Set better control limits, detect shifts, reduce defects. Compare X̄‑R, X̄‑S, or I‑MR in seconds. Download reports for audits, teams, and executive reviews today.
| Subgroup | Measurements | Mean (X̄) | Range (R) |
|---|---|---|---|
| 1 | 10.1, 9.9, 10.2, 10.0, 10.1 | 10.06 | 0.30 |
| 2 | 10.0, 10.1, 9.8, 10.2, 10.0 | 10.02 | 0.40 |
| 3 | 10.3, 10.1, 10.2, 10.4, 10.2 | 10.24 | 0.30 |
| 4 | 10.0, 10.1, 10.0, 9.9, 10.0 | 10.00 | 0.20 |
Control limits separate routine variation from special‑cause signals. In stable processes, about 99.73% of points should fall inside ±3σ limits. When a point crosses a limit or patterns cluster near a boundary, the probability of a non‑random shift rises. Using limits prevents over‑adjustment and supports consistent, evidence‑based decisions.
Use X̄‑R when measurements are grouped into small subgroups (typically 2–10) and ranges are easy to compute. Use X̄‑S when subgroup sizes are larger or when standard deviation better represents spread. Use Individuals/Moving Range for one‑at‑a‑time readings such as automated sensors or low‑volume inspection.
The calculator reports the center line, upper control limit, and lower control limit for the selected chart. For subgroup charts it also reports dispersion limits (R or S charts). It counts points beyond limits and highlights the first index that violates a rule. These outputs provide a quick health check of process stability.
For X̄‑R, UCLx = X̄̄ + A2·R̄ and LCLx = X̄̄ − A2·R̄. Range limits are UCLr = D4·R̄ and LCLr = D3·R̄. For X̄‑S, UCLx = X̄̄ + A3·S̄ and LCLx = X̄̄ − A3·S̄ with UCLs = B4·S̄ and LCLs = B3·S̄. For I‑MR, σ ≈ MR̄/1.128 and limits are X̄ ± 3σ.
The graph overlays your series with the three control lines. A healthy process shows random scatter around the center, with no sustained runs, cycles, or hugging of a limit. If you see multiple consecutive points on one side of the center line, investigate recent changes in materials, settings, operators, or measurement method.
Start with containment: verify the measurement system and confirm the data timestamp. Next, document suspected causes and test them with short experiments. When you remove a special cause, recalculate limits using the new stable window so the chart reflects the improved capability. Keep limits unchanged during investigations to avoid hiding signals. Track results weekly and share learnings widely.
No. Control limits reflect process variation over time. Specification limits come from customer or engineering requirements. A stable process can still be off‑spec, and an in‑spec process can still be unstable.
A practical minimum is 20–25 subgroups for X̄‑R/X̄‑S, or 100+ individual points for I‑MR when possible. Fewer points can work, but limits may be noisy and should be reviewed.
Subgroup charts assume equal subgroup sizes. Partial subgroups distort ranges and standard deviations, so the calculator uses only full subgroups to keep the constants and limits valid.
Confirm the measurement and timestamp, then look for a special cause: tool wear, setup change, material batch, operator shift, or environment. Document the cause and corrective action before recalculating limits.
Recalculate after a verified process change or after removing a special cause and establishing a new stable baseline. Avoid frequent recalculation during investigation, because it can hide real signals.
I‑MR is often best because readings are sequential and subgrouping is artificial. If you can form rational subgroups by time window or lot, X̄‑R/X̄‑S may be appropriate for added sensitivity.
Important Note: All the Calculators listed in this site are for educational purpose only and we do not guarentee the accuracy of results. Please do consult with other sources as well.