Hybrid Modeling for Optimizing Vaccine Distribution in Remote Health Stations

Abstract

This study aims to enhance the Expanded Programme on Immunization (EPI) vaccine distribution to distant Barangay Health Stations by testing a hybrid Genetic Algorithm (GA)-Particle Swarm Optimization (PSO) model. Incorporating GA's exploratory route generation and PSO's convergence guidance, the approach revealed a more efficient route, lowering delivery cost and allowing timely and reliable vaccine delivery. It simulated a hybrid Genetic Algorithm–Particle Swarm Optimization (GA–PSO) model for route planning in order to optimize the efficiency of delivery routes. GA explored new delivery routes through crossover and mutation, while PSO refined solutions toward optimal paths. The hybrid GA-PSO combined broad exploration with fast convergence, avoiding premature settling and uncovering hidden efficiencies, ultimately producing smarter, more reliable vaccine distribution strategies for remote health stations. In this study, both GA and PSO individually identified NP → NM → K with a cost of 25. The hybrid GA-PSO revealed a smarter shortcut, NP → SD → K, lowering the cost to 22. This hybrid approach achieved cheaper, faster, and more reliable vaccine distribution compared to single algorithms. It demonstrates the first application of a hybrid GA-PSO model to vaccine cold-chain logistics in remote health stations. Balancing GA’s creativity with PSO’s convergence prevents premature settling, uncovers hidden efficiencies, and offers a practical optimization framework for real-world public health supply chains.