Statistical Process Control with Rational Subgroups
The X-bar and R control chart is the foundational tool for continuous dimensional data in Statistical Process Control (ISO 7870-2 / AIAG SPC-3). It simultaneously tracks two fundamental aspects of any production process:
- Process Location / Central Tendency (X-bar Chart): Detects process shifts, drift, tool wear, and batch differences.
- Process Dispersion / Short-Term Variation (R Chart): Evaluates piece-to-piece consistency within each rational subgroup. Always evaluate the R chart first; if dispersion is unstable, X-bar limits are unreliable.
Standard SPC Equations & Constants
Subgroup Mean: X̄ᵢ = (x₁ + x₂ + ... + xₙ) / n
Subgroup Range: Rᵢ = Max(xᵢ) - Min(xᵢ)
Grand Mean: X̄̄ = Σ X̄ᵢ / k
Average Range: R̄ = Σ Rᵢ / k
Within-Subgroup Sigma Estimate: σ̂ = R̄ / d₂(n)
X-Bar Limit Factor: A₂ = 3 / (d₂(n) × √n)
Frequently Asked Questions
When should I use X-bar & S instead of X-bar & R?
When subgroup sample size n > 8 or 10, the subgroup sample standard deviation S is more statistically efficient than the sample range R. For n ≤ 5, the range R loses negligible statistical efficiency and is far simpler for line operators to compute.