Statistical software for manufacturing and process quality Detect a process change while it is still small, show whether the process meets the specification, and keep the evidence in a workbook an auditor can open.

A control chart is only as good as the rule set behind it. Too few rules and a slow drift runs for weeks before a point crosses a limit. Too many and the operators learn to dismiss the flags, which is worse than having no chart at all. The rule set is a decision to make deliberately, before the data arrives.

Analyse-it draws Shewhart variable charts — Xbar-R, Xbar-S and I-MR — and the attribute charts p, np, c and u. Detection rules come as WECO, Nelson, Montgomery or a set you define yourself. Phases mark a deliberate change such as a new lot or a re-tooling. The limits then recalculate either side of the change, rather than dragging the old process across the new one. Stratification separates lines, shifts or operators on the same chart.

Where a drift is small and sustained rather than sudden, CUSUM, EWMA and UWMA accumulate the evidence and signal earlier than any Shewhart rule will. Reading out-of-control signals covers how to tell a real shift from the false alarms any rule set produces.

A capability index is an estimate, and a single number hides how uncertain that estimate is. Quoted from thirty observations, a Cpk of 1.33 is consistent with a true value well below the limit your customer wrote into the contract. The confidence interval is the part of the result that answers the question actually being asked.

Analyse-it reports Cp, Cpl, Cpu, Cpk and Cpm for the within-subgroup variation, and Pp, Ppl, Ppu and Ppk for the overall variation, with confidence intervals on each. Nonconforming units and sigma metrics come from the same analysis. Where the data is not normal, transformations are applied and shown rather than assumed away, so the choice is visible to whoever reviews the study. Cp, Cpk and Ppk explained sets out what each index measures and why the within and overall figures differ.

Pareto analysis is simple arithmetic and is still the fastest way to stop a team working on the third most common failure. The value is in the stratification. A defect that dominates overall may belong almost entirely to one line, one shift or one supplier, and the combined chart hides that.

Analyse-it draws the Pareto chart with its cumulative percentage line and stratifies it two ways at once. Combine, reorder or colour the bars so the categories match how the plant actually records them.

A chart tells you when a process moved. Finding out why is ordinary statistics: comparing groups, fitting a model, testing whether a factor matters. Quality teams routinely own a package that charts well and then borrow a second one for this part, or fall back on spreadsheet formulas that nobody validates.

Every Analyse-it edition includes the full Standard edition. Hypothesis tests, one-way, two-way and multi-factor ANOVA, regression and the diagnostics that go with it are in the same ribbon as the control chart. Factorial designs let you vary the suspected factors together rather than one at a time, which is both faster and better at finding interactions.

Factorial designs covers how to plan one and read the interactions it finds.