INDICES Cg AND Cgk IN THE ASSESSMENT OF THE MEASURING DEVICE CAPABILITY

Abstract

The requirements for the measuring accuracy are increasing and at the same increasing also the requirements for the assessing of the quantifiable characteristics of the measuring devices, respectively measurement systems. The reviewing of the capability of the measuring device is one of the important methods of the quality management process. Only in the case when the measuring device is capable we can proceed to the assessment of the capability of the measuring and consequently also the production process. The article deals with the assessment of the capability of the measuring device by means of C^sub g^ and C^sub gk^ indices witch can determine the capability of the measuring device for the purpose. The conclusion of the article deals with the calculation of the indices by software R.

Headnote

Abstract: The requirements for the measuring accuracy are increasing and at the same increasing also the requirements for the assessing of the quantifiable characteristics of the measuring devices, respectively measurement systems. The reviewing of the capability of the measuring device is one of the important methods of the quality management process. Only in the case when the measuring device is capable we can proceed to the assessment of the capability of the measuring and consequently also the production process. The article deals with the assessment of the capability of the measuring device by means of C^sub g^ and C^sub gk^ indices witch can determine the capability of the measuring device for the purpose. The conclusion of the article deals with the calculation of the indices by software R.

Keywords: capability of measuring device, C^sub g^ and C^sub g^k indices, qualityTools package

(ProQuest: ... denotes formulae omitted.)

1. INTRODUCTION

Measurement is defined as "the assignment of numbers [or values] to material things to represent the relations among them with respect to particular properties." The process of assigning the numbers is defined as the measurement process, and the value assigned is defined as the measurement value. Measuring devices in general mean technical devices needed for the execution of measuring. Any measuring instrument or gage, reference material and auxiliary equipment inevitable for the execution of measuring process can be considered as a measuring device. Measurement system is the collection of instruments or gages, standards, operations, methods, fixtures, software, personnel, environment and assumptions used to quantify a unit of measure or fix assessment to the feature characteristic being measured; the complete process used to obtain measurements.

At the time when measuring is an inseparable part of many areas of human activity, it is a very important goal to reach correct and fair results with the corresponding accuracy. All measuring devices fall under regular calibration in terms of the Metrology manual. In practice it happens that the daily use of these measuring devices decreases the reliability of the measuring, which in the end can lead to incorrect results. Neither huge investment into new measuring devices, nor longer time spent by the measuring process may not be a guarantee of the high quality output. Just for these situations there are methods, which can confirm or deny the suspicion of the loss of the capability of the measuring device.

2. METHODS

Measurement System Analysis (MSA) focuses on understanding the measurement process, which considers the defect size and suitability of the measuring device, resp. measuring system.

The capability of the measuring device predicates of its functional capability and of the correctness of the measured data. It characterizes its appropriateness for the measuring of a particular attribute in the given scale. One of the methods of the assessing of capability is the determination of the capability of the measuring device by means of capability indices Cg and Cgk. These indices assess the measuring device from the point of view of and repeatability. Bias represents the difference between the accepted reference value and mean value of the results of the experiments. Measuring repeatability represents the closeness of the compliance between the results of the consecutive measurements realised at the same conditions of the measuring [2, 4].

This procedure of the determination of the capability is used at the measuring devices, at which no influence of the results by the measuring attendance occurs. It is based on the repeated measuring of the selected size of product, the control standard, whose nominal value lies in the middle of the tolerance of the measured parameter. The measuring is performed by one person with one measure and the same procedure in a relatively short time interval, whereby it is recommended to perform 50 (at least 25) repeated measurements. Within the measuring it is necessary to ensure the same conditions. It is assumed that the results of the measuring are governed by normal distribution [3, 8], The calculation of the Cg index (index of repeatability) of the monitored measuring device is set by the relation

... (1)

... (2)

where xi is z-th measured value, n number of measurements, x g average value of measurements, sg selective standard deviation of measuring and T represents tolerance of the measured dimension [1, 3, 6, 9], For the index Cgk (index of bias and repeatability) of the measuring device reads relation

... (3)

where xr is a reference value, which is preferably selected in the way that is selected in preference so that it lies in the middle of the tolerance of the measured dimension [1, 3, 6, 9],

The indices determine if the result of the measuring of the control standard lies with the probability of 99.73% in the selected tolerance zone of the scale of measuring device, which is set to 20% of the width of the tolerance of the measured dimension. The value of the index Cg makes only provision for the consistency of the measuring, the value Cgk consistency and also systematic deviation of the measuring process from the reference value given by the standard (partiality). It follows from the definition of the indices that Cg > Cgk.

If it reads that Cg> 1.33 and at the same time Cgk> 1.33 (for T > 50 pm), then the measuring device is proper and can be used in the production process. In case T > 50 pm the fulfilment of the conditions Cg > 1 and concurrently Cgk > 1 is sufficient. If the conditions are not fulfilled, then it is not possible to include the measuring device into the production process and it is necessary to make corrective measures (repair or adjustment of the measuring device, training of operators and others.) In general we can calculate indices Cg and Cgk by relation [2, 6]

... (4)

where k is a selected percentage of tolerance (e. g. in methodology by BOSCH k = 20 is chosen, by methodology FORD k = 15).

In [10] besides indices also repeatability %R and accuracy and repeatability %A&R are introduced, for which the following relations [10]

* ..., where the condition %R ≤ 15 must be fulfilled, (5)

* ..., where the condition 0 ≤ %A&R ≤ 15 must be fulfilled. (6)

Between value of repeatability %R and calculation of accuracy and repeatability %A&R is valid %R < %A&R.

3. IMPLEMENTATION OF THE ASSESSMENT OF CAPABILITY OF MEASURING DEVICE IN PRACTICE

In electronic management/control of engine charging for spark engines an electronically operated unit of flap is a central regulating member. It consists of a throttling flap with an electric drive and an angular sensor of the throttling flap (Figure 1). The monitored product is a part of the production process of an international organisation, which focuses on the design and production of highly - technological systems and components for car industry. The measuring took place in the production area of the enterprise at the temperature 22°C and humidity 50%. The goal of measuring is to check the capability of line measuring device (Figure 2).

In the measurement of the chosen size (in mm) was made 30 repeated measurements. Figure 2 Measuring device and measuring of the The measured values and basic numeric parameters of the throttling flap characteristics and sg are given in the table. The value of the tolerance is determined by the organisation to 0,05 mm (USL=40.024 mm, LSL=39.975 mm).

In the measurement of the selected size (in mm) were performed 30 repeated measurements.

The index of repeatability Cg of the measuring device is calculated by the relation:

...

The index of the bias and repeatability Cgk of the measuring device is expressed by the relation:

...

For the repeatability %R and accuracy and repeatability %A&R it reads

...

Because both conditions are met (Cg > 1.33 and at the same time Cgk > 1.33, equally also %R < 15, 0 < %A&R < 15 and %R < %A&R) measuring device is suitable and can be used in the production process.

4. PACKAGE QUALITYTOOL

For the assessment of capability we can also use the free software R. R is a language and an integrated environment for data analysis, statistical and mathematical computing, and graphical data processing and representation. This free software is very similar to the S language and runs on most UNIX/Linux platforms, Macintosh and Windows. R belongs to open source software and its latest version can be downloaded directly from its home page at http://www.r-project.org/.

R is a tool for accomplishing many conventional as well as modern statistical computing and analytical tasks. The basic environment contains several standard, recommended packages, and many more are available through the CRAN archive on the home page, including updates.

There are many packages for process quality assessment through statistical methods. For example, with the use of package quality Tools we can simply calculate indices Cg and Cgk [5]. Graphic output of the R package is in Figure 3. The assessment of capability can be analyzed by using the cg() function.

>libr ary (quality Tools)

>data=c(40.0005/40.0004/40.0001/40.0011/39.9999/40.0003/39.9998/40.0009/39.9988/40.0009/39.9999/40.0 001,40.0002,39.9999,40.0013,40.0011,39.9996,40.0012,40.000140.0004,40.0006,40.0012,39.9988,40.0009, 40.0001,39.9998,40.0011,39.9999,40.0009,39.9999)

>cg(data,target=40.0000,tolerance=c(40.025,39.975))

Conclusion

It is important to realize that for the ensuring of very exact parameters, attributes of various products it is not enough to deal with technological aspects of production process only but it is also important to ensure with the same concern the reliable measuring of the monitored parameters. The assessment of the capability of the measuring device is a process, which represents an important element of the consistent quality improvement of the production process, because the decisions on the quality are usually based on the control and measuring of various parameters of the products. The goal of such assessment is to prove that the measuring device is capable and suitable for measuring of the monitored attribute of quality. It provides for a possibility to identify the inadequacies as early as the initial phase of the production process and it helps to ensure the fulfilment of the quality requirements.

The given method of the assessment of the capability of the measuring device by means of indices Cg and Cgk is used at the complex monitoring of the capability of the production processes and at regular assessment of the capability of the measuring systems as a part of metrology management.

Acknowledgements

The work presented in this paper was supported by the Project VEGA 1/0922/12.

References

References

[1.] Dietrich, E.: Fähigkeitsnachweis von Messsystemen. In: http://www.q- das.com/fileadmin/files2/qdasguidelinemsa/Leitfaden_v21_me.pdf.

[2.] Fabiab, J. et al.: Statistical methods of quality control. Ceská spolecnosf pro jakost, Praha. 2007.

[3.] Hrubec, J.: Capability of instruments. In: http://josef.posta.sweb.cz/KONF/Hrubec.doc /.

[4.] Pernikar, J.: Evaluation of the capability of control means. In: http://gps.fme.vutbr.cz.

[5.] Roth, T.: Package qualityTools. In: http://cran.r-project.org/web/packages/qualityTools.

[6.] Shojae, R. et al.: Statistical Measurement system analysis of Ruston TA1750 Gas Turbine 1st Stage Nozzle. In.: Advances in Mathematical and Computational methods. In.: http://www.wseas.us/e- library/conferences/2010/Faro/MACMESE/MACMESE-03.pdf.

[7.] Technical support documentation. Rationale and Methodology for the Type 1 Gage Study.

[8.] Tavodová, K.: Competency assessment of the instrument. Acta Facultatis Technicae, XVI, 2011(1), Zvolen, s. 143-151., 2011.

[9.] Wisweh,L., Sandau, M.: Quality evaluation of measurement Instruments. In: http://home.mit.bme.hu/~kollar/IMEKO-procfiles-for-web/congresses/WC-16th-Wien- 2000/Papers/Topic%2025/Wisweh.PDF.

[10.] Internal documentation. Directive MSA.