Accelerating Chemical Process Development via Uncertainty-Aware Modelling and Experimental Design

A key challenge in scale-up is uncertainty, as models may not transfer reliably when physical phenomena or parameters change. This project will use Bayesian methods to quantify uncertainty and support reliable decision-making.

Another important question is what data to collect to build the models. Multiple sources of data are usually available to support scale-up: experimental data at lab- or pilot-scale under different operating conditions, and potentially also data from in silico experiments on high fidelity simulation models. These sources have differing costs: experiments at lab-scale are cheaper than a small-scale pilot plant, and high-fidelity model runs, while expensive, are significantly cheaper than experiments on a full-scale process. We aim to address this decision problem using Bayesian design of experiments ([3],[4]), which offers a principled machine learning framework for making data collection choices under uncertainty. However, most existing Bayesian design methods optimize a utility function related to parameter learning for a single model under equal costs. Here both model structure uncertainty and costs must be accounted for, and a more relevant utility would be linked to the ultimate goals of accurate prediction and successful plant design.

This project will use liquid–liquid extraction as a case study to develop and evaluate the proposed approach. We will:

1. develop a framework for quantifying model uncertainty during scale-up, incorporating chemical engineering knowledge into suitable prior distributions over model structures and parameter values structures, for example the dominant transport phenomena, choices of constitutive equations, centring prior distributions on plausible parameter values from established rules of thumb, etc.;
2. develop Bayesian decision-making methods to support data collection decisions, including choices regarding experimental scale and the trade-off between physical and simulation-based experimentation;
3. develop utility functions and decision criteria that target the ultimate goals of accurate large-scale prediction and confidence in process design decisions;
4. illustrate the methodology through a liquid-liquid extraction case study, targeting application areas in such as in pharmaceutical continuous downstream purification or recovery of critical metals from electronic waste streams.

The joint supervision team spans the Schools of Natural Sciences and Engineering. The main supervisor Dr Tim Waite (Mathematics) will guide the statistical experimental design and uncertainty quantification. Co-supervisor Dr Jiyizhe Zhang (Chemical Engineering) will provide guidance on chemical process modelling and expertise on application of AI methods in Chemical Engineering. Co-supervisor Professor Sami Kaski brings expertise in Bayesian inference, AI for decision making, and experimental design. The student will receive training in both ML/UQ and chemical process modelling. The collaboration will also provide a foundation for future EPSRC funding applications.

Desirable Student Background:

A good undergraduate degree (Master’s preferred), in Mathematics, Statistics, Computer Science, Data Science, or Chemical Engineering with experience in machine learning and/or Bayesian methods and computation. Experience with chemical process models would be helpful but is not necessary: training in this can be provided, and this aspect will be supported via collaboration with Dr Zhang’s group.

How to apply

Please apply under the University of Manchester application portal and select 'PhD in Artificial Intelligence' (OAA Application portal).

When applying, you’ll need to specify the full name of this project, the name of your supervisor, if you already having funding or if you wish to be considered for available funding through the university, details of your previous study, and names and contact details of two referees.

Your application will not be processed without all of the required documents submitted at the time of application, and we cannot accept responsibility for late or missed deadlines. Incomplete applications will not be considered.

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
• 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).
• Contact details for two referees (Please ensure referee email addresses are official university or workplace accounts, as we may need to verify the references.)
• English Language Certificate (if applicable)

We actively encourage applicants from diverse career paths and backgrounds and from all sections of the community, regardless of age, disability, ethnicity, gender, gender expression, sexual orientation and transgender status.

We also support applications from those returning from a career break or other roles. We are dedicated to supporting work-life balance and offer flexible working arrangements to accommodate individual needs. Our selection process is free from bias, and we are committed to ensuring fair and equal opportunities for all applicants.

We are dedicated to supporting work-life balance and offer flexible working arrangements to accommodate individual needs, including a part-time study option.

Join us and help grow the UK’s national capability in artificial intelligence, part of the UK’s modern industrial strategy. Everyone is welcome if you are passionate about uncertainty quantification, decision making with humans in the loop and decision making for machine learning systems and their applications in Science and Engineering, and looking to develop your research and innovation skills at the doctoral level to further your career in the AI sector.

Equality, diversity and inclusion is fundamental to the success of The University of Manchester, and is at the heart of all of our activities. We know that diversity strengthens our research community, leading to enhanced research creativity, productivity and quality, and societal and economic impact (Equality, diversity and inclusion | The University of Manchester).

We actively encourage applicants from diverse career paths and backgrounds and from all sections of the community, regardless of age, disability, ethnicity, gender, gender expression, sexual orientation and transgender status.

Funding Notes

Fully funded AI UKRI CDT 4 year program; The Project will be based in University of Manchester.

The UKRI AI CDT in Decision Making for Complex Systems is part of the TechFirst offering stipends of £30,000 for students eligible for home fee status. The start date is September 2027.

References

[1] JF Laub, J Zhang*, M Heyer, A Lapkin*. Automated Generation of Mechanistic Models for Chemical Process Digital Twins using Reinforcement Learning-Part II: Compartmentalization and Learning-Based Recalibration Computers & Chemical Engineering, 2025, 204, 109384.
https://www.sciencedirect.com/science/article/pii/S0098135425003874

[2] X Liang, Y Xie, J Zhang, W Huang, B Pan, XY Tai, J Xuan, A Lapkin, H Wang*. Autonomous in silico optimization framework for high-performance micromixers Chemical Engineering Journal, 2026, 529, 172957.
https://www.sciencedirect.com/science/article/pii/S1385894726004146

[3] Rainforth, T., Foster, A., Ivanova, D.R. and Bickford Smith, F., 2024. Modern Bayesian experimental design. Statistical Science, 39(1), pp.100-114.
https://projecteuclid.org/journals/statistical-science/volume-39/issue-1/Modern-Bayesian-Experimental-Design/10.1214/23-STS915.short

[4] Englezou, Y., Waite, T.W. and Woods, D.C. (2022) ‘Approximate Laplace Importance Sampling for the estimation of expected Shannon information gain in high-dimensional Bayesian design for nonlinear models’, Statistics and Computing, 32:82
https://link.springer.com/article/10.1007/s11222-022-10159-2

[5] de León, H.M., Straathof, A. and Haringa, C., 2025. Dynamic compartment models: Towards a rapid modeling approach for fed-batch fermentations. Chemical Engineering Science, 308, p.121396.
https://www.sciencedirect.com/science/article/pii/S0009250925002192?via%3Dihub

Supervisory team links:
https://timwaite.github.io
https://jiyizhezhang.github.io
https://kaski-lab.com