Measurement System Reanalysis

Measurement System Reanalysis- It is an experimental and mathematical method of determining how much the variation within the measurement process contributes to overall process variability. There are five parameters to investigate in it which are bias, linearity, stability, repeatability, and reproducibility. According to Automotive Industry Action Group (AIAG-2002), a general rule of thumb for measurement system acceptability is:

Under 10 percent error is acceptable.
10 percent to 30 percent error suggests that the system is acceptable depending on the importance of application, cost of the measurement device, cost of repair, and other factors.
Over 30 percent error is considered unacceptable, and you should improve the measurement system.

AIAG also states that the number of distinct categories the measurement systems divides a process into should be greater than or equal to 5. In addition to percent error and the number of distinct categories, you should also review graphical analyses over time to decide on the acceptability of a measurement system.

Any measurement process for a system typically involves measurement precision as well as measurement accuracy of the system variables subject to the constraints of the system. Requirement for statistically analyzing a system would involve a process to determine the variations from the mean (central) location which is imperative to analyze the measurement accuracy taking into consideration factors of bias, stability and linearity. These parameters of MSA (Measurement Systems Analysis) can be described as:

Bias refers to a probability of presence of certain factors in a system which can influence deviation from the standards in the system. Bias can lead to sampling of a data which on analysis appear to be different from the actual or anticipated data set. In order to measure the process measurement bias, for determinate measurement a process called calibration is needed which is of higher level than measuring the data average. In case of indeterminate measurement process owing to constraints, normally the data average values are compared with the standard values.
Stability refers to processes which are normally free from special cause variations. Analyzing a system for stability typical involve the standard statistical processes such as SPC (Statistical Process Control), scatter plots, ANOVA techniques and other standard deviation measurement tools. Determination of stability standards in a system requires data sampled to cover a wide range of possible variation factors and intensive piece meal statistical tests covering variations in human resources, tools, parts, time, space and location factors.
Linearity refers to different statistical results from measurements when subjected to different metric spaces. Linearity in a system is determined using higher levels of calibrations in measurement standards which often guided by inferences drawn from various interaction factors influencing a system. For instance, a non linearity in a system may result from equipment (or tools) not calibrated for various levels of operating range or poor design of system or any other system constraint.

Test Your Six Sigma Black Belt Skills By Taking Our Practice Tests on This Link