Scholarlink

<p>Load sharing systems, where multiple components work together to support a shared load, are widely used across various engineering and industrial applications. In many of these systems, the load distribution among the components follows an unequal load share rule, significantly impacting system reliability and performance. This paper examines a <i>k</i>-out-of-<i>m</i> load sharing system, modeled using the accelerated failure time model specifically tailored for unequal load sharing rules. We illustrate our findings with three cases: 1-out-of-3, 2-out-of-3 and 2-out-of-4 systems. The system components are assumed to be identical with Exponential and Weibull baseline distributions. We use the maximum likelihood estimation and two step estimation procedures to estimate the model parameters and evaluate the performance of these estimates through a simulation study, assessing bias and mean square errors. Additionally, we demonstrate the practical applicability of the model by analyzing real datasets.</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

Modeling Unequal Load-Sharing in k-out-of-m Systems Using Accelerated Failure Time Models

  • 期刊论文

  • 发表日期：

    2026年2月24日

查看全文
  • Sukumar V. Rajguru,
  • Santosh Shashikant Sutar

摘要

Load sharing systems, where multiple components work together to support a shared load, are widely used across various engineering and industrial applications. In many of these systems, the load distribution among the components follows an unequal load share rule, significantly impacting system reliability and performance. This paper examines a k-out-of-m load sharing system, modeled using the accelerated failure time model specifically tailored for unequal load sharing rules. We illustrate our findings with three cases: 1-out-of-3, 2-out-of-3 and 2-out-of-4 systems. The system components are assumed to be identical with Exponential and Weibull baseline distributions. We use the maximum likelihood estimation and two step estimation procedures to estimate the model parameters and evaluate the performance of these estimates through a simulation study, assessing bias and mean square errors. Additionally, we demonstrate the practical applicability of the model by analyzing real datasets.