Research and technological development can be a process of guided learning through experiments. The purpose of statistical methods is to make that process as efficient as possible. Optimal experimental design is a field of statistics that enables data to be collected as effectively as possible to inform objectives such as model selection, estimation of unknown model parameters and accurate predictions for unobserved treatments. An optimal design is efficient in achieving such objectives by being economic in the number of samples and the experimental costs.  Experimental design methods have very broad application across all the medical, physical and social sciences and technology. The increasingly complex systems from these areas present new challenges. They require realistic statistical models (probabilistic descriptions) which may incorporate hidden or auxiliary variables in hierarchies or strata (multi-level models) and may involve a large numbers of variables and limited experimental resources. Thus, careful consideration about the experimental design is needed when planning costly experiments to collect data on complex systems.