Validate Measurement System and Data Collection

By Dave Litten
About Validate Measurement System and Data Collection

Validate Measurement System and Data Collection

Although data collection plays a very important role in all Lean Six Sigma projects, before collecting data, the team must assess if the measurement system (measuring instrument, appraiser and environment in which measurement happens) is accurate and precise.

So, the team must perform Measurement System Analysis (MSA) – also known as Gage R&R.

Measurement system analysis determines if the measurement system can generate accurate data and is the accuracy is adequate to achieve your objectives.

Therefore, MSA make sure that the difference in the data are due to actual differences in what is being measured, and not to variation in measurement methods.

It is interesting to note that experience shows 30% to 50% of measurement systems are not capable of accurately or precisely measuring the desired metric. Measurements need to be precise and accurate. Accuracy and precision are different and independent properties:

Lean6Sigma Precision Accuracy

lean6sigma precision accuracy

Data may be accurate in that may reflect the true values of the property, but not precise in that measurement units do not have enough discriminatory power periods

The opposite is also true, data can be precise yet in an accurate in that they are precisely measuring something that does not reflect the true values. Sometimes data can be neither accurate normal precise.

Statistical definitions

From a statistical viewpoint, there are four desirable characteristics that relate to precision and accuracy of continuous data:

  • But no systematic differences between the measurement values we get and the “true value”
  • The ability to get the same result if we take the same measurement repeatedly, or if different people take the same measurement
  • The ability of the system to produce the same results in the future that it did in the past
  • The ability of a system to detect meaningful differences, that is, good discrimination

Yet another desirable characteristic is called linearity. This is the ability to get consistent results from measurement devices and procedures across a wide range of uses.

It must be obvious, that having on calibrated measurement devices can affect all these factors.

Once the team ascertains that the measurement is good, then a Data Collection Plan is prepared.

The Data Collection Plan (DCP) includes the measures whose data needs to be collected, how much data to collect, data source, who will collect the data, etc.

Unlike conventional data collection, in Lean Six Sigma projects, data is collected on both the CTQ and the potential causes identified in a Cause & Effect Matrix.

Due to the quantity of data involved in most businesses, it isn’t practically viable to collect data of the entire population.

Hence the team needs to resort to statistical sampling methods. Then, the data collection can be executed.

From time-to-time a Six Sigma project team needs to validate the data collected. Sometimes, the data collectors need to be trained and retrained.
After the data collection is complete, it is ready for a process capability assessment.

Many projects can get delayed because of poor data quality or the delay in collecting sufficient data.

Pen