SPC Control Charts (I-MR, X̄-R, X̄-S)
Control chart core values (CL, UCL, LCL, σ̂) for I-MR, X̄-R and X̄-S plus Nelson Rule evaluation and capability from σ̂.
Description
The SPC engine covers the three classic Shewhart control charts: I-MR (individuals + moving range), X̄-R (subgroup mean + range, 2 ≤ n ≤ 10), and X̄-S (subgroup mean + std-dev, 2 ≤ n ≤ 25). Control limits are computed from the classic AIAG/Montgomery SPC constants (d2, A2, A3, D3, D4, B3, B4, c4). σ̂ is derived from the baseline subgroup; `baselineEnd` enables a Phase-1/Phase-2 analysis. The analyze(inputs) dispatcher additionally supports mode 'nelson' for the eight Nelson rules (1–6 enabled by default) and 'capability' for Cp/Cpk from a supplied CL and σ̂. The flat output shape (e.g. i.cl, xbar.ucl) matches the fixture reference values 1:1.
Formulas
Assumptions
- Data is approximately normally distributed (especially relevant for I-MR)
- Baseline measurements are independent and from a stable process
- Subgroups are rationally formed: within-subgroup variation < between-subgroup
Limitations
- Subgroup size n must match chart type: I-MR n=1, X̄-R 2 ≤ n ≤ 10, X̄-S 2 ≤ n ≤ 25
- Autocorrelation or non-normality distorts the control limits; I-MR is especially sensitive
- Nelson rules 7 and 8 are disabled by default — enabling them raises the false-alarm rate
References
- Montgomery, Introduction to Statistical Quality Control, 8th Ed. — Chapters 6-7
- AIAG — Statistical Process Control (SPC) Reference Manual, 2nd Edition
- Nelson, L. S. (1984), The Shewhart Control Chart — Tests for Special Causes, Journal of Quality Technology, 16(4), 237–239