection of CO2 into the same reservoir it originates from.
The reservoir is the S Field, Sarawak Basin, offshore Malaysia, and contains
contaminated natural gas with ~30% methane and ~70% carbon dioxide. The re-injection, following separation of the CO2 on the platform, is incorporated in the field
development plan for adherence to the Paris Agreement and Malaysia’s vision of net
zero-carbon emissions by 2050. The in-situ CO2 storage strategy in this offshore
Miocene formation is associated with heterogeneous carbonates and extreme reservoir
conditions (temperature ~150°C, pressure ~34 MPa), requiring an extensive assessment
as well as advanced technology solutions. The main objectives of this study are to firstly
reduce subsurface CO2 storage risks for the S Field through improved prediction and
advanced understanding of fluid-rock interactions. Secondly, and strongly interlinked
with the first point, is to provide a systematic evaluation of the geological, including
geochemical and geomechanical, properties at different scales. This study characterises
reservoir and caprock properties and investigates CO2-water-rock (CWR) interactions
through experimental and modelling approaches. Findings from both laboratory and
modelling studies are used for risk assessment for CO2 storage in the S Field. An
integrated experimental program was developed to evaluate reservoir and caprock
characteristics and their geochemical, geomechanical, mineralogical, and petrophysical
analyses during CWR. For the CWR interaction study, carbonates and caprocks were
reacted with CO2 saturated brines for up to 35 days and six months respectively, under
reservoir pressure and temperature conditions. The samples used for all experiments
were core samples from a recent appraisal well, and synthetic brine was formulated
using the actual reservoir brine composition. Sensitivity analysis for CWR interaction to
quantify mineralogical changes was performed by geochemical modelling using
PhreeqC. The carbonate formation has heterogenous properties with the highest porosity
(40%) and permeability (3000 mD) observed in the gas zone. The mineralogy in the gas
zone is dominated by calcite (80-90 wt %), while mineralogy in the aquifer zone is co-dominated by calcite (50%) and dolomite (40%). The differences in the petrophysical
properties and mineralogical composition are due to diagenesis and differences in
depositional environment. These differences also influence the geomechanical
properties, resulting in stronger carbonates in the aquifer zone compared to the gas
zone. The highest peak stress in the stress-strain curve for aquifer zone samples is 40
MPa with static Young’s modulus is 9 GPa compare to 16 MPa of peak stress in gas
zone samples and Young’s modulus of 5 GPa. The primary caprock (Seal A), composed
of Miocene mudrocks and siltstones has proven to be an effective seal with a thickness
of ~500m. The measured porosity is in the range of 2-10% and permeability of 0.5 µD.
The main mineral components of Seal A are illite and quartz; some minor amounts of
swelling clays (smectite) have been observed. Minimal changes in both mineralogy and
petrophysical properties were observed in both reservoir carbonates and caprock shale
after exposure to scCO2. Geomechanical properties such as strength and stiffness were
similar after exposure to scCO2. Based on the results, the risk assessment shows a low
impact with negligible action required to conduct CO2 storage in the S Field. The data
derived from the laboratory works were used to develop new and improved correlations
on subsurface characterization, specifically in carbonate reservoirs with similar
heterogeneity and depositional environments. For example, value creation from the
integrated subsurface characterization is the new correlation between porosity-permeability and the relationship of porosity and rock strength in carbonate formation.
In addition, findings from the laboratory and sensitivity studies can be used as input for
reservoir model calibration and 3D coupled modelling for future CO2 storage
development plans in the S Field. Applications of the developed methodology
combining various laboratory work in this study have proven satisfactory in studying
the identified subsurface uncertainty for CO2 storage development. It had been used as a
base methodology for other eight further potential CO2 sites in hydrocarbon fields
offshore Malaysia. The unique findings from the extensive and long duration of CO2-
water-rock interaction experiments at the extreme pressure and temperature conditions
in this study can be a reference or analogue data to other hydrocarbon fields with similar
conditions. On top of that, the comprehensive geological risk assessment using bow tie,
risk matrix, and heat map developed from this study can guide other high CO2 field
development plans in carbonate reservoirs
The comparative abundance and behaviour of sharks in the Cayman Islands (BWI)
Get PDFThis thesis investigates the ecology of coastal sharks, in particular the Caribbean reef shark
(Carcharhinus perezi), using data collected in the Cayman Islands from 2009 to 2019. Relative
abundance was estimated by deploying 936 Baited Remote Underwater Video Systems
(BRUVS) and individual shark behaviour monitored by recording the movement of 66
acoustically tagged Caribbean reef sharks. This work was complemented by information
gathered on 24,442 dives by 69 SCUBA divers participating in a citizen science programme
(the ‘Sharklogger Network’). Additionally, population sizes for two species were estimated
through the application of mark-recapture models to the sighting histories of sharks individually
identified on 557 BRUVS videos deployed between 2015 and 2018. The study recorded eight
shark species of which Caribbean reef and Atlantic nurse shark (Ginglymostoma cirratum) were
the most abundant, with estimated local coastal population sizes of 180 and 336 respectively.
Immature and mature Caribbean reef, nurse, and hammerhead (Sphyrna) sharks were recorded
throughout the year, suggesting resident breeding populations. Most Caribbean reef and nurse
sharks showed relatively small home ranges (< 20 km) and high site-fidelity. However, some
individuals showed maturity-based seasonal movements indicating a distinct mating and
pupping season, with a few detected moving more than 100 km
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!