Turn raw readings into clear process insights fast. Auto-calculate averages, ranges, and limits securely here. Download tables, PDFs, and charts for every review cycle.
| # | Value | Moving Range | Typical Use |
|---|---|---|---|
| 1 | 10.0200 | — | Start point |
| 2 | 10.0800 | 0.0600 | Variation between consecutive points |
| 3 | 9.9700 | 0.1100 | Variation between consecutive points |
| 4 | 10.1000 | 0.1300 | Variation between consecutive points |
| 5 | 10.0500 | 0.0500 | Variation between consecutive points |
| 6 | 10.1200 | 0.0700 | Variation between consecutive points |
| 7 | 9.9500 | 0.1700 | Variation between consecutive points |
| 8 | 10.0100 | 0.0600 | Variation between consecutive points |
| 9 | 10.0700 | 0.0600 | Variation between consecutive points |
| 10 | 10.0300 | 0.0400 | Variation between consecutive points |
| 11 | 9.9900 | 0.0400 | Variation between consecutive points |
| 12 | 10.1100 | 0.1200 | Variation between consecutive points |
An Individuals chart is used when you collect one value at a time, such as torque, fill weight, thickness, or cycle time. It fits low-volume work, destructive testing, and sensors that stream single readings. With 25–30 points, you gain a baseline for typical variation and can separate routine noise from special causes. With fewer than 10 points, limits are unstable, so treat conclusions as directional.
The Moving Range series uses consecutive differences, so it reacts quickly to sudden shifts between adjacent observations. A high MR can indicate a setup change, mixed materials, probe instability, or technique variation. For a stable process, most MR points sit below the MR UCL, and the average MR should not trend upward. If one MR spike appears, inspect the two values that created it before changing settings.
Control limits describe expected behavior, not customer requirements. The calculator flags points beyond UCL/LCL, runs of eight on one side of the center line, and six-point trends. Treat signals as prompts to investigate, not automatic rejection of product. Verify measurement system health and correlate events with maintenance, raw material lots, or environmental shifts. Record the suspected cause, action taken, and when stable conditions returned.
When you enter LSL and USL, the tool estimates within-sigma using MR̄/d2 and reports Cp and Cpk. Cpk near 1.00 means the process is close to the edge, while 1.33 or higher is often targeted in mature production. Low Cp with reasonable Cpk suggests the spread is too wide; low Cpk with higher Cp suggests the mean is off-center. Use capability only after stability is confirmed.
Use time-ordered data from one measurement system and one product family. Avoid mixing machines or gauges unless that mix is the process you control. Keep the sampling interval consistent and capture context: lot, operator, tool number, cavity, and ambient conditions. After improvements, start a new baseline and archive the prior chart to support audits and lessons learned. For regulated industries, save the exported report with date, version, and reviewer signatures for full traceability.
Use 25 to 30 time-ordered points when possible. With fewer than 10 points, limits can swing widely and signals may be misleading. If data are scarce, chart anyway but confirm findings with additional sampling.
MR highlights short-term jumps that can be hidden when the mean is stable. A single process upset often shows as a large moving range even if the Individuals points remain near the center line.
No. Specifications reflect customer requirements, while control limits reflect process behavior. Mixing them can cause over-adjustment or missed signals. Use specs only for Cp/Cpk, and use chart limits to judge stability.
It suggests a sustained shift in the process average. Investigate recent changes such as setup, tool wear, material lots, maintenance, or measurement system drift. Confirm the shift with context data before resetting limits.
Recalculate after you remove a confirmed special cause or after an intentional process change that permanently alters behavior. Keep the previous chart and limits as evidence of improvement and for audit traceability.
Cp/Cpk requires both LSL and USL and a valid within-sigma estimate. If you did not enter specs, entered equal limits, or MR̄ is near zero, capability cannot be computed reliably and is left blank.
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.