Scholarlink

<p>This paper introduces a new lifetime distribution derived by applying a transformation to the inverted Dagum distribution. We examine its statistical properties, including moments, hazard rate, quantile functions, and order statistics. The three unknown parameters are estimated using twelve methods, and a comprehensive Monte Carlo simulation reveals that the Kolmogorov-based estimator outperforms alternatives like maximum likelihood and least squares for some criteria. The proposed distribution demonstrates superior flexibility in modeling real data, outperforming established competitors such as the Dagum, inverted Dagum, extended Dagum, Gamma, and Weibull distributions in goodness-of-fit tests for two datasets: bladder cancer remission times and Thorium sediment measurements. Furthermore, we adapt the process capability index <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\({C}_{pk}^{*}\)</EquationSource> <EquationSource Format="MATHML"><math> <mmultiscripts> <mi>C</mi> <mrow> <mi mathvariant="italic">pk</mi> </mrow> <mrow> <mrow /> <mo>∗</mo> </mrow> </mmultiscripts> </math></EquationSource> </InlineEquation> to the new model and validate its utility in quality control through three engineering applications: electronic component failure times, carbon fiber tensile strength, and ball size measurements. The results highlight the new distribution’s practical relevance in real data analysis and quality control.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A New Lifetime Model: Properties, Estimation, and Applications in Quality Control

  • 期刊论文

  • 发表日期：

    2025年7月31日

查看全文
  • Kadir Karakaya,
  • Erdem Cankut

摘要

This paper introduces a new lifetime distribution derived by applying a transformation to the inverted Dagum distribution. We examine its statistical properties, including moments, hazard rate, quantile functions, and order statistics. The three unknown parameters are estimated using twelve methods, and a comprehensive Monte Carlo simulation reveals that the Kolmogorov-based estimator outperforms alternatives like maximum likelihood and least squares for some criteria. The proposed distribution demonstrates superior flexibility in modeling real data, outperforming established competitors such as the Dagum, inverted Dagum, extended Dagum, Gamma, and Weibull distributions in goodness-of-fit tests for two datasets: bladder cancer remission times and Thorium sediment measurements. Furthermore, we adapt the process capability index \({C}_{pk}^{*}\) C pk ∗ to the new model and validate its utility in quality control through three engineering applications: electronic component failure times, carbon fiber tensile strength, and ball size measurements. The results highlight the new distribution’s practical relevance in real data analysis and quality control.