Scholarlink

<p>The Vehicle Routing Problem with Time Windows (VRPTW) represents a critical challenge in Home Health Care (HHC), where punctuality and operational efficiency are essential to ensure patient well-being and service quality. This paper introduces a pragmatic hybrid optimization framework specifically designed for HHC operations, integrating lightweight temporal-spatial clustering with multi-route tabu search within a cluster-first, route-second architecture to improve computational tractability and service responsiveness. Our methodology structures the routing process into three phases: demand segmentation through time-aware clustering, construction of initial feasible routes via a greedy heuristic, and iterative route improvement using a multi-route tabu search algorithm. To systematically evaluate the contribution of temporal refinement and urgency management, three progressively enhanced variants are developed: a baseline clustering model (M1), a temporally refined sub-clustering extension (M2), and a priority-aware variant (M3) incorporating representative-based initialization and differentiated penalty mechanisms reflecting heterogeneous patient criticality. Experimental evaluation on a synthetic benchmark of 40 instances, with problem sizes ranging from 50 to 200 clients, demonstrates that the priority-aware variant achieves a 71% reduction in average delay for high-priority patients and significant decreases in time-window violations, while maintaining competitive travel costs and fleet utilization levels. Comparative experiments against established metaheuristics, including GRASP and Adaptive Large Neighborhood Search (ALNS), indicate that the proposed framework provides comparable routing quality with lower computational effort under identical iteration budgets. Rather than aiming to outperform state-of-the-art solvers in pure routing cost, the objective is to offer an interpretable and computationally efficient patient-centered decision-support approach tailored to HHC constraints. Overall, the results show that coupling structured clustering with interpretable metaheuristic optimization provides a scalable solution for large-scale, time-sensitive home healthcare logistics.</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

Efficient Home Health Care Vehicle Routing with Time Windows via Hybrid Clustering and Tabu Search

  • 期刊论文

  • 发表日期：

    2026年6月30日

查看全文
  • Mohamed Salim Amri Sakhri

摘要

The Vehicle Routing Problem with Time Windows (VRPTW) represents a critical challenge in Home Health Care (HHC), where punctuality and operational efficiency are essential to ensure patient well-being and service quality. This paper introduces a pragmatic hybrid optimization framework specifically designed for HHC operations, integrating lightweight temporal-spatial clustering with multi-route tabu search within a cluster-first, route-second architecture to improve computational tractability and service responsiveness. Our methodology structures the routing process into three phases: demand segmentation through time-aware clustering, construction of initial feasible routes via a greedy heuristic, and iterative route improvement using a multi-route tabu search algorithm. To systematically evaluate the contribution of temporal refinement and urgency management, three progressively enhanced variants are developed: a baseline clustering model (M1), a temporally refined sub-clustering extension (M2), and a priority-aware variant (M3) incorporating representative-based initialization and differentiated penalty mechanisms reflecting heterogeneous patient criticality. Experimental evaluation on a synthetic benchmark of 40 instances, with problem sizes ranging from 50 to 200 clients, demonstrates that the priority-aware variant achieves a 71% reduction in average delay for high-priority patients and significant decreases in time-window violations, while maintaining competitive travel costs and fleet utilization levels. Comparative experiments against established metaheuristics, including GRASP and Adaptive Large Neighborhood Search (ALNS), indicate that the proposed framework provides comparable routing quality with lower computational effort under identical iteration budgets. Rather than aiming to outperform state-of-the-art solvers in pure routing cost, the objective is to offer an interpretable and computationally efficient patient-centered decision-support approach tailored to HHC constraints. Overall, the results show that coupling structured clustering with interpretable metaheuristic optimization provides a scalable solution for large-scale, time-sensitive home healthcare logistics.