There is always variation. There always will be. Variation is concerned with differences.

Differences between people; between products; between processes; between periods of time. The question is: what are the differences telling us?

Knowledge of variation helps us understand the losses. Losses from the failure to distinguish between special causes and common causes. Losses from tampering increases variation.

Knowledge of variation helps us understand the theory of the loss function. Reducing variation drives costs out of a system.

Capability charts are charts of data over time. The control limits (upper and lower) show us how much variation can be expected from a system or process – they define what can be predicted about future performance. Control limits are calculated from the variation between the data points.

There are two things that matter: how to choose what to measure and how to turn measures into capability charts. When you are ready to make a chart, these things are explained in detail in the Vanguard guide to capability charts.