Translating Uncertain Knowledge into a Solvable Mathematical Model for Process Optimisation 

ELIGIBILITY: This project is for a student taking the MSc in Statistics and Sustainability. The student must possess knowledge of numerical optimisation methods and linear programming techniques.

This project aims to bridge the gap between human language and mathematical optimisation. By developing a method that translates natural language descriptions of process optimisation goals and constraints into a formal mathematical framework. This approach will enable users to leverage their domain expertise without requiring deep knowledge of optimisation techniques. We seek to automate the process of resource allocation for sustainability and health objectives.

Large language models will be used to translate human language into a formal language for optimisation. Since human knowledge is often uncertain, one of the goal of the projects is to model and include this uncertainty (via probabilistic statements) into the optimisation process.