MindStat SPC — Statistical Process Control & Quality

Variables Control Charts

Continuous-measurement control charts with Western Electric / Nelson rules and 3σ limits from Rbar/d2, Sbar/c4, or pooled SD.

📐 Which chart should I pick?

X̄-R — subgroups of 2–10 measurements (e.g., 5 pieces per shift). Fast to compute, easiest to read.

X̄-S — larger subgroups (>10) or variable subgroup size; uses S instead of R.

I-MR — individual measurements, no subgrouping (e.g., one lab reading per batch).

CHART OPTIONS

Chart Type

Sigma Estimation

Significance (k-sigma)

USL (optional)

LSL (optional)

📋Data format: CSV — rows = subgroups, columns = measurements within subgroup. First row = header (optional numeric). For I-MR: single column of individual measurements.

📥 Data

Data Grid

Edit cells directly. Click 'Apply to CSV' to transfer data.

📊 Statistical Interpretation

Attributes Control Charts

p, np, c, and u charts for count-based quality data. Includes Nelson rules.

🔖 Which attributes chart?

p — proportion defective when the sample size varies.

np — count of defective units when the sample size is constant.

c — count of defects per inspection unit (constant unit size).

u — defects per unit when unit size varies.

CHART OPTIONS

Chart Type

K-sigma

📋Data format: CSV — for p/np: Inspected,Defective. For c: Defects (single column). For u: Units,Defects.

📥 Data

Data Grid

Edit cells directly. Click 'Apply to CSV' to transfer data.

📊 Statistical Interpretation

CUSUM Chart

Tabular Cumulative Sum chart for detecting small sustained shifts in the process mean. Two-sided with C+ and C- statistics.

PARAMETERS

Target mean (mu0)

K (slack, default 0.5sigma)

H (decision interval, default 5sigma)

📋Data format: Single column of sequential measurements. First row = header (optional).

📥 Data

Data Grid

Edit cells directly. Click 'Apply to CSV' to transfer data.

📊 Statistical Interpretation

EWMA Chart

Exponentially Weighted Moving Average chart with exact and steady-state control limits. Good for detecting small shifts.

PARAMETERS

Lambda (0.05-0.4)

L (sigma multiplier)

Target mean (mu0)

📋Data format: Single column of sequential measurements.

📥 Data

Data Grid

Edit cells directly. Click 'Apply to CSV' to transfer data.

📊 Statistical Interpretation

Run Chart

Time-ordered plot with runs analysis (above/below median). Four pattern tests: too few/many runs, trends, oscillation.

DATA

📋Data format: Single column of time-ordered measurements.

📥 Data

Data Grid

Edit cells directly. Click 'Apply to CSV' to transfer data.

📊 Statistical Interpretation

Process Capability Analysis

Full capability study: Cp, Cpk, Pp, Ppk, Cpm, Cpmk, sigma level, PPM defective, with normality test and histogram.

SPECIFICATIONS & DATA

USL (required)

LSL (required)

Target

Subgroup size (for Cp/Cpk, 0=individual)

Confidence level

Apply 1.5σ long-term shift (Motorola convention) (unchecked = ISO/academic short-term sigma)

📋Data format: Single column of measurements (or rows of subgroups for within-subgroup sigma estimation).

📥 Data

Data Grid

Edit cells directly. Click 'Apply to CSV' to transfer data.

📊 Statistical Interpretation

Capability Indices Calculator

Interactive calculator and benchmark comparison for Cp, Cpk, Pp, Ppk with visual gauge and industry benchmarks.

📐 Formulas, meanings, and acceptance thresholds

Cp = (USL − LSL) / (6σ_within)
Potential capability — ignores mean drift.

Cpk = min[(USL − X̄) / (3σ_within), (X̄ − LSL) / (3σ_within)]
Actual capability — penalizes drift from the midpoint.

Pp, Ppk — same formulas but with long-term sigma (actual performance).

Acceptance thresholds (AIAG / IATF 16949)

• Cpk < 1.00 — not capable

• 1.00 ≤ Cpk < 1.33 — marginally capable

• 1.33 ≤ Cpk < 1.67 — capable — AIAG minimum for existing processes

• Cpk ≥ 1.67 — excellent — new process standard

• Cpk ≥ 2.00 — Six Sigma world-class (~3.4 DPMO)

Note: Always check normality before interpreting Cp/Cpk.

CALCULATOR

USL

LSL

Target

Process Mean

Within Sigma

Overall Sigma

📊 Statistical Interpretation

Gauge R&R — ANOVA Method

Two-way ANOVA for measurement system analysis.

🔧 Terminology

EV — Equipment Variation (repeatability).

AV — Appraiser Variation (reproducibility).

GRR — Total Gauge R&R = √(EV² + AV²).

PV — Part-to-part variation you want to detect.

%GRR / Tolerance — < 10% acceptable · 10–30% conditional · > 30% unacceptable (AIAG MSA).

ndc — Number of Distinct Categories ≥ 5 → reliable discrimination.

DATA & OPTIONS

Tolerance (USL - LSL)

Number of distinct categories

📋Data format: CSV — Operator,Part,Measurement. Typical: 3 operators, 10 parts, 2-3 replicates each.

📥 Data

Data Grid

Edit cells directly. Click 'Apply to CSV' to transfer data.

📊 Statistical Interpretation

Gauge R&R — Xbar & Range Method

Classical AIAG short method: EV, AV, GRR, PV, TV computed from range statistics. No ANOVA required.

DATA & OPTIONS

Tolerance (USL - LSL)

K1 multiplier (5.15 for 99%, 6 for 99.73%)

📋Data format: CSV — Operator,Part,Measurement.

📥 Data

Data Grid

Edit cells directly. Click 'Apply to CSV' to transfer data.

📊 Statistical Interpretation

Bias & Linearity Study

Assess measurement system bias at multiple reference points and linearity of the bias across the measurement range.

DATA

📋Data format: CSV — Reference,Measured. Multiple measurements per reference value for repeatability.

📥 Data

Data Grid

Edit cells directly. Click 'Apply to CSV' to transfer data.

📊 Statistical Interpretation