Quality improvement methodologies and the analyses that are performed within your organization are only as good as the data on which they are based. That is why it is essential to make sure that your measurement system - your gages, personnel, methods and procedures - is stable and capable of measuring your data before continuing with your process improvement efforts.
Measurement System Analysis Defined
Measurement System Analysis, often referred to as MSA, is used to assess the statistical properties of process measurement systems. Measurement systems can include your collection procedures, gages, and other test equipment used to collect data for analyzing process problems.
The purpose of measurement system analysis is to ensure and demonstrate that your measuring procedures and systems provide
1. adequate resolution,
2. results that are not unduly biased, and
3. little variability in comparison with specified tolerances.
Importance of Measurement System Analysis
Establishing the adequacy of your measurement system using a measurement system analysis process is fundamental to measuring your own business process capability and meeting the needs of your customer (specifications). Take, for instance, cycle time measurements: It can be measured in seconds, minutes, hours, days, months, years and so on. There is an appropriate measurement scale for every customer need/specification, and it's the job of the Quality professional to select the scale that is most appropriate.
| Measurement | Scale / Resolution / Granularity |
|
|
| Cycle Time | second, minute, hour, day, week, month, year& |
| Tolerance | millimeter, centimeter, decimeter, meter& |
| Cost | cent, dime, dollar, hundred dollars& |
Measurement System Analysis Example
This concept can easily be displayed with an example. Let's examine the case of a customer awaiting home installation of cable for their television. The customer has been told that installation with take place on a specified day between 1 and 4 p.m., and the installation duration will be half an hour. In the customer's view, the cable company will be on time if they arrive anytime between 1 and 4 p.m. and the cable company will be producing a defect in service if they do not show up for installation between those times. It is a defect for the cable company to show up outside of these times because the customer may be at work up until 1 p.m. or may have another appointment after 4:30 p.m.
For years, the cable company has measured their performance on whether they showed up during the specified time period by writing the hour in which they arrived. If they arrived at 1:34 it was recorded as a '1' and within the time period; if they arrived at 3:59 it was recorded as a '3,' again, within the time period. This worked well because it displayed if they were early (recording a 12 or before) or late (recording a 4 or later) in every installation event.
The problem with this measurement system arises when a service technician arrives outside of the customer specified limits of 1 and 4 p.m. In this case, the cable company has no way to determine how much of a defect is being produced - their measurement system does not display adequate resolution. The appropriate resolution would measure in minutes not hours, which would allow the cable company to quantify if the service technician showed up at 4:01 or 4:59. This would help the cable company determine the root cause of the defect and restructure their businesses schedule more effectively.
| Cable Business Scale | Measurement System Analysis Issue |
|
|
| Day | Not fine enough resolution |
| Hour | Not fine enough resolution |
| Minute | Proper measurement system resolution |
The next time you charter a process improvement project and are preparing to collect data, your measurement system analysis can help you ensure that you are measuring
1. as your customer feels the problem, and
2. at a great enough resolution to view customer perceivable defects.
Measurement System Analysis To Come
In future articles we will delve into the details surrounding Gage R&R (repeatability and reproducibility), bias, stability, and linearity of your data. We'll describe how the quality of measurement data relates to continuous improvement in a practical format.
Are you skilled in establishing measurement systems to the correct resolution and without bias? Would you like to see your name and bio in print online? Why not submit a piece of this article for publishing? If interested, feel free to drop us an email and we'll give you more details.
