Route optimization in urban waste management using locally adjusted discrete cuckoo search: a hybrid metaheuristic approach.

A sudden surge in urbanization and population has escalated challenges in cities in waste generation, making efficient route optimization for waste management a critical necessity. This study proposes a hybrid metaheuristic framework, Locally Optimized Discrete Cuckoo Search (LO-DCS), for an effective route optimization in urban waste management. The proposed algorithm adapts the classical Cuckoo Search algorithm to discrete routing problems by integrating permutation-based random walk, 2-opt local optimization and K-means clustering. The input data is obtained from waste bin coordinates which were extracted using Google Earth Engine and georeferenced satellite imagery within a predefined region of interest. The proposed framework was implemented on real-world urban datasets from Bengaluru city using multiple performance indicators, including travel distance, fuel consumption, carbon emission and operational time. Extensive experiments involving 30 independent runs were performed to assess stability and robustness. Comparative analysis with Genetic Algorithm (GA), Particle Swarm Optimization (PSO), Discrete Spider Monkey Optimization (DSMO) and Quantum-based Avian Navigation algorithm (QANA) demonstrates the competitive performance of LO-DCS across all the bin clusters. Statistical significance test was used to validate the results using Wilcoxon and Friedman tests with Holm correction. Furthermore, optimality gap analysis using exact solvers confirms that LO-DCS produces near-optimal solutions for moderate-sized bin-cluster instances. The experimental results show that LO-DCS achieves an average improvement of approximately 85% across the clusters for all the key performance indicators (distance, fuel consumption, CO