SPCVisualsPaste your data, get a control chart
Paste a column of numbers, or rows of subgroups, and get a real control chart: limits from the data, Nelson rules 1–4 evaluated, out-of-control points marked.
How you make a control chart in Excel
Excel has no control chart type, so you build one from a line chart: your readings in one column, then three helper columns holding the centre line, the UCL and the LCL repeated down every single row, and all four plotted as series. Format the three limit series as flat lines with no markers and it looks right.
It looks right, and then it quietly stops being right. If those helper columns are formulas over the growing data column — the natural way to build it — every new reading recalculates the limits. An out-of-control point drags the limits out until it is back inside them, so the chart erases the signal it exists to raise. Limits have to be frozen at a baseline, and a spreadsheet gives you no way to say that is what they are.
Nothing you paste leaves this calculation. It is not stored, not logged and not sent anywhere else — the numbers are computed and returned, and that is all. Paste straight out of a spreadsheet — tab-separated columns are understood as they land.
What the chart is telling you
The centre line is the process average and the outer lines are three sigma either side of it, computed from short-term variation. A point outside them is unlikely enough to be worth investigating. So are several patterns that stay inside them, which is what the run rules catch.
Rules evaluated here
- 1 — beyond 3σ. One point past a control limit.
- 2 — shift. Nine in a row on the same side of the centre line.
- 3 — trend. Six in a row rising, or six falling.
- 4 — alternating. Fourteen in a row zig-zagging up and down.
Rules 5 to 8 — the zone rules — are deliberately not applied. They assume the data is roughly normal and that consecutive readings are independent, and a pasted sample cannot be assumed to be either. Firing them here would paint red points on somebody's data that mean nothing. In the product you switch them on knowingly, per characteristic.
The part most people get wrong
These limits are computed from this data, which is exactly what you want for a one-off look and exactly what you do not want on a live chart. On a real process you lock limits to a stable baseline window, so that later drift shows up as drift instead of quietly widening the limits until nothing ever signals again.
Specification limits are not control limits
This is the most repeated argument in quality, so here it is with your own numbers: enter a tolerance above and tick draw the spec limits.
- Control limits come from the process: what it does when nothing unusual is happening. They are computed, never chosen.
- Specification limits come from the drawing: what the customer will accept. They are chosen, never computed.
A process can be perfectly in control and entirely incapable, or capable and wildly out of control. They answer different questions.
And on a subgrouped chart the two must not share an axis at all. The plotted points are subgroup averages, and averages vary less than the parts they came from by a factor of √n. So a chart whose every point sits comfortably inside the tolerance can be produced by a process making parts outside it — the tool says so explicitly when your data does this. That is why putting spec limits on an X̄ chart is not a stylistic preference: it hides the very thing you needed to see.
What the Excel version will not do
- Recalculating limits hide the thing you are looking for
- A special cause inflates R̄, which widens the limits, which swallows the special cause. The worse the excursion, the more effectively the chart conceals it.
- The run rules are eight columns of nested IFs
- Nine points on one side, six increasing, two of three beyond two sigma, fourteen alternating — each is its own formula, and every one of them breaks silently the day somebody inserts a row in the middle.
- Nobody maintains it after its author leaves
- The file works until it needs a new characteristic, a changed subgroup size, or a re-baseline after a tool change. Then it belongs to whoever can read the formulas, and usually that is nobody.
- It cannot tell anyone
- A chart in a workbook on somebody’s desktop raises no alarm at two in the morning. The point trips, and it gets found on Thursday.
Questions people ask about this
Which control chart should I use?
One reading at a time gives an I-MR chart. Several parts measured together give an X̄-R chart, or X̄-s once the subgroups are large enough that the standard deviation carries more information than the range. Counts of defects in equal-sized batches give a c-chart. Leave the selector on "detect from the data" and the tool picks from the shape of what you paste.
Can I put specification limits on a control chart?
On an individuals chart it is merely unhelpful. On a subgrouped chart it is actively misleading: the plotted points are averages, which vary √n less than the parts they came from, so every point can sit inside the tolerance while the process makes out-of-spec parts. Enter a tolerance here and the tool says so explicitly when your data does that.
How many points do I need for a control chart?
Twenty to twenty-five subgroups is the usual guidance for establishing limits. Fewer will draw a chart, but the limits carry enough uncertainty that both false alarms and missed signals become likely.
Is my data in control?
The chart marks any point outside the limits and any run pattern that fired. Bear in mind these limits were computed from the same data you pasted, which is right for a one-off look and wrong for monitoring: on a live process you freeze limits to a stable baseline so later drift shows up as drift.
Excel drew it once. Then the line ran another shift.
Same readings, except the limits stay frozen where you locked them, every run rule is evaluated on every new point, the chart is on a screen on the floor rather than in a file on a laptop, and somebody gets told when a rule trips.
More control charts tools
- Control limit calculator — Give it subgroups or individual readings and it returns the control limits for the chart, the range or sigma chart beneath it, and every constant it used to get there.
- Nelson & Western Electric rules checker — Every run rule evaluated on your data, each violation named in plain English with what it usually indicates — a shift, a trend, tool wear, two machines mixed.
- X̄-R chart calculator — Subgroups in, X̄ and R charts out — limits from A₂, D₃ and D₄, run rules evaluated, and the range chart shown first because it decides whether the averages chart can be trusted.
- X̄-s chart calculator — For subgroups big enough that the range wastes them. Limits from A₃, B₃ and B₄, sigma recovered with the c₄ correction, and the s chart read first because it decides whether the averages chart can be trusted.
- I-MR chart calculator — For processes that give you one number at a time — a batch, an oven, a destructive test. Limits from the mean moving range, run rules evaluated, and an honest note about what an individuals chart cannot see.
- p, np, u and c chart calculator — Counted data rather than measured: proportion defective, number defective, defects per unit. Limits that step with the sample size instead of pretending it never changed, and an honest warning when the counts are too low for three-sigma limits to mean anything.
Everything else
- Cp / Cpk calculator — Paste measurements — or type a mean and a sigma — with your tolerance, and get Cp, Cpk, Pp, Ppk, the sigma level and the expected parts per million out of spec.
- Cpk confidence interval & sample size — A Cpk of 1.33 from 30 pieces has a 95% interval of roughly 0.97 to 1.69. See the uncertainty in your own number, and how many parts would settle it.
- Cpk ↔ PPM ↔ sigma level converter — What PPM is a Cpk of 1.33? What Cpk does 3.4 PPM need? Both conventions shown side by side, because the 1.5 sigma shift is why two sources disagree by a factor of ten.
- Control chart constants — The whole table, n = 2 to 25, with the formula each constant belongs to. The same values the charts on this site are computed from.
- Levey-Jennings chart with Westgard rules — Paste your QC results with the assigned mean and SD from the package insert, and get the chart with every sigma band drawn and all six Westgard rules evaluated — each one named, and labelled random or systematic error.
- Bland-Altman plot — Paired readings from two methods, plotted as difference against average — with the bias, the limits of agreement, their confidence intervals, and a test for whether the disagreement depends on the magnitude.
- Pareto chart generator — Categories and counts in, ranked bars and the cumulative line out — with the vital few named, an "Other" bucket that always sorts last, and an honest warning when the distribution is flat and there is no dominant cause to attack.
- Box and whisker plot generator — One column per group, boxes side by side. Quartiles, the 1.5×IQR fences, whiskers that stop at real readings and outliers drawn individually — with the quartile method stated, because that is why your plot and Excel's disagree.
- Histogram and normality test — A histogram with the fitted normal curve, both standard bin rules with a reasoned recommendation, and an Anderson-Darling test that refuses to tell you your data is normal — because no test can.
- Gage R&R calculator (ANOVA) — Paste a crossed study — part, operator, reading — and get the full ANOVA: repeatability and reproducibility separated, the part-by-operator interaction tested rather than assumed away, %GRR, %Tolerance and ndc against AIAG's bands.
- Grubbs outlier test — The G statistic against its critical value, the p-value, and the point implicated — plus a straight answer about what to do next, which is almost never to delete the reading.
- Scatter diagram and correlation — Paired readings plotted, with Pearson r, Spearman rho, r², the least-squares line and its p-value — and the two reported together, because when they disagree the disagreement is the finding.
- Standard deviation calculator — The usual summary statistics, and the thing a spreadsheet will not tell you: the within-subgroup sigma a control chart uses, beside the overall sigma a textbook means. The gap between them is why Cpk and Ppk disagree.