Noise is unexplained variability or any influence that prevents us from attaining identical measured values, or identical results, on repeated versions of an experiment. Whilst typically this may be due to the lack of control of a process, it is important when conducting Design of Experiments (DOE) analysis to take into consideration this noise. There are two key noise types to consider, experimental and observational.