To be bell-shaped with a distribution means that most of the values are in the center, and fewer and fewer values tail out to either side symmetrically.
Use: If you look at a frequency distribution or a histogram and it is bell-shaped, you know that no one, or more than one factors have unduly influenced the process. You may actually be wanting to see a distribution that is not bell-shaped because if it is not bell-shaped, you know that you have assignable-cause variation. Think of assignable-cause variation as a ‘smoking gun’ for an opportunity for improvement.
