Applied measurement uncertainty and knowledge embedding
The GUM (Guide to the expression of measurement uncertainty) is a standard for statistical methods used for obtaining a quantifiable statement of the uncertainty of a measurement. The GUM method utilizes the central limit theorem, to statistically combine all the contributing factors of a measurement in a t-distribution, used to express where the true value is within a confidence interval. It is only through a statement of uncertainty that traceability can be achieved as stated in the ISO standard 17025:2017. On an industrial research project, this method has been implemented on instruments measuring metal packaging for food and beverages. This facilitated understanding the performance of the measurement under different conditions, with varying environmental factors and measurement tasks. Furthermore, this method gave a better indication of design considerations that should be targeted to improve the performance of the measurement. The approach has been applied on some instruments with geometric features measured and is controlled with a tolerance ranging from +/-0.003mm to +/-0.300mm. The uncertainty approach created a better understanding of the error propagation and allowed for an accepted and internationally recognized quantification of the performance of the measurements, giving better confidence in the results. In addition, a significant part of the project was dedicated to embedding knowledge of measurement uncertainty within the engineering team at Sencon and creating a software tool to help create an uncertainty budget for new or changed designs.
Ahmed Khafaga is a KTP associate with experience in dimensional metrology and measurement uncertainty. He is currently leading a Knowledge Transfer Partnership project funded by Innovate UK. The partnership is between Coventry University and the Sencon UK Ltd, the latter is a manufacturer of gauging systems for the food and beverage metal packaging industry, and the project’s goal is to create a methodology for quantifying measurement uncertainty across the product range. Ahmed has graduated from Southampton University with a Master’s degree in Mechatronics engineering in 2017 and has obtained his Bachelor’s degree in Electrical and Electronics Engineering from the University of Sussex in 2016.