PhD Studentships: School of Built Environment, Engineering and Computing

Leeds Beckett University

Leeds, UK 🇬🇧

Qualification Type:PhD
Location:Leeds
Funding for:UK Students, EU Students, International Students
Funding amount:£18,622
Hours:Full Time
Placed On:12th May 2023
Closes:7th June 2023

The School of Built Environment, Engineering and Computing at Leeds Beckett University is currently accepting applications for up to 7 funded full-time PhD Research Studentships

The School of Built Environment, Engineering and Computing at Leeds Beckett University brings together a range of related sectors from built environment and construction; human and natural environments; engineering industries; and computing and IT industries in a vibrant setting, with a rapidly growing research, enterprise, and student cohort.

Following our success in REF 2021 the school is offering up to 7 full-time studentships from October 2023. The studentships cover projects related to the built environment, engineering, and computing sectors. Studentship funding lasts for the first three years of a four-year registration period.  The fourth year is a self-funded ‘writing-up’ year.

We invite applications from talented people to apply for studentships on the following project areas:

Artificial intelligence-based document classification tools

Document classification is an essential tool that has a significant impact on many industries, including government, healthcare, and finance. The ability to organize vast amounts of data quickly and accurately into relevant categories helps decision-makers access and analyse information efficiently. Document classification can also uncover patterns and trends that would be difficult to detect otherwise, such as in the context of environmental remediation, where it can help specialists pinpoint the critical information needed to develop effective strategies for cleaning up contaminated sites. To address the challenges associated with document classification: 1. We will collaborate with the National Nuclear Laboratory (NNL) to develop a tool that uses two artificial intelligence methods: natural language processing (NLP) and deep learning (DL) techniques to classify NNL documents into three categories: not useful, offering insights, and not offering insights. 2. We will utilize state-of-the-art techniques such as knowledge distillation and transfer learning to simplify the DL network’s structure, reduce training time, and improve model performance. 3. We will explore different deep neural networks (NNs), such as FNNs (Feedforward), RNNs (Recurrent), CNNs (Convolutional), TCN (Temporal Convolutional), hybrid RNNs and CNNs, and build a NN ensemble based on the outputs of the best three NNs. 4. We will use various techniques to determine the minimum data sets required for effective model training, such as data mining and time-frequency analysis. The outcomes of this project will be a valuable tool, methods for simplifying NN structures, and the best NN models that will enable NNL specialists to categorize documents quickly and accurately. Ultimately, this project will contribute to the efficient organization and retrieval of vast amounts of data, improve decision-making processes, and lead to world leading research.

Cyber Security Innovations in Simulations, Serious Games, and Edutainment

Hands-on cyber security and hacking challenges are an effective way to engage learners and unlock the “security mindset”. Constructivist approaches to teaching suggest that students learn best by “learning through doing”. One such approach is through capture the flag (CTF) security challenges, where solving technical challenges reveal flags which can be used to prove and score that they have been solved. Most CTF challenges are created as one-use static challenges. We have developed unique solutions that enable randomised and replayable hacking and defensive challenges. SecGen is a platform for generating randomised VMs containing hacking challenges; Hackerbot is a chatbot that can help students to learn about defense and monitoring; and we have developed a cloud-based lab infrastructure for conducting computer security exercises.

We invite Ph.D. proposals that build on these frameworks and techniques, to develop and evaluate innovative technical solutions to cyber security simulations, serious games, and edutainment. Proposals will be technical in nature and will propose, implement, and measure the impact of new innovations.

Model the impact of behind-the-meter EV charger on smart dwellings’ energy demand and control strategy as function of commuting patterns

The student will model, using computer simulations, the impact of electric vehicles (EVs) on electricity demand, in combination with residential demand, with focus on the load generated by EVs used for commuting. The simulated EV charger will be placed “behind the meter”; I.E., the energy demand of the plugged-in EVs will be metered and billed together with the energy usage of every other appliance in the house. The simulation will be built upon the output dataset of the RED WoLF project’s Pilot smart houses and on the Governing AI Algorithm (GAIA) managing energy demand in these houses, located in UK, Ireland, France, and Luxembourg. These dwellings are equipped with thermal storage, battery storage and PV arrays. The timing of electricity intake and storage is managed by an AI algorithm with the optimization objective of reducing carbon emissions, as well as electricity bills when a dynamic tariff is available.

Development of low carbon high performance cement free concrete for structural applications

The project will primarily focus on the development of cement free concrete which can potentially be used for structural applications, e.g., floors, DPC, foundations, walls, tunnels, bridges, sea defense walls & highways applications, etc. This entails evaluation of the mechanical, physical and durability properties. To further address the sustainability issue, research will also focus on incorporating various types of recycled plastic.  Along with physical experiments, machine learning techniques will be utilized to analyse material properties to reduce the demand for resources such as labour and time.

The safe and optimum implementation of plant extracts to provide potable water treatment for rural communities in the developing world

In developed countries, the most suitable method of removing coliforms and turbidity from drinking water is coagulation. The most used chemical coagulants are alum and ferric: these chemicals are cost-prohibitive in developing regions. Freely available natural plant extracts have been available for water purification for many centuries. However, the science and engineering application of the use of plant extracts have not been fully developed. Initial investigations by the DoS have shown that M. oleifera can be used to improve water quality by around 85%, which is closely comparable to alum.

The safe implementation of such plant extracts into the human food chain still needs to be developed. An initial study by the DoS has shown that M. oleifera extract proved non-toxic at concentrations from 0 to 1000 mg/l, based on the ecotoxicity results. However, when tested on human epithelial and intestinal cell lines (HeLa and CACO-2) the extract started toyield a cytotoxicity response at concentrations of >50 mg/l.

Research question: Can plant extractsfor water purificationbe safely introduced into the human food chain?

Application of Hyper-Spectral Imaging to Assess the Level of Aflatoxins in Pistachio Nut

The consumption of pistachios has been increasing, given their important benefit to human health. In addition to being an excellent nutritional source, they have been associated with chemical hazards, such as mycotoxins, resulting in fungal contamination and its secondary metabolism. Aflatoxins are the most common mycotoxins in pistachio and the most toxic to humans, with hepatotoxic effects. Exposure to aflatoxins is associated with an increased risk of liver cancer.  In this project, a hyper-spectral image dataset of sample pistachio nuts with different known level of aflatoxin contamination is first generated. Then, an investigation into spectral band selection for aflatoxin infection detection will be conducted. Finally, the application of Artificial Intelligent for detection and classification of the infected individual nuts will be investigated.

BIM4Regs – A framework for facilitating automated building regulations compliance diagnostics in BIM Model

The advent of Building Information Modelling and its associated frameworks and standards promises a new era, towards enhanced productivity in the construction industry. Consequently, BIM has been touted as an effective way of addressing issues affecting productivity, profitability, sustainability, safety, and overall effectiveness of the construction industry (). Following its contribution to driving the Construction 2025 agenda, which is aimed at 33% lower cost and 50% faster delivery, among others (HM government, 2016), the need to extend BIM to building regulations compliance diagnostics has been envisioned by the UK government through the BIM4Regs initiative.

The proposed study aims to investigate the optimal approach for facilitating automated building regulation compliance diagnostics. The study would explore state-of-the-art technologies to develop framework for automated building compliance diagnostics and decision support system. It would also involve the development of a BIM4Regs prototype, using Building Regulations Approved Document B as a case study.

Creating an Optimised Retrofit Matrix for Structural Integrity and Moisture Risk Assessments For Retrofitting Homes

The UK has set a target to achieve Net Zero emissions by 2050, and the Committee on Climate Change (CCC) has emphasised that almost all the 29 million homes in the UK will need to be retrofitted to meet this goal. However, the longevity of insulation measures is variable and warranties for retrofit products tend to be limited to 25 years. Interpretation and solutions for managing moisture risks associated with retrofit measures are also inconsistently implemented. There are also growing concerns that many retrofit products may fail or underperform after several decades meaning that some retrofits may need replacing to safeguard the home’s structure and the health of occupants. This creates a loop of retrofitting and re-retrofitting that goes against the concept of sustainable retrofitting of existing housing stocks.

To address this issue and promote sustainable retrofitting practices, this PhD study proposes to investigate the structural integrity and moisture risks associated with various energy retrofitting solutions. This will inform risk-based retrofit decision-making by gaining a better understanding of the potential structural and moisture risks associated with retrofitting products. The study will adopt a mixed method comprising both Numerical study (to assess the structural integrity and moisture risks) and Qualitative study (to explore the importance of these issues to stakeholder decision-making). The findings will be combined with a Life Cycle Assessment (LCA) of retrofit solutions to develop an Optimised Retrofit Matrix (ORM), to support policy design and decision making which will be trailed by landlords and other stakeholders.

Homes as a system: incorporating occupant behaviour into domestic energy efficiency policy to achieve Net Zero.

The pursuit of Net Zero is an urgent global priority. A substantial contributor to carbon emissions is space heating in domestic dwellings.  In response, the UK government has committed to substantial energy reduction, with policy focused on demand reduction via insulation and low carbon heating systems. Existing research shows that the effectiveness of technical solutions to reduce domestic heat demand relies on occupants using systems as predicted. However, energy efficiency is not the main driver behind energy use in homes.  For example, people may prioritise comfort, convenience, or have social norms that result in energy use which may not maximise efficiency. The home is therefore a system that the includes the building fabric, heating system and the users. User variation is not currently factored into decision making or modelling, yet materially affects the energy efficiency of the home and may determine what retrofits are most appropriate.

This PhD acknowledges that domestic energy retrofit is a socio-technical challenge and seeks to define the social factors that must be accounted for to maximise retrofit success.

Retail Property Location Optimisation

This project aims to provide digital solutions to the crisis facing the UK retail highstreets and retail property locations through multidisciplinary approach. It focuses on investigating retail property location performance with a view of developing retail location optimisation model to serve as decision making tool for various retail real estate activities. As such, the research project would develop intelligent digital applications to serve as location decision tools for retail property stakeholders around property investment, development, planning and strategic management of UK city centre and the highstreets. The project argues that (1) a rethink of the distribution of physical retail spaces would be appropriate to ensure various classes of retail real estates are optimally positioned in locations that best meet the need of teaming retail consumers; (2) stakeholders’ decisions pertaining to retail real estate should be guided by retail consumers choices to attain retail location optimisation and success in retail real estate’ businesses. 

The proposed research will rest on three broad underlying principles namely: (1) retail consumers (directly or indirectly) controls the retail property markets and retail property location performance (Adebayo et al, 2019); (2) spatial behaviour of retail consumers can be scientifically assessed and scored based on interconnectedness of streets (that is, space syntax theory) (Hillier and Hanson, 1989); (3) historical data on retail property performance could help in predicting retail location performance and resilience index using predictive machine learning models.

You must have a good honours degree (1st Class or an Upper 2:1) and/or a Masters degree in a relevant subject (completed with Distinction or a High Merit), and a desire to pursue a PhD. Studentships will be awarded to the strongest applications assessed on the applicant’s academic excellence, the strength of the research proposal and how the proposal fits with the research project/area identified.

The studentship lasts for 3 years and includes Full Home tuition fees plus a stipend of £18,622 – 2023/24 rate).  The studentship will only fully fund those applicants who are eligible for Home fees with relevant qualifications.  Applicants normally required to cover International fees will have to cover the difference between the Home fee (home fee approximately £4,596 at current rates) and the International tuition fee rates (international fee approximately £14,000 per annum at current rates).

How to Apply

To apply, please visit: How to apply – Leeds Beckett University

Please make sure that you complete the application form in full.  Submit your application form to: researchadmissions@leedsbeckett.ac.uk, along with:

  • Your research proposal (Max 4 pages). This should show how you will address and research the project area for which you are applying. Indicatively it should include discussion of research context, research questions, research methods, and the potential significance/originality of the PhD.
  • Scanned copies of your degree certificates (e.g. undergraduate or masters degree certificates) and transcripts.
  • Scanned copies of your English language requirements.
  • Scanned copies of your passport and previous UK visas if applicable.

Please state clearly that you are applying for School of Built Environment, Engineering and Computing studentships and the name of the studentship project area on your application form and research proposal.

For more information on the admissions process generally, please visit: How to apply – Leeds Beckett University

The closing date for applications is 7th June 2023.  Shortlisted candidates will be invited for the interview week commencing 4th July.  We regret that we may not be able to respond to all applications.  Applicants who have not received a response within four weeks of the closing date should consider their application has been unsuccessful on this occasion.


POSITION TYPE

ORGANIZATION TYPE

EXPERIENCE-LEVEL

DEGREE REQUIRED

You ad could be here!