Analysis of Urban Waste Management Vehicle Routing Problem and Its Variants Based on Genetic Algorithm

Abstract

This study addresses the intricacies of Vehicle Routing Problems (VRP) in a complex road network through a comprehensive genetic algorithm-based approach. Initially, a grid map simulation was constructed, incorporating ten waste disposal points to mimic real-world garbage collection routes. The first phase of the study focused on a basic VRP, applying a genetic algorithm to derive the shortest possible driving distances and routes for each vehicle, which were visually represented. Subsequently, the research enhanced this model by introducing constraints related to vehicle payload and disposal point capacities. These modifications, including changes in chromosome representation and fitness function, led to an advanced capacity-constrained VRP model. Further refinements were made by incorporating time windows and processing times as additional constraints, evolving the model to address both time and capacity limitations in VRP. The final phase of the study expanded the model to cater to dynamic routing scenarios, responding to unforeseen events in real-time. This paper presents a holistic approach to VRP, offering scalable solutions adaptable to various constraints and real-world complexities. The findings are represented graphically, providing clear insights into the efficiency of the proposed models.