Scholarlink

<p>Conventional planning for highway communication networks prioritizes coverage, often overlooking critical factors such as disaster resilience and equitable service distribution. To address this gap, we propose a novel hybrid optimization framework that integrates Mixed-Integer Linear Programming with a Non-dominated Sorting Genetic Algorithm to optimize coverage, risk exposure, and fairness simultaneously. Unlike baseline models a roadside unit (RSU)-only deployment and a coverage-driven drone base station (DBS) allocation our approach explicitly embeds AI-driven risk prediction and fairness constraints into the decision-making process. Numerical experiments demonstrate the framework’s superiority. While the RSU-only model saturates at ≈&#xa0;72% coverage with over 25% risk exposure, and the baseline DBS model achieves nearly 90% coverage at the cost of high risk (≈&#xa0;21%) and fairness disparity (<i>Gini</i> ≈ 0.31), our model achieves a balanced trade-off. It delivers ≈&#xa0;85% coverage, nearly halves the risk exposure to ≈&#xa0;12%, and significantly improves fairness (<i>Gini</i> ≈ 0.18). Sensitivity analyses on the risk weight <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(\beta \)</EquationSource> </InlineEquation>, maximum drone range <InlineEquation ID="IEq2"> <EquationSource Format="TEX">\({D}_{\text{max}}\)</EquationSource> </InlineEquation>, and communication radius <InlineEquation ID="IEq3"> <EquationSource Format="TEX">\({R}_{\text{comm}}\)</EquationSource> </InlineEquation> demonstrates the framework’s adaptability to local priorities and resource constraints. The AI component operates as a calibrated statistical risk estimator rather than a standalone learning system, ensuring transparent, reproducible, and methodologically grounded integration within the optimization framework. Prioritizing resilience and fairness over marginal coverage gains leads to safer, more reliable highway networks. It lays the foundation for risk-aware, real-time optimization of future connected mobility systems.</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

Reliability-Aware Multi-Objective Optimization for Sustainable Smart Highway Systems

  • 期刊论文

  • 发表日期：

    2026年3月14日

查看全文
  • Armin Mahmoodi,
  • Said M. Easa

摘要

Conventional planning for highway communication networks prioritizes coverage, often overlooking critical factors such as disaster resilience and equitable service distribution. To address this gap, we propose a novel hybrid optimization framework that integrates Mixed-Integer Linear Programming with a Non-dominated Sorting Genetic Algorithm to optimize coverage, risk exposure, and fairness simultaneously. Unlike baseline models a roadside unit (RSU)-only deployment and a coverage-driven drone base station (DBS) allocation our approach explicitly embeds AI-driven risk prediction and fairness constraints into the decision-making process. Numerical experiments demonstrate the framework’s superiority. While the RSU-only model saturates at ≈ 72% coverage with over 25% risk exposure, and the baseline DBS model achieves nearly 90% coverage at the cost of high risk (≈ 21%) and fairness disparity (Gini ≈ 0.31), our model achieves a balanced trade-off. It delivers ≈ 85% coverage, nearly halves the risk exposure to ≈ 12%, and significantly improves fairness (Gini ≈ 0.18). Sensitivity analyses on the risk weight \(\beta \) , maximum drone range \({D}_{\text{max}}\) , and communication radius \({R}_{\text{comm}}\) demonstrates the framework’s adaptability to local priorities and resource constraints. The AI component operates as a calibrated statistical risk estimator rather than a standalone learning system, ensuring transparent, reproducible, and methodologically grounded integration within the optimization framework. Prioritizing resilience and fairness over marginal coverage gains leads to safer, more reliable highway networks. It lays the foundation for risk-aware, real-time optimization of future connected mobility systems.