Before carrying out any measurements, whether for process control or experimental work, you need to ensure that the measurement system is capable of measuring accurately and precisely. A measurement system analysis is done to make sure that any observed differences in data are due to what is being measured and not the measuring system.

Measurement is fundamental to science, the effective use of statistics is dependent on the quality of measurement. Before starting statistical analysis, the measurement system should meet three tests.

  • The measurement units are small enough to discriminate between items.
  • The item measures what it is intended to measure.
  • Re-measurement of the same item gives the same result.

Good Measurement System

A good measurement system should be

  • Accurate: It should produce a number that is “close” to the true value
  • Linear: it produces accurate and consistent results over the entire range of concern.
  • Repeatable: If the same item is measured repeatedly, the measurements are close to one another.
  • Reproducible: It produces the same results when used by any properly trained individual.
  • Stable: When applied to the same item, it should produce the same results in the future as it did in the past.

Measurement System Analysis – Bias

The difference between the averages measured value and a reference value.
Measurement System Analysis 01

Measurement System Analysis – Repeatability

The variation when the same equipment is used several times on same item by same operator. (Precision – the closeness of agreement between individual test results)
Measurement System Analysis 02

Measurement System Analysis – Reproducibility

Is the variation in the average of the measurements made by different operators, using the same measuring equipment to measure the identical characteristic on the same item.

Measurement System Analysis 03

Measurement System Analysis – Stability

Is the total variation in the measurements obtained with a measurement system on the same item when measuring a single characteristic over an extended time period. The system is stable if the results are the same at different points in time.

Measurement System Analysis 04

If you are interested in measurement system analysis (MSA), or need training for your employees in how to qualify measurement equipment used in a process or prior to purchase. Then contact us for a free consultation.

Alternatively you can continue to our quality training page for an overview of other training we provide.

Go to Quality Training