Measurement Systems Analysis Instruction Guide

1 DEFENSE SEGMENT Instruction Guide Measurement Systems Analysis (MSA) Instruction Guide 3/8/2017 All rights reserved under the copyright laws. Printed copies of this document are uncontrolled. DEFENSE SEGMENT Instruction Guide This Guide contains three components: 1. The MSA Flowchart is intended to assist in determining the next steps in fulfilling the MSA requirements, addressing the various types of measurements that may be encountered. The flowchart will direct the participant to various locations in the Instruction Guide , to assist in fulfilling the requirements for their particular situation. 2. The MSA Procedures is a comprehensive body of knowledge that is intended to cover the various MSA requirements that may be encountered.

2 3. The MSA Report is intended to serve as a means for a MSA participant to document all of the inputs that may have influenced the output (results) of the study. It may assist in determining the root cause(s) if a MSA component fails to deliver the desired results. The last column is intended to provide some additional guidance for the participant on potential future action. All rights reserved under the copyright laws. Printed copies of this document are uncontrolled. DEFENSE SEGMENT Instruction Guide All rights reserved under the copyright laws. Printed copies of this document are uncontrolled. DEFENSE SEGMENT Instruction Guide Measurement Systems Analysis (MSA) 1 General The purpose of MSA is to quantify the amount of accuracy and variation that exists in the measuring process MSA will assist in enabling the supplier to distinguish acceptable from unacceptable product, improve processes, and comply with Oshkosh requirements This Guide is intended to support the information contained in the Automotive Industry Action Group (AIAG) MSA Reference Manual.

3 Definitions Bias The difference between the observed average of measurements and a reference value. Bias is evaluated and expressed at a single point within the operating range of the Measurement system. Kappa -- Statistical measure of inter-rater agreement for qualitative (attribute) items, taking into account the agreement occurring by chance. Linearity The difference in bias errors over the expected operating range of the Measurement system. Stability -- Addresses the necessary conformance to the Measurement system standard or reference over the operating life of the Measurement system. Gage Requirements The gage shall be calibrated in accordance with a documented calibration procedure Graduations on the Measurement device should be one-tenth of the tolerance range or smaller (for example of a micrometer can measure to the nearest , it must not be used to measure a feature with a tolerance of less than ).

4 For critical characteristics this could restrict the one-tenth rule to the process range rather than the tolerance range. Measurements should be recorded to one decimal place smaller than the tolerance. (For example if the tolerance is the measurements should be reported to a minimum of three decimal places- ). Analog devices should be recorded to the smallest graduation. For example if the smallest scale graduation on the caliper dial is , then the Measurement results should be recorded to . MSA Components All rights reserved under the copyright laws. Printed copies of this document are uncontrolled. DEFENSE SEGMENT Instruction Guide Graphical Analysis of Results: This Analysis should be evaluated prior to any other statistical Analysis .

5 A template for this can be found in the PPAP workbook GR&R X&R . Interpretation of Range chart: The range control chart is used to determine whether the process is in control. The ranges of multiple readings by each appraiser on each part are plotted on a standard range chart including the average range and control limits. If all ranges are in control, all appraisers are performing the Measurement task in similar way. If one appraiser is out of control, the method used differs from the others. If all appraisers have some out of control ranges, the Measurement system is sensitive to appraiser technique and needs improvement to obtain useful data. Interpretation of Average chart: The averages of measurements by each appraiser by each part are plotted by appraiser.

6 This plot will assist in determining consistency between appraisers. Approximately 50% or more of the averages should fall outside the control limits. If less than 50% fall outside the control limits then either the Measurement system lacks adequate resolution or the sample does not represent the expected process variation. Stability Study design/ Analysis : Obtain sample(s) and establish the Reference Value relative to a traceable standard. Depending on the focus of the study it may be desirable to have samples at the low end, middle, and high end of the specification. On a periodic basis measure the sample(s) 3 to 5 times. The sample size and frequency should be based on understanding of the Measurement system.

7 Some factors to consider are calibration cycle, time between repairs, operating condition stress, environment changes etc. All rights reserved under the copyright laws. Printed copies of this document are uncontrolled. DEFENSE SEGMENT Instruction Guide Plot the data on and Xbar and Range control chart in time order. Analyze the control chart to identify if there are any points out of control. Identify cause of out of control point(s), make correction and continue to monitor Bias Study design: Control chart Analysis should indicate that the Measurement system is stable before the bias is evaluated. Obtain a sample and establish its reference value relative to a traceable standard.

8 If one is not available, select a production part that falls in the mid-range of the production measurements and designate it as the master sample. Note: The same part can be chosen if a stability study was completed. Measure the part a minimum of 10 times in a controlled environment (gage room), and compute the average of readings. This average is used as the reference value . Enter data into template Linearity Study design: The collection of samples for this study expands on the collection used for the Bias study. Whereas the method used for the bias study choses a part from the mid-range of the production measurements, the linearity study expands this collection to include a minimum of 5 parts from the entire range of the production measurements.

9 Enter data into template Study Results: For the Minitab study results below (5) parts were chosen that represented the expected range of measurements. Each part was measured by layout inspection to determine its master Measurement , and then one operator randomly measured each part 12 times. Total Variation Study Variation from a Crossed Gage R&R study using the ANOVA was used as the input Process Variation ( ). All rights reserved under the copyright laws. Printed copies of this document are uncontrolled. DEFENSE SEGMENT Instruction Guide Interpretation of Linearity Results: Because the R2 value is high ( ), we can see that the relationship between the master part Measurement and bias is close to linear.

10 Therefore it is reasonable to assess linearity for this data set. However since the % linearity is relatively large ( ), this indicates a problem. As can be seen from the graph, smaller parts tend to measure too high, while larger parts tend to measure too low. A slope close to zero (horizontal line) indicates there is no linearity problem, while a slope that is not close to zero (vertical line) indicates a linearity problem. The greater the slope, the worse the linearity. Interpretation of Bias Results: The average bias of ( ) and %bias ( ) indicate that there is some variability due to bias. Also the p-value is Because the p-value is less than the chosen level of significance (a-level), typically set at , we reject the null hypothesis that bias equals 0.