ПОСЛІДОВНИЙ АДАПТИВНИЙ МЕТОД ПРИЙНЯТТЯ РІШЕННЯ ПРО ПРИДАТНІСТЬ ПРОДУКЦІЇ
Journal Title: Вимірювальна техніка та метрологія - Year 2018, Vol 79, Issue 1
Abstract
The article analyzes the requirements for increasing the probability of conformity assessment of products. It is shown that to increase the probability of measuring inspection traditionally more precise measuring systems are used; perform multiple observations of the investigated quantity, taking for the result of the measurement the average arithmetic value: set tolerance intervals. An analysis was conducted of each of the ways to increase the probability of assessing the compliance of the object and showing the disadvantages of each of them. It is shown that the results of the evaluation of the object of inspection parameters are inevitably related to the uncertainty due to the imperfection of this procedure, equipment and the influence of external random factors. Reducing this effect is especially important within certain limits around the limit values of the tolerance interval with simultaneous reduction of the volume of additional measurements. Therefore, the decree is about suitability or unsuitability of the object of inspection account should be taken of both the possible uncertainty of the result and the requirements regarding the cost and complexity of the inspection procedure, as well as the given reliability. To avoid these disadvantages and limitations offered adaptive successive decision-making method for compliance. At the beginning of the decision-making process, based on the actual relationship between the length of the tolerance interval and the uncertainty of measurement, are found the initial relative tolerance limits, upper and lower, which correspond to the probability of making a decision about the correspondence of рв = 95 %. In accordance with these relative tolerance limits determine tolerance limits for the primary measurement, with which the primary measurement result is compared. In the case when the primary measurement result goes beyond the tolerance interval, they pass to the adaptive procedure, performing additional measurements each time until their average value falls within the tolerance interval. Number of successive steps determine for the permissible «residual« probability of making a false decision on the correspondence of the object. This probability depends on the parameters of the probability density function of the investigated quantity. Application of the method allows increase probability of making a correct decision about compliance for each inspected object and in comparison with known methods, does not significantly affect the decrease in productivity and increase the cost price of the product. The practical use of the algorithm has shown that the 5 additional stages of the research, makes it possible to reduce the probability of false solutions at least thrisely, an increase in inspection operations does not exceed 60 %.
Authors and Affiliations
Yevgen Volodarsky, Larysa Kosheva, Maryna Klevtsova
Keywords
Related Articles
- EP ID EP530856
- DOI 10.23939/istcmtm2018.01.058
- Views 88
- Downloads 0
Дослідження термометричного матеріалу HfNi1-xRhxSn
The electron energy state, magnetic and transport characteristics of thermometric materials HfNi1-xRhxSn were investigated in the T80¸400 K temperature range and at charge carriers concentration from Rh 9,5 1019 A N ×...
КОРОЗІЙНА СТІЙКІСТЬ МЕТАЛЕВИХ АМОРФНИХ СТОПІВ ТА МЕТРОЛОГІЧНА НАДІЙНІСТЬ ТЕРМОПЕРЕТВОРЮВАЧІВ НА ЇХ ОСНОВІ
Provided analytical review of typical investigation kinds and methods and classification of main indices of estimate of material corrosion process . Analyzed corrosion resistance of materials in different structural st...
Коригування похибок діодних цифрових термометрів
The digital diode thermometer error factors and error modelling results as multiarguments function are provide in this paper. The some ways of digital diode thermometer errors correction are discussed also.
ЗАСТОСУВАННЯ МЕТОДУ СЕРІЙ ДЛЯ ДОСЛІДЖЕННЯ ВЗАЄМНОЇ КОРЕЛЯЦІЇ СПОСТЕРЕЖЕНЬ
When processing the results of measurements big role important presence of correlation values. To find the standard uncertainty need to know the effective number of uncorrelated observations. No correlation can consider...
ІСТОРИЧНІ ПОСТАТІ У ЦАРИНІ МЕТРОЛОГІЇ ГАЛИЧИНИ – ПРОФЕСОР ДОКТОР Й. К. ЛІСҐАНІҐ (1719–1799)
In 2019, 100 years since the founding of the Department of Information and Measurement Technologies of the Lviv Polytechnic National University, there is an urgent need to look back in the historical persp...