Menu Close

1030. Measurement System Analysis in Statistics: Core Terminology

brprakash 2 Comments
1023. Measurement System Analysis in Statistics: Core Terminology

Measurement System Analysis in Statistics: Core Terminology

Measurement System Analysis in Statistics deals with evaluating the accuracy, precision, and overall reliability of data collection tools. Whenever operational decisions rely on quantitative measurements, teams must verify that the measurement system itself is dependable. Deploying Measurement System Analysis in Statistics ensures that observed variations reflect true process changes rather than measurement error.

A complete measurement system includes not only the physical gauge or instrument, but also the operators, calibration procedures, fixtures, software, and surrounding environmental conditions.

1. Accuracy vs. Precision

Understanding measurement reliability begins by distinguishing between Accuracy (location relative to the target) and Precision (spread across repeated trials).

                  ┌──────────────────────────────────────────────┐
                  │      Measurement System Analysis (MSA)       │
                  └──────────────────────┬───────────────────────┘
                                         │
       ┌─────────────────────────────────┴─────────────────────────────────┐
       ▼                                                                   ▼
┌──────────────────────────────┐                                    ┌──────────────────────────────┐
│     Accuracy (Location)      │                                    │      Precision (Width)       │
│  * Bias                      │                                    │  * Repeatability (Equipment) │
│  * Linearity                 │                                    │  * Reproducibility (Operator)│
│  * Stability                 │                                    │                              │
└──────────────────────────────┘                                    └──────────────────────────────┘

2. Core Terminology and Operational Definitions

Measurement Accuracy

The degree of closeness between the average of repeated measurements and a certified reference standard or true value.

Measurement Precision

The degree of agreement or repeatability among independent measurement results obtained under stipulated conditions.

Bias

The mathematical difference between the average of observed measurements and a known reference standard:

\text{Bias} = \bar{X}<i data-path-to-node="28" data-index-in-node="28">{\text{measured}} - X</i>{\text{reference}}

Linearity

The change in bias across the expected operating range of the measurement gauge. A linear gauge maintains consistent accuracy whether measuring small or large parts.

Stability

The variation in average measurement values over an extended period when measuring the same master reference block. It measures resistance to calibration drift over time.

Resolution (Discrimination)

The smallest increment of measurement that an instrument can reliably detect and display. As a rule of thumb, an instrument should possess a resolution capable of dividing the specification tolerance or process variation into at least 10 distinct intervals.

3. Summary of Measurement Location and Spread Attributes

Term Category Operational Definition Practical Example
Bias Accuracy Difference between average measured value and reference standard Scale reads 100.5\text{ g} for a certified 100.0\text{ g} weight
Linearity Accuracy Change in bias across the instrument’s operational range Bias is +0.1\text{ mm} at 10\text{ mm} but +0.8\text{ mm} at 100\text{ mm}
Stability Accuracy Change in measurement baseline over time Daily checks on a reference block shift upward over 3 months
Resolution Discrimination Smallest unit of scale division A digital caliper reading to 0.01\text{ mm} vs 0.1\text{ mm}

4. Worked Operational Examples

Example 1: Calculating Measurement Bias

A quality laboratory uses a digital micrometer to measure a certified master gauge block known to be exactly 10.000\text{ mm}. An inspector measures the block 5 times, recording:

10.004, 10.006, 10.003, 10.005, 10.007\text{ mm}

  • Step 1: Calculate Sample Mean (\bar{X})

    \bar{X} = \frac{10.004 + 10.006 + 10.003 + 10.005 + 10.007}{5} = \frac{50.025}{5} = 10.005\text{ mm}

  • Step 2: Calculate Measurement Bias

    \text{Bias} = 10.005\text{ mm} - 10.000\text{ mm} = +0.005\text{ mm}

  • Interpretation: The micrometer has a positive systematic bias of +0.005\text{ mm}, consistently overestimating component dimensions.

Example 2: Evaluating Instrument Resolution Rule of 10

A machining line produces shafts with a strict tolerance range of 50.00\text{ mm} \pm 0.10\text{ mm} (Total Tolerance = 0.20\text{ mm}).

  • Rule of 10 Target Resolution:

    \text{Required Resolution} \le \frac{\text{Tolerance}}{10} = \frac{0.20\text{ mm}}{10} = 0.020\text{ mm}

  • Assessment: Using a standard ruler with 0.5\text{ mm} markings is completely inadequate. The operator must use a digital caliper or micrometer with at least 0.01\text{ mm} resolution to ensure reliable defect detection.

Frequently Asked Questions (FAQ)

Q1: What is the difference between accuracy and precision?

Accuracy refers to how close measurements are to the true or target value, whereas precision describes how close repeated measurements are to one another.

Q2: How does calibration impact measurement bias?

Calibration adjusts the measurement instrument against a known standard, directly reducing systematic bias toward zero.

Q3: Can a measurement system be precise but inaccurate?

Yes. A gauge can produce tightly clustered, identical readings across repeated trials (high precision) while being consistently offset from the true reference value (high bias/low accuracy).

Q4: Why is stability evaluated over time rather than in a single trial?

Stability measures drift caused by environmental changes, wear, or electronic degradation, which can only be detected through periodic tracking against a reference standard over weeks or months.

Six Sigma Practice Exam Questions

1. An inspector measures a master pin certified at 5.000\text{ cm}. The average of 10 repeated measurements is 5.012\text{ cm}. What is the calculated bias?

A) -0.012\text{ cm}

B) +0.012\text{ cm}

C) +0.000\text{ cm}

D) 5.012\text{ cm}

  • Correct Answer: B) +0.012\text{ cm}

  • Explanation: Bias is calculated as average measured value minus reference value: 5.012 - 5.000 = +0.012\text{ cm}.

2. When a gauge exhibits zero bias at a setting of 10\text{ mm} but demonstrates a bias of +0.05\text{ mm} at 100\text{ mm}, the gauge has an issue with:

A) Stability

B) Linearity

C) Resolution

D) Repeatability

  • Correct Answer: B) Linearity

  • Explanation: Linearity measures changes in bias across the operating scale of the gauge.

3. A manufacturing tolerance for a stamped plate is 1.00\text{ mm}. According to standard MSA guidelines, what is the minimum recommended gauge resolution?

A) 0.10\text{ mm}

B) 0.50\text{ mm}

C) 0.01\text{ mm}

D) 1.00\text{ mm}

  • Correct Answer: A) 0.10\text{ mm}

  • Explanation: Instrument resolution should be at least 1/10\text{th} (10%) of the tolerance range: 1.00 / 10 = 0.10\text{ mm}.

4. Tracking a calibration standard daily on an X-bar control chart to monitor long-term baseline drift evaluates which measurement property?

A) Reproducibility

B) Linearity

C) Stability

D) Discrimination

  • Correct Answer: C) Stability

  • Explanation: Stability evaluates performance consistency over extended time horizons.

5. Which term describes the degree of agreement among independent measurement results obtained under identical operating conditions?

A) Bias

B) Accuracy

C) Precision

D) Linearity

  • Correct Answer: C) Precision

  • Explanation: Precision reflects the dispersion or agreement of repeated measurements under identical conditions.

This article aligns with standard body-of-knowledge practices for professional quality certification curricula.

Written by Ravi Prakash—Quality Expert (38+ yrs exp). Connect on LinkedIn or Contact Us.

Support Our Website

Your support helps us continue delivering good content and useful data. If you’d like to contribute, here’s how:

If you would like to follow this Masterclass series on quality engineering and process statistics, check out our previous and upcoming lessons:

Posted in Continuous Improvement, Measure Phase, Process Improvement, Quality, Quality Tools, Six Sigma, Statistics
Tagged , , , , , , , , , ,

About the Author: brprakash

View all posts by brprakash | Website