Conventional Xbar-R, I-MR and P charts suit typical process monitoring, but production often presents small shifts, short runs, rare events, overdispersion or mixed data across conditions — cases where ordinary charts miss signals or alarm too often. Advanced control charts are designed for these situations: EWMA and CUSUM detect small shifts, median-range charts are more robust, Laney P′/U′ corrects over- and under-dispersion, Z-MR standardizes data across batches, G/T handles zero-inflated rare events, zone charts integrate run rules, and precontrol fits short runs and changeovers.
Choose by data characteristics: EWMA or CUSUM for slow drift and small shifts in continuous processes; median-range for outliers or skew; Laney P′/U′ for high-volume count data with abnormal fluctuation; Z-MR when sample sizes or conditions differ across batches; G/T for extremely rare defects dominated by zeros; zone charts for standardized automatic run-rule testing; precontrol for short batches with frequent changeovers. This tool provides all eight charts online with automatic rule testing and AI interpretation, replacing manual table lookups and programming.
Select the chart that matches your data type and paste the observations as subgroups or individuals. The tool computes control limits from the corresponding theory, marks out-of-control points and names the rule violated, then offers an AI interpretation of likely assignable causes. Confirm the special cause on the shop floor before acting — control charts signal abnormality, they do not replace root-cause analysis.
EWMA: z_i = λx_i + (1−λ)z_{i−1} with pointwise limits ±L·σ√(λ/(2−λ)(1−(1−λ)^{2i})); Tabular CUSUM: C_i⁺ = max(0, C_{i−1}⁺ + x_i − μ0 − k) against decision limit h; Laney: σ_z = MR̄/1.128 corrects the P/U limits; G/T applies the power transform t^0.2777 to event intervals. Each chart page shows the specific formulas behind its own control limits.