# Articles

## Your Step-by-Step Guide to MSA – Measurement System Analysis

When collecting data from a particular process it essential have a correct measurement system to achieve correct results and conclusions. Measurement System Analysis provides a great tool to ensure the quality of the data collected from a process.

MSA – Measurement System Analysis has three purposes:
1. Examine how much of the total observed variability comes from the measurement or tool;
2. Delimit the components that cause a variability in the measurement system; and
3. Assess whether the measurement or tool is capable.

There are a variety of statistical properties that can be determined for evaluation of measurement system: stability, bias, variation (which can be broken down into repeatability and reproducibility) and linearity. Among the criteria mentioned above repeatability and reproducibility are the most essential for determining the quality of the measurement system which can be determined by Gage R&R study.

In a Gage R&R study:
(1) repeatability measures the ability of the measurement system to give the same result at repeated measurements when same operators, methods, and equipment are used and
(2) reproducibility addresses the variability that occurs when different operators perform the same type of work (measuring).

Consider a case where plastic parts are produced from an injection molding machine, it is required to assess if the part observed is scratched or not. If the part has no scratches on it, is considered as “OK” whereas if the part has at least 1 scratch on it, it is considered as “not OK”.

In order to assess the repeatability and reproducibilityof the case give above following steps should be taken:
• A sample of 50 parts are collects from the injection molding machine (the required sample size depends on the nature of the data to be collected)
• Each part shall be given a serial number from 1 to 50
• Each operator inspects all the parts from 1 to 50 and decided if they are OK or not. (Operators do not know the serial number assigned to parts while they are deciding on the quality of the parts. The assistant of the operator knows the serial number of the parts they give to assessment operator and write down the decision based on the answer given for that specific part
• Each operator assesses all parts from 1 to 50 again and the assistant write the decision for each part next to previous decision given for the same part number by the same operator.
• After writing down all the decision for all the parts, the percentage of matching on decisions for the same part number by same operator is evaluated. If the matching between the 1st and 2nd assessment of same part by same operator is above 90% the system is considered highly repeatable if the value is %70 it is usually accepted but improvement of MSA is advised
• To evaluate the reproducibility of the system matching of decisions for the same part by different operators are investigated. In this case again if the decision match more than 90% the measurement system is considered highly reproducible whereas above 70% is considered acceptable and any value below 70% implies a non-reproducible measuring system. This means that operators decision on parts inspected are subjective not objective.

At 6Sigma.us we are committed to helping people find solutions! We provide hands-on implementations of Lean and Six Sigma at our locations, at your workplace or online. Visit our schedule of classes and find a solution that meets your needs, or contact us and we will surely help you find the right fit.

SixSigma.us offers both Live Virtual classes as well as Online Self-Paced training. Most option includes access to the same great Master Black Belt instructors that teach our World Class in-person sessions. Sign-up today!