Variability Reduction in Manufacturing
About 1 min read
Even a well-run process produces some variation from one item to the next. A core goal of quality management is reducing variability, because consistent, predictable output is the essence of quality. This idea is central to the teaching of W. Edwards Deming and to later methods such as Six Sigma.
Questions you may have include:
- Why does variation matter?
- What is the difference between common and special causes?
- How do you reduce variation?
Variation is the enemy of consistency
Customers value products that perform the same way every time. The more output varies, the more likely some items fall outside acceptable limits and become defects. Hitting the target on average is not enough; the spread around that target must also be small.
Common and special causes
Quality pioneers distinguished two kinds of variation. Common-cause variation is the natural, built-in scatter of a stable process. Special-cause (assignable) variation comes from a specific, identifiable disturbance — a worn tool, a bad batch of material, a changed setting. The two require different responses: special causes are tracked down and removed, while reducing common-cause variation means improving the process itself.
Measuring and reducing variation
Control charts, introduced by Walter Shewhart, help tell the two kinds of variation apart by showing whether a process is stable or being disturbed. Once a process is stable, variation is reduced by standardizing methods, improving equipment, and refining the process — the systematic aim of approaches like Six Sigma, whose very name refers to keeping output far within acceptable limits.
Summary
Reducing variability makes output consistent and predictable, which is what quality really means. Separating common-cause from special-cause variation — often with control charts — lets you remove specific disturbances and then steadily tighten the process, the foundation of methods such as Six Sigma.