This project aims to provide a comprehensive probabilistic framework and toolbox for decision-making on the basis of measurements, building on the techniques developed for Bayesian analysis and inference.

Decisions are made on the basis of measurements, e.g. whether to accept or reject a part, which course of treatment to apply, whether or not a pollutant exceeds specification limits. Due to measurement uncertainty, there is a risk that the wrong decision could be made. The probabilistic tools and techniques to be developed in this project will allow the expected costs of wrong decisions to be evaluated and subsequently balanced against the cost of providing more or more accurate measurement information. This project will develop a methodology and toolbox for decision-making on the basis of measurements. Applications include co-ordinate metrology in the control of manufacturing processes and product conformance and analytical chemistry in environmental monitoring.