A Genetic Algorithm-Based Traffic Light Optimization Model for Efficient Home Healthcare Service Delivery in Türkiye

Abstract

Home healthcare (HHC) has become a crucial service model to address the rising needs of aging populations and patients with chronic conditions. However, efficient planning and resource allocation remain major challenges, especially in geographically dispersed regions. This study proposes a novel optimization-based operational model incorporating a traffic light algorithm to prioritize patient visits based on health status in Diyarbakır, Türkiye. The algorithm classifies patients into three categories (green, yellow, and red) allowing proactive and dynamic care management. A genetic algorithm is applied to solve the complex multi-objective routing and scheduling problem while considering numerous real-world constraints such as minimum team size, gender composition, and vehicle capacity. The model integrates demographic data from 2011–2023 and minimizes total visit duration while maximizing the number of patients served. Key decision variables include team size, staff gender distribution, patient condition, location, and travel time. The optimization process demonstrates significant improvements in performance metrics across generations, reducing penalty values and achieving more balanced, efficient outcomes. Results indicate that the model effectively aligns healthcare delivery with patient needs, operational limitations, and service quality goals. Unlike previous studies focusing mainly on cost or time, this model uniquely emphasizes clinical prioritization through color-coded patient conditions, integrating cultural and practical constraints. The study highlights the importance of tailored, region-specific solutions and offers a framework that can be adapted for broader applications. Future work should explore integrating machine learning for dynamic risk scoring and incorporating logistical elements such as traffic and real-time availability.