What is MSA?
A Measurement system is a process by which we assign a number to a characteristic of a product or service. The first step in assessing a system is to understand this process, and determine whether it will satisfy our requirements.
Measurement System Analysis, often referred to as MSA, is used to assess the statistical properties of process measurement systems.
- MSA primarily deals with analyzing the effect of the measurement system on the measured value in quantifiable terms
- Emphasis is on the effect due to equipment and personnel
- We test the system to determine the numerical values of its statistical properties and compare them to accepted standards. It is a process used to standardize the methods of analysis to ensure and demonstrate that your measuring systems have adequate resolution, do not excessively bias results, and possess a small variability in comparison with specified tolerances. It helps you assess, monitor, and reduce measurement system variation. The objective of a measurement system analysis study is to make sure that your measurement system - gages, methods, and procedures are stable and capable of measuring data, before continuing with your process improvement efforts. It ensures that your measuring procedures and systems provide:
- Adequate resolution
- Results that are not unduly biased, and
- Little variability in comparison with specified tolerances
To evaluate a measurement system determine:
- If it has adequate discrimination
- If it is statistically stable over time
- If statistical properties are consistent over the expected range and acceptable for process analysis or control
- If the sum of all variables is an acceptable level of measurement uncertainty
Overall Objective of MSA
Uncertainty of Measurement
It is the range within which the true value of a characteristic is estimated to lie. Such data can be expressed as the statistical distribution of a series of measurements, standard deviations, probability, percentages, and error as the difference between actual value minus the true value, as points on a control chart or diagram.
- Determining these fundamental issues are most meaningful if made relative to process variation
- Reporting measurement error as only percent of tolerance is inadequate for the worldwide market where emphasis is on continual process improvement
Use of Data in Measurement System Studies
The data collected using a measurement system is used:
- To control process
- In estimating the existence of relationship between variables that can affect the outcome of a process
- To conduct analytical studies to increase the knowledge about the system of causes and its effect on processes
- To focus on measurement systems where readings can be repeated on each part, and reproduced by different operators.
A measurement is not always exact. Measurement system variation affects individual measurements and decisions based on data. Measurement system errors are classified into five categories: bias, repeatability, reproducibility, stability, and linearity. You need to know the extent of variation before deciding on the following applications.
- Establish criteria for suitability and acceptability of new measuring equipment
- Compare one measuring device against another
- Evaluate suspect equipment
- Compare the performance of an equipment before and after its repair
- Calculate measurement system variation
- Establish acceptability of manufacturing process
- Manage & improve the measurement process
Where to start?
- Evaluate the components of the measuring system, and control the variation in them as much as possible to ensure that an item of measuring equipment complies with the requirements for its intended use
- Expand your consideration of Measurement Process Variation to Measurement System Statistical Properties & Measurement Uncertainty.
- Follow the basics of SPC.
Process Accuracy Measurements
Stability (or drift) is total variation in measurements obtained with a measurement system on same master or parts when measuring a single characteristic over an extended time period (a time period is days, not hours). Stability is the key to predictability. Stable processes are those that are free from special cause variation. In terms of measuring equipment, stability is determined by using a control chart. As measurements are taken, points within the limits indicate that the process has not changed and the prediction is made that it is not likely to change in the future. Statistical process control (SPC), scatter plots, or other forms of statistical analysis are used to measure process stability.
Linearity is the difference in the accuracy values through the expected operating range of the equipment. Selecting the parts throughout the operating range of the instrument can determine the linearity. The accuracy of these parts is determined by the difference between the master measurement and the observed average measurement.
Difference between observed average of measurements and reference value. The reference value, also known as accepted reference value or master value, is a value that serves as an agreed upon reference for measured values. A reference value can be determined by averaging several measurements with a higher level of measuring equipment.
Process Precision Measurements
Gage R&R statistically isolates different types of variation in the measurement process. These types of variation include:
- Repeatability = equipment variation = within variation
- Reproducibility = appraiser variation = between variation
- Residual or pure error
- Variation due to interaction effects. For example, out of several inspectors, one might have a tendency to read one gage differently than others.
Gage R&R- Gage Repeatability and Reproducibility can be applied to any kind of measurement (attribute or variables, indeterminate or determinate). The two most common methods used and supported by statistical software are the ANOVA method (Analysis Of Variance) and the average and range method.
Repeatability refers to the variation in measurements obtained with one measurement instrument when used several times by one assessor while measuring the identical characteristics on the same part.
Reproducibility refers to the variation in the average of measurements made by different assessors using the same measuring instrument while measuring the identical characteristics on the same part.
R&R is the combination of repeatability and reproducibility variation, and is considered as the total measurement variation excluding within part variation and variation in central location.
A material or substance with one or more properties, which are sufficiently well established to be used for the calibration of an apparatus, assessment of a measurement method, or for assigning values to materials.
- Measurement Uncertainty is the sum of all the probabilities (percents) assigned to the variables that make up the measurement system.
- The total of these probabilities should be weighed, and carry importance in proportion to the seriousness, and criticality of the measurements being made.
- Decisions resulting from measurement system analysis include:
- Using the system as is, taking into account its uncertainty.
- Improving the system to control the variation in the contributing factors.
- Considering other measurement systems of higher levels of discrimination and capability.
Product and process conformance are determined by the measurements taken by a measurement system. If the measuring process is changing over time, the ability to use the data gathered in making decisions is reduced. In essence, 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.