Statistical tolerance intervals of distributions in tasks associated with acceptable risk

Abstract


The calculation of tolerance bounds for distributions is a necessary task to effectively address acceptable risk. This is reflected, among other things, in the requirements of international standards. The article provides theoretical information about statistical tolerance intervals and directions of their application. Formulas for calculating one-sided tolerance intervals are presented. As an example, the problem of calculating the characteristics of structural strength of materials in aircraft construction using experimental data is considered. A comparative analysis of the values of the lower tolerance limit using different laws of data distribution is carried out.The use of non-parametric methods for calculation of tolerance limits is shown. Using the basic relation for the calculated indices, the confidence probabilities are obtained, allowing to use the minimum elements of the samples as the left one-sided boundary. At given values of the confidence probability, the dependences of the data coverage values on the sample size are plotted. Statistical modelling methods were used to calculate the values of the lower tolerance boundary. The R software package was used for calculations.

Full Text

9

About the authors

A. R Davydov

Perm National Research Polytechnic University

V. V Noskova

Perm National Research Polytechnic University

References

  1. Цепелев, В.С. Основные сведения о БЖД: учебное пособие / В.С. Цепелев, Г.В. Тягунов, И.Н. Фетисов. – 3-е изд., испр. – Екатеринбург: Изд-во Урал. ун-та, 2014. – 120 с. – (Безопасность жизнедеятельности в техносфере: в 2 ч.; ч. 1).
  2. Розенталь, О.М. Сравнение методов оценивания риска ошибочного контроля качества природных вод / О.М. Розенталь, Л.Н. Александровская // Аналитика и контроль. – 2016. – Т. 20, № 1. – С. 53–61
  3. Об оценке погрешностей расчетов, выполняемых при обосновании безопасности объектов использования атомной энергии / С.Н. Богдан, О.М. Ковалевич, С.А. Шевченко, Д.А. Яшников // Ядерная и радиационная безопасность. – 2017. – № 2. – С. 38–53.
  4. Закс, Ш. Теория статистических выводов: пер. с англ / Ш. Закс. – М.: МИР, 1975. – 776 с.
  5. Первадчук, В.П. Прикладной статистический анализ результатов механических испытаний: учеб.-метод. пособие. – Ч. 1: Алгоритмы анализа и определение статистических толерантных интервалов характеристик материалов / В.П. Первадчук, А.Р. Давыдов, Т.А. Осечкина. – Пермь: Изд-во Перм. нац. исслед. политехн. ун-та, 2017. – 86 с.
  6. Cho, K.S. Tolerance intervals in statistical software and robustness under model misspecification / K.S. Cho, H.K. Tony Ng // Journal of Statistical Distributions and Applications. – 2021. – Vol. 8(10).
  7. Hanson, L. Tolerance Limits for the Class of Distributions with Increasing Hazard Rates / L. Hanson, L.H. Koopmans // The Annals of Mathematical Statistics. – 1964. – P. 1562–1570.
  8. Осечкина, Т.А. Ядерные оценки при расчетах границ толерантных интервалов / Т.А. Осечкина // Современные наукоемкие технологии. – 2019. – № 5. – С. 64–68.
  9. R Core Team. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing. – Vienna, Austria. – 2023. [Электронный ресурс]. – URL: https://www.R-project.org/ (дата обращения: 05.01.2025).
  10. Young, D.S. tolerance: An R Package for Estimating Tolerance Intervals / D.S. Young // The Pennsylvania State University. Journal of Statistical Software. – 2010. – Vol. 36, iss. 5. – P. 39.

Statistics

Views

Abstract - 14

PDF (Russian) - 4

Refbacks

  • There are currently no refbacks.

This website uses cookies

You consent to our cookies if you continue to use our website.

About Cookies