Physics-Informed, Explainable, and Uncertainty-Aware AI for polymer extrusion modelling

About the Project

Polymer extrusion is a fundamental stage in polymer processing, producing polymeric materials for a wide range of applications including packaging, construction, and healthcare. Yet, controlling the complex thermomechanical behaviour of polymer melts remains a major industrial challenge. Conventional monitoring tools and existing AI-based soft sensors can measure/predict key melt properties and process parameters with high accuracy, but they typically offer limited physical interpretability, uncertainty quantification, generalisation, and scalability. As a result, their deployment in large-scale industrial extrusion lines remains constrained.

This PhD project aims to develop the next generation of explainable, physics-informed, and uncertainty-aware AI models for polymer extrusion that are scalable across different materials, screw geometries, machine sizes, and operating conditions. The models should have the desired accuracy and computational feasibility required for real-time estimation of difficult-to-measure extrusion parameters and integrating into real-time optimisation frameworks. Building on recent advances in soft sensing and adaptive learning, the research will embed first-principles knowledge, such as conservation laws of mass, momentum, and energy, and non-Newtonian polymer rheology, directly into deep learning architectures. These physics-informed neural networks (PINNs) and hybrid grey-box models will ensure physically consistent, data-efficient, and generalisable predictions across different processing conditions.

To enhance trust and interpretability, the project will integrate explainable AI (XAI) techniques to reveal how process variables, material properties, etc., influence melt temperature, pressure, viscosity, and other relevant parameters. Furthermore, uncertainty quantification methods will allow the models to express confidence in their predictions, enabling more reliable, risk-aware decision-making and safe process optimisation in real-time operation.

A major focus will be upscaling to industrial extrusion systems. Using transfer learning, domain adaptation, and similarity-based feature engineering, the developed models will be tested and validated across laboratory, pilot, and production-scale extruders in collaboration with industrial partners. The ultimate goal is to deliver generalised, interpretable, and uncertainty-aware AI models that maintain physical fidelity, predictive confidence, and robustness across diverse processing environments.

Eligibility

Students with a First class/2.1 degree in Materials Science, Physics, Mathematics or an aligned Engineering subject are encouraged to apply. An MSc in a related filed would also be acceptable. Experience in developing machine learning models using MATLAB and/or Python TensorFlow/Pytorch would be preferable. Also, an MSc and/or publications in a relevant field should also be useful.

Self-funded candidates can also apply.

Funding

This is a 3.5-year PhD. Excellent candidates will be nominated for competence-based faculty funding (application deadline for faculty funding is 19th December 2025. If no suitable applications are received in December, a further round of applications will be considered for the deadline 13th March 2026).

For more information, visit our funding page or search our funding database for specific scholarships, studentships and awards you may be eligible for.

The start date for funded students is October 2026. The start date for self funded students is negotiable.

We recommend that you apply early as the advert may be removed if the position is filled.

Before you apply

We strongly recommend that you contact the main supervisor for this project (Dr Abeykoon - chamil.abeykoon@manchester.ac.uk) before you apply. Please include details of your current level of study, academic background and any relevant experience and include a paragraph about your motivation to study this PhD project.

How to apply

You will need to submit an online application through our website here: https://uom.link/pgr-apply

When you apply, you will be asked to upload the following supporting documents:

• Final Transcript and certificates of all awarded university level qualifications
• Interim Transcript of any university level qualifications in progress
• CV
• You will be asked to supply contact details for two referees on the application form (please make sure that the contact email you provide is an official university/ work email address as we may need to verify the reference)
• Supporting statement: A one or two page statement outlining your motivation to pursue postgraduate research and why you want to undertake postgraduate research at Manchester, any relevant research or work experience, the key findings of your previous research experience, and techniques and skills you’ve developed. (This is mandatory for all applicants and the application will be put on hold without it.
• English Language certificate (if applicable). If you require an English qualification to study in the UK, you can apply now and send this in at a later date.

If you have any queries regarding making an application please contact our admissions team FSE.doctoralacademy.admissions@manchester.ac.uk