Do we really need the MSA procedure?

Today's production companies collect a large amount of data, through measurements and inspections, on the basis of which decisions will be made with an impact on processes and business in general.

Inspections performed either automatically (more and more often) or by specialized personnel, can also generate erroneous decisions, with effects in artificially increasing defects or sending non-compliant products to customers. In both situations we lose money and even customers.

To prevent these inconveniences, MSA is used: statistical technique for the analysis of the measurement system.

What should an MSA procedure contain ?

First of all, it should be mentioned that the existence of this procedure is not mandatory according to the IATF 16949 standard, but it helps organizations to consistently use best practices.

Accordingly, an MSA procedure should contain:

• description of the measurement systems for which MSA studies must be performed, all of which are described in the Control Plans
• the correct methodologies to be used in each case:

1. Gage R&R or another method
2. continuous or attributive data
3. data sampling methods - how many operators, how many samples and how many repetitions for each type of MSA study
4. the criteria for accepting the MSA study, considering also the specific requirements of the clients in this direction

• prioritization criteria depending on the criticality of products and processes
• the knowledge necessary to be able to carry out and interpret MSA studies as well as the possible trainings necessary to accumulate this knowledge

Remember that the application of MSA requires relatively important resources (human, time), precisely for this reason, but the procedure must be applied strictly, otherwise the results will be irrelevant, so I kind of wasted everything I spent. Not to mention that we will not decrease / prevent customer complaints and internal defects.