Skip to content

Free template

Updated

MSA study workbook

A measurement system analysis (MSA) checks that a gauge and the way it is used give readings you can trust. A gage R&R study measures the scatter between repeated readings and between people, but it does not tell you whether the average reading is right, whether it stays right across the gauge's range, or whether it drifts over time. Those are three separate studies: bias, linearity and stability. This workbook has a sheet for each. The bias sheet compares the average of repeat readings with a reference value and tests whether the gap is real. The linearity sheet measures five reference parts across the range and fits how the bias changes. The stability sheet puts repeat readings of a master part on an X-bar and range chart and flags points outside the limits. The PDF has the same three forms to print, a one-page guide to which study to use and a worked example. The acceptable limits are your own rule or your customer's; none is built in.

Free to use: print it, copy it and edit it for your team. Enter your name and work email once to download.

Page 1 of the MSA study workbookA table of five questions about a gauge, each with the study that answers it, what it tells you, when to do it and what you need, above a list of terms and a checklist before any study.
Which study whenBias studyLinearity studyStability study with X-bar chartt tests and intervalsLimits as % of toleranceWorked example
Filled-in MSA study workbook: a micrometer: bias, linearity and stability
The same sheet filled in with the worked example (page 5 of the PDF, illustrative numbers). Open full size (opens the picture in a new tab)

When to use it

When to use the MSA study workbook

  • When a gage R&R looks fine but you still doubt the readings, for example a gauge that always reads a little high.
  • When a new gauge arrives, comes back from repair or is moved to a different job, and you want to know if it is accurate.
  • When a gauge is used over a wide range of sizes and you need to know whether the bias is the same at both ends.
  • When readings of a master part seem to move with the season, the shift or the temperature.
  • Not for judging how much of the variation is the people: that is a gage R&R.

How to fill it in

  1. 1

    Choose the study from the question

    Is the average right: bias. Is it as right at both ends of the range: linearity. Does it hold over time: stability. The guide sheet sets out which to use when and what each needs.

  2. 2

    Fix the conditions and the limit

    Use a gauge in calibration and a reference part whose value is known from a better source. Write the unit and the tolerance on the sheet, and decide your acceptable limit, in percent of the tolerance, before you measure.

  3. 3

    Bias: measure one reference part 10 to 15 times

    Enter the readings and the reference value. The sheet works out the average, the bias, the t test and the bias as a percent of the tolerance.

  4. 4

    Linearity: five parts across the range, 10 readings each

    Enter the reference values and the readings. The sheet takes the bias of every reading against its reference value, fits a line and reports the slope and the linearity as a percent of the tolerance.

  5. 5

    Stability: a master part, 20 or more small groups over time

    Measure the same master part in groups of 2 to 6 at regular times. The sheet draws an X-bar chart with limits worked out from your own groups and flags any group outside them.

  6. 6

    Decide and act

    A significant bias can be adjusted out or corrected for. Linearity that is too large means restricting the range, correcting or replacing the gauge. A stability flag means finding what changed at that time.

A filled-in example

Illustrative, not a benchmark

An example (illustrative): a 0 to 25 mm digital micrometer measuring a feature with a 0.300 mm tolerance, with an acceptable limit of 5% of the tolerance. All numbers are made up.

  • Bias: a 12.500 mm gauge block is measured 15 times. The average is 12.512, so the bias is 0.012 mm, which is statistically significant but 4.0% of the tolerance.
  • Linearity: five gauge blocks from 5 to 25 mm are measured 10 times each. The bias grows from 0.010 mm at 5 mm to 0.015 mm at 25 mm, a linearity of 1.8% of the tolerance.
  • Stability: 20 groups of 3 readings of a master part are plotted on an X-bar chart. Group 14, measured after the room warmed up, falls outside the limits; the range chart stays in control.

Bias and linearity are small next to the tolerance, so the micrometer is fit for this use under the 5% rule. The stability flag points to a temperature effect, which the other two studies could not show.

Common mistakes

  • Using a reference value of unknown quality

    A bias study is only as good as the reference. Use a certified block or a part measured by a better method, and note where its value came from.

  • Confusing significant with important

    A significant bias means the gap is bigger than the scatter would explain. Whether it matters depends on the tolerance, which is why the sheet also gives it as a percent of the tolerance.

  • Running a linearity study on parts that do not span the range

    Five parts bunched in the middle of the range say nothing about the ends. Choose references spread from near the bottom to near the top of the range you actually use.

  • Judging stability from too few groups

    A handful of groups cannot show drift. Collect at least 20, spread over the time period that matters, such as several weeks.

  • Setting the acceptable limit after seeing the result

    Decide the limit first, from your own rule or your customer's. A limit chosen after the fact can make any gauge pass.

Download the template

Free to use: print it, copy it and edit it for your team. Enter your name and work email once to download.

Run this template in LeanSuite

In LeanSuite quality issues are logged as tags, ranked in the QA Matrix and traced to their root cause, and digital instructions and checklists with number limits, required photos and signatures build quality into every task.

See it in a 45-minute demo

FAQ

MSA study workbook: common questions

  1. Bias is the gap between the average reading and a reference value. Linearity is how much that bias changes across the range of the gauge. Stability is how much the readings of the same master part change over time. Together with repeatability and reproducibility they describe how good a measurement system is.

  2. A gage R&R measures the scatter in readings, between repeats and between people. A bias study measures whether the average of the readings is right. A gauge can be very repeatable and still read consistently wrong.

  3. 10 to 15 repeat readings of one reference part, by one person with the same gauge and method, is a common choice. More readings narrow the confidence interval for the bias. The sheet has room for 20.

  4. There is no single number. It depends on the tolerance and on your customer's requirements. This workbook asks for your own limit, in percent of the tolerance, and tells you whether the result is within it.

  5. A control chart shows whether readings of a master part stay inside the limits expected from their own scatter. A point outside, or a pattern, means something changed in the gauge, the master part or the conditions at that time.

More free templates

All templates
Browse all 236 templates

Can't find the template you need?

Tell us what you're looking for. We'll make it and send it to your inbox, free.

A shift handover sheet, a red tag log, a changeover checklist for one machine: if your team uses it on the floor, ask.

By sending this form you agree to our Terms of Service and Privacy Policy.

Pass it on

We want LeanSuite to be the best place on the internet for lean help. If this was useful, send it to someone on your team or in your network who needs it.