Scholarlink

<p>This research introduces a simple yet efficient evolutionary algorithm (EA) specifically designed to address the Quadratic Assignment Problem (QAP), a notoriously difficult optimization problem frequently encountered in practical domains such as facility layout planning, resource allocation, circuit board design, and logistics. EAs excel at exploring the solution space while focusing on promising regions, mitigating the risk of getting stuck in suboptimal solutions. The proposed hybrid EA (augmented with a hill-climbing local search) incorporates carefully designed operators and mechanisms tailored to the QAP. Rigorous tests on the QAPLIB standard library of benchmark instances, originally provided by CORL@L, demonstrate its competitive performance, striking a balance between solution quality and computational efficiency compared to cutting-edge methods. The approach requires minimal parameter tuning and has a very low implementation complexity, making it practical and convenient to use. This research highlights the significant potential of carefully adapted EAs for effectively addressing complex combinatorial optimization challenges, as exemplified by their successful application to the QAP.</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

An evolutionary algorithm tailored to the quadratic assignment problem

  • 期刊论文

  • 发表日期：

    2025年10月10日

查看全文
  • Wissam Marrouche

摘要

This research introduces a simple yet efficient evolutionary algorithm (EA) specifically designed to address the Quadratic Assignment Problem (QAP), a notoriously difficult optimization problem frequently encountered in practical domains such as facility layout planning, resource allocation, circuit board design, and logistics. EAs excel at exploring the solution space while focusing on promising regions, mitigating the risk of getting stuck in suboptimal solutions. The proposed hybrid EA (augmented with a hill-climbing local search) incorporates carefully designed operators and mechanisms tailored to the QAP. Rigorous tests on the QAPLIB standard library of benchmark instances, originally provided by CORL@L, demonstrate its competitive performance, striking a balance between solution quality and computational efficiency compared to cutting-edge methods. The approach requires minimal parameter tuning and has a very low implementation complexity, making it practical and convenient to use. This research highlights the significant potential of carefully adapted EAs for effectively addressing complex combinatorial optimization challenges, as exemplified by their successful application to the QAP.