When developing and manufacturing measurement systems it is necessary to ensure that the particular measurement and testing process is suitable for the intended use. Such a measurement process is dependent on a series of influential factors such as operating personal, environment, evaluation method, the measuring object and the mounting fixture.
To find out if a measuring and testing process is capable under those influences, a measurement systems analysis (MSA) can be carried out. In this context, measurement error or controlling of the measuring device are being assessed systematically.
Our software provides the opportunity to import and subsequently evaluate data from different sources. Data can be assorted, pictured and analyzed according to the end-users requirements.
By preparing data it is possible to draw a conclusion, e.g. for error cause at failure or for possible trends of measurement values.
Evaluation or analysis should be divided into partial analysis. These include
- statistical evaluation of measurement data regarding measuring equipment capabilities,
- statistical error distribution,
- correlation analysis and similarity analysis of measurements,
- curve comparison of measurements,
- and trend analysis of measurement progression.
Import of test reports or test executions according to the IRS XML-format. Other formats can be imported from different sources, e.g. from databases. Customer specific formats or sources can be processed through a plug-in structure.
In order to make comparisons of analytical data of specific attributes, test executions are grouped. Defined groups, sorting and used filters can be saved as predefined sets and applied by a quick selection. Results can be compared by grouping, e.g. temperature classes of climate sets or measurement progressions for long-term tests. Measurements in different test environments can be evaluated separately which is helpful for developing measuring and testing systems in order to detect errors.
History & Statistics
With this type of evaluation statistical value analysis can be implemented and visualized. Sliding limits can be displayed and limit violations highlighted. Also, statistical values are determined, e.g. average, standard deviation, specification limit, measuring equipment indices and process capability indices.
Different diagram types can be used for visualization, e.g. scatter and line diagram with variable axes (time, index or serial number, etc.). A distribution curve with overlaid histogram and all relevant parameters can be displayed as well.
In this partial analysis statistical error distribution is determined and visualized on two levels. The first level evaluates the results of test executions of a group. From the error distribution of the first level conclusions there can be concluded how the result spreads overall.
The second level evaluates the results of groups which can be compared with other groups.
The primary goal of the graph comparison function is to demonstrate correlations between different measurements. Another type of presentation is a value chart. Paramount is to compare several measurements at the same time within one group by presenting them all together in one diagram.
The linear context between measurements can be detected through similarity analysis. With the correlation analysis it can be determined if a test step is linked to another one.
Trend analysis is used for determining trends to control or average of a measurement series. It is necessary to recognize, prioritize and evaluate all trends fully automatic as the analyzing data is very large.