Control ChartControl Chart Builder
SPC — I-MR, X-bar/R & p-charts
What is a Control Chart?
A control chart (SPC chart) plots a process metric over time against statistically derived control limits. It answers one question: is the variation you see just common-cause noise, or has something changed? Points inside the limits with no unusual patterns mean a stable, predictable process. Points outside — or non-random patterns — signal a special cause worth investigating.
A stable process, a sustained shift, hourly subgroups, and an attribute chart with one bad week.
Chart Type & Data
Only common-cause variation detected — the process is stable and predictable.
The within-subgroup sigma estimate is the one to carry into a Cp/Cpk capability study.
Out-of-Control Signals (0)
No rule violations. The three rules checked: a point beyond a limit, 8 consecutive points one side of center, and a 6-point trend.
Everything is editable — click any field.
Want an editable copy?
Download the free SPC template — enter your measurements and it computes the center line, 3-sigma control limits, and moving ranges, with the chart drawn automatically. Opens in Excel and Google Sheets.
What is a control chart?
A control chart plots a process metric in time order against a center line (the process average) and control limits set three standard deviations either side. Walter Shewhart's insight was that every process varies, but there are two kinds of variation: common-cause — the routine noise of the process — and special-cause — something specific changed. The chart tells you which one you are looking at, so you neither chase noise nor ignore a real change.
Control charts are the backbone of statistical process control (SPC) and the standard tool of the Control phase in a Six Sigma DMAIC project: after an improvement, the chart is what proves the gain holds.
Control chart vs. run chart
A run chart is just the data over time; a control chart adds statistically derived limits, which is what turns "that looks high" into "that is a signal." Reacting to individual points on a run chart — adjusting the process every time a number looks off — is called tampering, and it usually increases variation. The control limits tell you when acting is justified.
From stability to capability
A control chart answers "is the process stable?" — not "is it good enough?" Once the chart shows control, carry the within-subgroup sigma into a Cp/Cpk capability study to compare the process against specification limits, and express the result as a sigma level. If the chart shows special causes, find them first with a Fishbone diagram — computing capability on an unstable process predicts nothing.
Frequently asked questions
- What is a control chart used for?
- A control chart monitors a process metric over time and separates common-cause variation (the noise inherent to the process) from special-cause variation (something changed). Points inside the control limits with no unusual patterns indicate a stable, predictable process; points outside the limits or non-random patterns signal a specific, findable cause. It is the core tool of statistical process control (SPC) and the Control phase of a DMAIC project.
- How are control limits calculated?
- Control limits sit three standard deviations either side of the center line, with sigma estimated from short-term variation rather than the overall spread. An I-MR chart uses the average moving range: UCL/LCL = mean ± 2.66 × MR-bar. An X-bar/R chart uses the average subgroup range: UCL/LCL = X-double-bar ± A2 × R-bar, where A2 depends on subgroup size. A p-chart uses binomial limits: p-bar ± 3 × √(p-bar × (1 − p-bar) ÷ n), which widen for smaller samples.
- What is the difference between control limits and specification limits?
- Control limits come from the voice of the process — its own historical variation — and describe what the process actually does. Specification limits come from the voice of the customer and describe what the process should do. They are unrelated numbers: a process can be in perfect statistical control while producing out-of-spec parts, and vice versa. Control charts answer "is the process stable?"; a capability study (Cp/Cpk) answers "is it good enough?"
- Which control chart should I use?
- If you measure a continuous value one at a time (a cycle time, a monthly metric, a batch result), use the I-MR chart. If you collect small samples of measurements at intervals (for example 5 parts every hour), use X-bar/R — the R chart tracks consistency and the X-bar chart tracks the average. If you count defectives out of a sample rather than measuring a value, use the p-chart. Measurement charts detect process changes with far less data than attribute charts.
- What are the out-of-control rules?
- This tool checks three widely used rules: (1) any point beyond a 3-sigma control limit, (2) eight consecutive points on the same side of the center line, indicating a sustained shift in the process average, and (3) six consecutive points steadily rising or falling, indicating a trend such as tool wear or drift. These are the highest-signal subset of the Western Electric / Nelson rules.
- Is this control chart maker free?
- Yes. It runs entirely in your browser with no sign-up. Paste your data, pick a chart type, and get the center line, control limits, flagged signals, and a print-ready export. A free Excel template with the same formulas is also available for teams that chart on a spreadsheet.
Related tools
- Process Capability (Cp, Cpk) — is the stable process actually meeting spec?
- Sigma Level & DPMO Calculator — express the defect rate on the Six Sigma scale
- Pareto Chart — rank the causes behind flagged signals
- Fishbone Diagram — run the root-cause hunt on a special cause
- Hour-by-Hour Board — the line-side companion for plan-vs-actual tracking