Trustworthy AI for Nuclear Decommissioning Knowledge Retention (Ref: FP-SJ-2026-2R)

Funded PhD Project (UK Students Only)

About the Project

The UK’s nuclear decommissioning programme faces a critical knowledge challenge. Decades of operations have produced vast archives of documents, drawings, and records—many only recently digitised. These materials are often fragmented and difficult to interpret without the expertise of staff who are ageing and retiring. This loss of tacit knowledge threatens safety, efficiency, and the ability to train future specialists.

This EngD project will explore how existing AI technologies, such as large language models, can be adapted for nuclear decommissioning knowledge retrieval while addressing their current limitations: lack of transparency, traceability, and engineering context. The researcher will develop methods to combine AI with Model-Based Systems Engineering (MBSE) principles, creating a structured backbone that links AI-generated insights to plant systems, processes, and original data sources. This approach will make AI outputs more verifiable, auditable, and trustworthy.

The outcome will be a proof-of-concept platform tested on real decommissioning data, demonstrating how AI and MBSE can work together to preserve knowledge, support safe decommissioning, and accelerate workforce training. This research will help transform decades of legacy information into a reliable digital resource for the UK’s nuclear future.

Name of primary supervisor/CDT lead:
Siyuan Ji s.ji@lboro.ac.uk

Name of secondary supervisor:
Wenheng Zhang

https://www.lboro.ac.uk/research/mbse/
https://www.lboro.ac.uk/schools/meme/research-and-innovation/research-groups/systems-engineering-and-complexity/

Entry requirements:
Applicants should have, or expect to achieve, at least a 2:1 honours degree (or equivalent) in relevant subject (e.g., computer science, data science, mathematics, physics, electronic engineering). A relevant master's degree and/or experience is desirable, but not essential. The ideal candidate will have working knowledge in computer science, data science, with programming experience in Python and a passion in artificial intelligence and engineering. Knowledge in MBSE is not essential as the candidate will receive formal training in this area during their first year.

English language requirements:
Applicants must meet the minimum English language requirements. Further details are available on the International website (http://www.lboro.ac.uk/international/applicants/english/).

Funding information:
The studentship is for 4 years and provides a tax-free stipend of about £22,780 per annum for the duration of the studentship plus tuition fees at the UK rate. Due to funding restrictions, this is only available to those eligible for UK fees.

Bench fees required: No

Closing date of advert: 25 June 2026

Start date: October 2026

Full-time/part-time availability: Full-time 4 years

Who is eligible to apply?: UK Only

Project search terms:
Machine Learning, Systems Engineering, Large Language Model, Trustworthy AI, Information Retrieval, Model-Based Systems Engineering, Nuclear Decommission

Email address Wolfson:
ws.phdadmin@lboro.ac.uk