The reality in the world is that variation exists. If variation did not exist, every product would be right the first time. The same would hold for providing a service. But, the reality is that we live in a world where variation does exist and therefore every now and then we will have to deal with a defect, mistake, flaw, or error. Another reality is the variation can be random, meaning that no undue influence is coming from the people, the machinery, the materials, the methods, the measurements, or from the environment. If the variation in a process is random, it is at a state of normality. Other names for normality include common-cause, chance variation, and random variation.
Use: If we are trying to improve upon a defect rate, we have to understand the variation that is causing the defect. We have to be able to visualize the variation to attack it. The team needs to know whether the variation is assignable-cause variation – meaning that there is undue influence from one or more of the 6M’s (Man, Machine, Material, Method, Measurement, Mother Nature). That is the ‘smoking gun’ for root-cause analysis. The team needs to know whether it is unexpected variation, or whether it is common-cause variation. In other words, the team needs to know what they expect to see first; then, if they see something they don’t expect to see, that is their opportunity to find the root cause. Normality is what they expect to see.
Normality is able to be seen in the form of a frequency distribution or a histogram being bell-shaped. One can also perform sophisticated tests to check for normality. Such tests can be found using sophisticated software such as Minitab.
