Scholarlink

<p>The main objective of the present study is to predict the optimal availability of biscuit manufacturing plants using genetic algorithm (GA), particle swarm optimization (PSO), and Hybrid GA-PSO algorithms. The biscuit manufacturing system (BMS) is a complex industrial entity having six components configured in series. For availability investigation, a novel stochastic framework is developed for BMS. Under various operating situations, the dynamic behaviour of the plant is captured by this framework. Markov birth–death process is used to model the system's behaviour. As well, the most crucial subsystem is identified by employing a thorough reliability, availability, maintainability, and dependability (RAMD) examination. By considering component failure and repair rates as exponentially distributed, the performance of the BMS evaluated under a set of assumptions. It is revealed that the hybrid GA-PSO provides the optimal availability 0.99950181 at population size 50 after 5 iterations and it outperforms over GA and PSO. The robustness of the algorithm shown with the help of descriptive summary statistics as well as with non-parametric statistical tests. It is observed that the convergence rate of hybrid GA-PSO is very fast in comparison to traditional GA and PSO. The same methodology may be opted for performance evaluation of similar kinds of other manufacturing industries.</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

Stochastic modeling and availability optimization of biscuit manufacturing plants using Markovian approach and hybrid GA-PSO algorithm

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年6月18日

查看全文
  • Ashish Kumar,
  • Naveen Kumar,
  • Monika Saini

摘要

The main objective of the present study is to predict the optimal availability of biscuit manufacturing plants using genetic algorithm (GA), particle swarm optimization (PSO), and Hybrid GA-PSO algorithms. The biscuit manufacturing system (BMS) is a complex industrial entity having six components configured in series. For availability investigation, a novel stochastic framework is developed for BMS. Under various operating situations, the dynamic behaviour of the plant is captured by this framework. Markov birth–death process is used to model the system's behaviour. As well, the most crucial subsystem is identified by employing a thorough reliability, availability, maintainability, and dependability (RAMD) examination. By considering component failure and repair rates as exponentially distributed, the performance of the BMS evaluated under a set of assumptions. It is revealed that the hybrid GA-PSO provides the optimal availability 0.99950181 at population size 50 after 5 iterations and it outperforms over GA and PSO. The robustness of the algorithm shown with the help of descriptive summary statistics as well as with non-parametric statistical tests. It is observed that the convergence rate of hybrid GA-PSO is very fast in comparison to traditional GA and PSO. The same methodology may be opted for performance evaluation of similar kinds of other manufacturing industries.