Control Chart Selector, Shewhart Limits & Detection Power

The formula

This is the expression the tool evaluates. Every term is named underneath, with the unit it must be supplied in.

How the result is derived

Step by step, from the values you type to the figure on screen.

1. The 7 inputs are read from the form on every keystroke: Grand Average (X double bar), Average Subgroup Range (R bar), Standard Deviation of All Readings, Subgroup Size (n), Number of Subgroups (k), Limit Multiplier and Shift to Detect.
2. Each value is checked against the accepted range in the input table below. A value outside its range stops the calculation rather than producing a misleading figure — the results blank out and a message appears.
3. The validated values are substituted into the expression above, which resolves Upper Control Limit, Mean Chart together with every supporting figure in one pass — no value is carried over from a previous entry.
4. The supporting outputs — Lower Control Limit, Mean Chart, Upper Limit, Dispersion Chart, Lower Limit, Dispersion Chart, Within-Subgroup Sigma, Chart Demanded (1 = I-MR, 2 = Xbar-R, 3 = Xbar-s), Subgroups to Detect the Stated Shift, Limit Inflation if Overall SD is Used and Standard Error of the Sigma Estimate — come from the same pass, so they always describe the same case as the headline figure.
5. Results are rounded for display only. The full-precision value is used throughout the chain, so reading a rounded intermediate figure back into the tool by hand can shift the last digit.

What each input means

Where to read each value on the floor, the unit it must be in, and the range the tool accepts.

What the tool returns

The headline figure and every supporting value it is built from.

Worked example

Substituting

The honest limits are 177.3467 to 182.6533. The shortcut limits, built on the standard deviation of all 125 readings, run from about 176.2 to 183.8 - and a roll sitting at 183 would plot comfortably inside them while being, by the process own within-subgroup behaviour, four sigma off target. The gap between the two is the between-subgroup drift, and it is roughly the same size as the within-subgroup noise: this process is not one process, it is a sequence of slightly different ones.

How to use it

1. Work through the input groups in order — Baseline Study, Subgrouping and Chart Design. The defaults are a realistic case, so you can change one value at a time and watch what moves.
2. There is no calculate button. Every figure recalculates as you type or drag, which is what makes this usable for a what-if sweep rather than a single answer.
3. Read Upper Control Limit, Mean Chart in the dark results panel — that is the headline figure, expressed in units.
4. Check the supporting rows underneath (Lower Control Limit, Mean Chart, Upper Limit, Dispersion Chart, Lower Limit, Dispersion Chart, Within-Subgroup Sigma, Chart Demanded (1 = I-MR, 2 = Xbar-R, 3 = Xbar-s), Subgroups to Detect the Stated Shift, Limit Inflation if Overall SD is Used and Standard Error of the Sigma Estimate) before acting on the headline — they are where an implausible input usually shows itself first.
5. Reset to defaults returns every field to the reference case, which is the quickest way to check whether a surprising result came from the tool or from an input you had changed earlier.

Where this is used

• Process planning — establishing Upper Control Limit, Mean Chart before a trial is booked, so machine time and material in Quality Systems, Traceability, Utilities & Factory Decisions are committed against a calculated figure rather than an estimate.
• Costing and quotation — Upper Control Limit, Mean Chart is an input to the cost sheet, and quoting from a worked number rather than a remembered one is what keeps a margin intact.
• Troubleshooting — when the floor result drifts from plan, entering the measured values (starting with Grand Average (X double bar)) shows how much of the gap in Upper Control Limit, Mean Chart each variable explains.
• Teaching and study — the accepted ranges bracket normal Quality Systems, Traceability, Utilities & Factory Decisions practice, so moving one variable at a time shows the shape of the relationship rather than a single answer.

Reading the result

Typical bands and what each one is telling you.

Assumptions and limits

• These are trial limits from a baseline study. ISO 7870-2 expects any subgroup with an assignable cause to be investigated and removed and the limits then recomputed, so the first pass is never the last. The reported standard error of sigma assumes the subgroup ranges are independent; for a subgroup size of 1 the moving ranges overlap by one reading and are correlated, so the true uncertainty is larger than the figure shown, and the individuals chart is the least trustworthy of the three for that reason among others. All the constants assume the individual readings are approximately normal and independent - autocorrelated data such as a continuously monitored dyebath temperature will produce limits that are far too narrow, and count data such as neps or defects per unit needs an attribute chart, not this one. Detection power is calculated for a sustained step change and for the beyond-limits rule alone; run rules change both the detection and the false alarm figures. A negative limit inflation means the overall standard deviation came out smaller than the within-subgroup estimate, which normally signals subgrouping that is not rational or too few subgroups to estimate either quantity reliably. Finally, nothing here is a capability index: control limits describe what the process does, specification limits describe what the customer asked for, and drawing the two on the same chart is the most common and most damaging mistake in factory SPC.
• Every input is bounded to the range normal practice occupies (Grand Average (X double bar) -100000 to 100000 units, Average Subgroup Range (R bar) 0.0001 to 100000 units and Standard Deviation of All Readings 0.0001 to 100000 units, and so on for the rest). Those bounds are guard rails against typing errors, not a claim that the formula fails one unit outside them.
• The calculation is deterministic: the same inputs always give the same result. It carries no allowance for machine condition, operator skill, ambient conditions or lot-to-lot material variation unless an input above explicitly represents one.
• Nothing is sent anywhere. The maths runs in your browser, so the numbers you type never leave the page.

Standards and further reading

• ISO 7870-1 - Control charts, Part 1: General guidelines, for chart selection and the role of rational subgrouping.
• ISO 7870-2 - Control charts, Part 2: Shewhart control charts, which tabulates d2, A2, D3, D4, B3 and B4 and asks for 20 to 25 subgroups before limits are fixed.
• ASTM E2587 - Standard Practice for Use of Control Charts in Statistical Process Control, for variables and attribute chart construction.
• ISO 22514-2 - Statistical methods in process management, Capability and performance, Part 2, for the within-subgroup against overall distinction that separates control limits from performance indices.

Questions people ask