Constructing an experimental design by colouring vertices

Abstract

Factorial designs are used to plan experiments where there are a large number of factors that can affect a response of interest. Often, the number of observations that are to be collected means that observations are collected over a number of days, with days being treated as "blocks". In this talk, we give an introduction on how an experiment based on a blocked factorial design is constructed. Further, we demonstrate how designs with selected interactions can be constructed via proper vertex colourings in graph theory.