Measurement systems need to be adequate, reliable, and correct in their formation. Two major things contribute to total variation – the process and the measurement system. One of the key things you need to do is determine the amount of variability caused by measurement errors, as this is something you need to think about when defining a measurement system. It’s the convergence of people with the techniques used and the variation of how your devices are going to work in the measurement process.
There are a number of important elements in the measurement system – people, standards, devices, environments, instruments and methods. A fishbone diagram can illustrate the challenges in a measurement system with a specific kind of problem. The measurement systems variation should represent a very tiny percentage of the total variation in population of data.
Many different causes of measurement errors can come into play in the context of Lean Six Sigma projects, these include:
- Devices – An example is gauges. You could have issues with calibration or need to deal with challenges related to wear and tear or damage to measuring devices. Maybe there’s a lack of acuity or precision in the actual capability of the measurement devices.
- Methods – If your methods have inherent issues with how you control and maintain quality within your measurement system, that’s going to create a challenge. You may face potential problem areas, such as measuring color acuity or finding devices to measure the shininess of a surface. One thing that’s also really difficult to manage is the sustainability and the susceptibility to human error in the use of methods, as you move through the process.
- Documentation – There could be a lot of issues in this area. For example, you could interpret an instruction in different ways, depending on who’s going to do the measurement. The documentation could also be incomplete, contain errors, or allow you to do measurements in an incorrect sequence.
- Environment – Environmental factors that could come into play include earth tremors, relative humidity, temperature, electrical surges, drops in power, wind, or fluctuations in barometric pressure.
- Person – According to experts, a 100% human inspection is only about 80% to 85% effective. So if you’re highly reliant on human beings to get it right, you could experience a lot of error in your measurement system.
Measurement Systems Analysis (MSA) refers to the analysis of precision and accuracy of measurement methods. It is an experimental and mathematical method of determining how much the variation within the measurement process contributes to overall process variability. Characteristics contribute to the effectiveness of a measurement method which is