Scholarlink

<p>In today’s industrial manufacturing, two-sided assembly lines are commonly employed to produce large-scale, high-volume products such as automobiles and trucks. This research develops three mixed-integer linear programming models and one constraint programming model to address the type-I two-sided assembly line balancing problem (TALBP-I), aiming to minimize both the number of mated stations and workstations. A branch, bound, and remember (BBR) algorithm and a beam search (BS) heuristic are then proposed to solve the TALBP-I. Since complete enumeration requires significant computation time, the proposed BBR employs a combined task enumeration method, utilizing incomplete task enumeration to achieve high-quality solutions by eliminating tasks that cause sequence-dependent idle time and imbalance between sides. The proposed BS also reduces sequence-dependent idle time and promotes load balance between sides by applying heuristic rules when extending partial solutions. A comparative study is conducted to evaluate the performance of the formulated models and the developed methods. The computational results demonstrate that the proposed methodologies outperform both published results and existing algorithms re-implemented for a fair comparison. Moreover, BBR solves all instances optimally within an average of 10&#xa0;s, while BS also finds all optimal solutions more quickly when solving larger instances. The comparative study demonstrates that the proposed BBR and BS are the new state-of-the-art methods for TALBP-I.</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

Enhanced exact and heuristic methods for the type-I two-sided assembly line balancing problem

  • 期刊论文

  • 发表日期：

    2026年1月19日

查看全文
  • Zixiang Li,
  • Ibrahim Kucukkoc,
  • Yuxin Liu,
  • Zikai Zhang,
  • Liping Zhang

摘要

In today’s industrial manufacturing, two-sided assembly lines are commonly employed to produce large-scale, high-volume products such as automobiles and trucks. This research develops three mixed-integer linear programming models and one constraint programming model to address the type-I two-sided assembly line balancing problem (TALBP-I), aiming to minimize both the number of mated stations and workstations. A branch, bound, and remember (BBR) algorithm and a beam search (BS) heuristic are then proposed to solve the TALBP-I. Since complete enumeration requires significant computation time, the proposed BBR employs a combined task enumeration method, utilizing incomplete task enumeration to achieve high-quality solutions by eliminating tasks that cause sequence-dependent idle time and imbalance between sides. The proposed BS also reduces sequence-dependent idle time and promotes load balance between sides by applying heuristic rules when extending partial solutions. A comparative study is conducted to evaluate the performance of the formulated models and the developed methods. The computational results demonstrate that the proposed methodologies outperform both published results and existing algorithms re-implemented for a fair comparison. Moreover, BBR solves all instances optimally within an average of 10 s, while BS also finds all optimal solutions more quickly when solving larger instances. The comparative study demonstrates that the proposed BBR and BS are the new state-of-the-art methods for TALBP-I.