Title: Designing optimal experiments Authors:  Kalliopi Mylona - King's College London (United Kingdom) [presenting]
Abstract: Experimenters have benefited from the statistical methodology for nearly a century by being able to draw accurate and precise conclusions from experimental data at a low cost. However, this approach has evolved mainly in a world where linear models are used to examine the effects of a few factors on a single response variable. The ``big data'' revolution in the industry, the life sciences, and the physical sciences, facilitated by modern instrumentation for automatically collecting data and conducting experiments, limits the relevance of much of this methodology. To get the most out of the elaborate statistical and mathematical modelling used for the data analysis and the specially designed experiments, new approaches to these problems are required. Novel tools are presented for the design of complex experimental scenarios.