Decisions on accepting or rejecting products are taken on the basis of measurement values. As all processes, the measurement process also shows variation so no measurement is perfectly correct. The objective of Measurement Systems Analysis (MSA) is to determine how large the area of uncertainty is around the measured value and what are sources of variation that contribute to this uncertainty. Do not underestimate the importance of this: the larger the area of uncertainty, the bigger the chance that products are wrongly rejected (increased internal failure cost) or wrongly accepted (increased external failure cost).

QS Consult view and approach

In June 2010 edition 4 of the MSA manual from the Automotive Industry Action Group (AIAG) was released. Originally this manual was part of the QS-9000 standard and it is still seen as the standard for MSA. It is the base for the QS Consult course on MSA. Note that MSA is not an easy topic and many companies have serious problems understanding the methods, the demands and how to fulfil these demands. Performing MSA studies is a time consuming and expensive business, so it is very important to do what is needed but not more than is needed. We need to be cost conscious in the implementation of MSA.

Services offered

• MSA training at various levels (beginner to specialist).
• Support with the implementation of an MSA system.
• Setting up appropriate MSA studies, evaluating the results and recommending actions for improvement.

Additional information

Tips to make MSA simpler and cheaper:

• Determine what kind of measurement errors can occur in the system. Do not investigate what cannot exist (no linearity for a system with small measurement range, no operator effect in automatic measurement systems, etcetera)!
• Measurement systems must be investigated. Do not do an MSA study for every instrument. Instruments are controlled through calibration.
• Make sure there are very clear and unambiguous measurement instructions.
• Make a graphical representation of the results to get a better insight in the behaviour of the measurement system.

MS Excel 2003 tools (Dutch version only)

• Bias
• Gage R&R
• Attributive measurement systems