DEVELOPMENT OF OPTIMIZATION ALGORITHM FOR A DYNAMIC WASTE COLLECTION ROUTE PLANNING IN LAGOS MUNICIPALITY

ABSTRACT The rapid urbanization of Lagos Municipality has intensified the challenges of municipal solid-waste collection, often leading to inefficient routes, high operational costs, and increased environmental impact. This research presents the development of an optimization algorithm for dynamic waste-collection route planning, integrating Genetic Algorithm (GA) and Ant Colony Optimization (ACO) into a hybrid computational model. The system aims to generate adaptive, near-optimal collection routes that minimize total distance, fuel consumption, and travel time while responding to real-time variations in waste generation and traffic flow. The study adopts a cluster-first, route-second approach in which waste bins are assigned to multiple vehicles based on capacity constraints before the hybrid GA–ACO algorithm performs route optimization. The model was implemented using Python (v3.10) and MATLAB R2023a, incorporating IoT-enabled sensor data, GIS-based mapping, and an interactive Flask web interface. Simulation results on Lagos Mainland datasets demonstrated that the hybrid model achieved up to 18.6% fuel savings and 17–20% route distance reduction compared to standalone GA and ACO algorithms. Furthermore, the system’s feedback mechanism enables dynamic route recalculation when bin-fill thresholds or traffic conditions change. This work establishes a robust computational framework for smart waste-management systems in developing cities, combining artificial intelligence, IoT, and optimization theory. The proposed model contributes both practically to improve Lagos’ waste-collection efficiency—and academically to the advancement of metaheuristic hybridization for dynamic vehicle-routing problems.