The measurement system is used to measure the measured characteristics quantitatively or qualitative evaluation A collection of instruments or measuring tools, standards, operations, methods, fixtures, software, personnel, environment and assumptions; The whole process used to obtain measurement results. In the ISO/TS16949 standard, a measurement systems analysis method (MSA for short) is provided, which is one of the five manuals of the standard.
Measurement system analysis There is also special software for analysis. ·Determine whether the data used is reliable:
·Evaluate new measuring instruments
·Compare two different measurement methods
·Evaluate the measurement methods that may have problems
·Identify and resolve measurement system errors
1) Measuring tool: any device used to obtain measurement results.
2) Measurement system:
Measuring tool (equipment)
Surveyor
Measured workpiece ( parts ) Procedure, methods
Interaction of the above points
The ideal measurement system should only produce "correct" measurement results each time it is used. Each measurement result should always be consistent with a standard value. A measurement system that can produce ideal measurement results should have the statistical characteristics of zero variance, zero bias and zero probability of error classification of any measured product.
Statistical characteristics of the measurement system
1. The measurement system must be under statistical control, which means that the variation in the measurement system can only be caused by ordinary reasons rather than special reasons. This can be called statistical stability.
2. The variation of the measurement system must be smaller than that of the manufacturing process.
3. The variation shall be less than Tolerance zone 。 four measurement accuracy It shall be higher than the one with higher accuracy in process variation and tolerance zone. Generally, the measurement accuracy is one tenth of the one with higher accuracy in process variation and tolerance zone. 5. The statistical characteristics of the measurement system may change with the change of the measured items. If so, the maximum variation of the measurement system should be less than the smaller of the process variation and the tolerance zone.
Standards for measurement system analysis
1. National standards;
2. First level standards (connecting national standards with private companies, scientific research institutions, etc.);
3. Level II standards (transferred from Level I standards to Level II standards);
4. Working standard (transferred from the second level standard to the working standard).
ArcGIS measurement system
Spatial Analyst does not distinguish between four different types of measured values when processing or operating values. ratio Values can perform most mathematical operations, but if multiplication, division, or square root operations are performed on interval values, ordinal values, and nominal values, the results are usually meaningless. On the other hand, it is meaningful to perform subtraction, addition, and Boolean operations on interval values and ordinal values. raster data Attribute processing within sets and between raster data sets is most efficient when using nominal measurement systems. ratio
The value of the ratio measurement system is expressed as a linear proportion of a value to a fixed zero point. These values can be mathematically manipulated to produce predictable and meaningful results. Examples of ratio measurements include age, distance, weight, and volume.
interval
Time, calendar year, Fahrenheit scale, and pH are examples of interval measurements. These values are linearly calibrated, but they are not proportional to the true zero in time or space. Since there is no real zero point, the exact value of ratio and proportion is not significant, although it is possible to compare the measured values relatively.
Ordinal number
The order value can determine the position. These measurements can show positions (such as the first, second, and third), but do not establish magnitude or relative scale. It is impossible to indicate the degree of good, bad, beautiful, healthy or strong things by ordinal numbers. For example, the speed of the first ranked athlete in a competition may not be twice that of the second ranked athlete. It is impossible to know how much faster the first runner is than the second runner just by distinguishing the winner from the position.
nominal
The values associated with this measurement system are used to distinguish between instances. These values can also establish groups, classes, members, or categories associated with objects. These values represent traits rather than quantities, so they are independent of fixed points or linear proportions. A coding scheme for land use, soil type or any other attribute can be used as nominal measurements. Other nominal values include social security number, postal code and telephone number.
