1. The Normal Distribution (Bell Curve)
A continuous probability distribution representing symmetrical data. In a perfectly optimized environment, most data points cluster around the central peak (the Mean), while extreme events taper off evenly into the "tails". In reality, human and process data rarely fit a perfect Normal Curve without CAPA intervention.
2. The Four Statistical Moments
Standard reporting stops at two moments. This Master Class platform utilizes all four to expose the hidden realities of your floor operations:
- 1st Moment - Weighted Mean (μ): The exact center of gravity. Higher weighted inputs drag this center toward themselves.
- 2nd Moment - Standard Deviation (σ): Measures the spread or inconsistency of your process. A massive SD means total chaos; a tight SD means strict operational control.
- 3rd Moment - Skewness (γ): Measures asymmetry. A Positive Skew (tail pointing right) means your team is struggling and bulk performance is low. A Negative Skew means bulk performance is highly mature.
- 4th Moment - Excess Kurtosis (κ): Measures "Tail Fatness" and extreme outlier risk.
3. The Kurtosis Scenarios (Tail Risk)
Kurtosis is the ultimate lie-detector for process stability. It tells you what is hiding in the extremes:
- 🔴 Leptokurtic (Excess Kurtosis > 1.0): "The Black Swan Scenario"
The curve has a very sharp peak but extremely "fat, heavy tails".
Interpretation: The process looks fine on average, but it is highly prone to catastrophic, unpredictable extreme failures. When it fails, it fails massively. Requires immediate Root Cause Analysis on outliers.
- 🟠Platykurtic (Excess Kurtosis < -1.0): "The Flatline Scenario"
The curve looks like a plateau with thin, non-existent tails.
Interpretation: Total lack of standard operating procedures. Performance is highly dispersed across the board. No one is exceptionally bad or exceptionally good; the entire workforce is just wildly inconsistent. Requires complete baseline process standardization.
- 🟢 Mesokurtic (-1.0 to 1.0): "The Standard Baseline"
Matches a normal Gaussian distribution.
Interpretation: Extreme events are rare and mathematically predictable. The process is under standard statistical control.
4. Kernel Density Estimation (KDE)
Unlike basic tools that draw a fake, symmetrical bell curve over messy data, KDE mathematically traces the Actual Process Wave. If your data has two peaks (Bimodal) because you have two different shifts performing differently, the KDE curve will visually show both humps.
5. Outlier Standard Deviation Range
This setting controls your "Red Lines". In standard Six Sigma, any data falling beyond ±2 or ±3 Standard Deviations is considered an extreme statistical anomaly. These points require immediate Root Cause Analysis (RCA).
6. CAPA Corrector (Corrective and Preventive Action)
Pressing the CAPA button triggers an algorithmic normalization. It mathematically redistributes your raw, skewed data into an optimized, symmetrical Normal Curve. It shows you exactly what your data should look like if all systematic errors were resolved.