Multi-Objective Vehicle Routing Problem With Time Windows Menggunakan Genetic Algorithm

Abstract

This study focuses on the development and application of the Multi-Objective Vehicle Routing Problem with Time Windows (MO-VRPTW) model using Genetic Algorithm (GA) to optimise the waste transport system in Batu Bara Regency, North Suma- tra. The objective of this study is to design a vehicle route optimisation model that considers three main objective functions, namely minimising total distance travelled, total travel time, and risks due to road conditions, with service time windows and vehicle capacity constraints. The development stage was carried out by constructing a mathematical formulation of MO-VRPTW that realistically represents the waste transportation problem. Furthermore, the model was implemented using a Genetic Algorithm, which included the processes of initial population formation, fitness-based selection, crossover, and mutation to obtain the optimal solution. The application stage was carried out using field data, including waste collection points (TPS), final disposal sites (TPA), travel distance, travel time, vehicle capacity, and road condi- tions. The results of the study show that the Genetic Algorithm-based MO-VRPTW model is capable of producing more efficient routes, reducing total distance and travel time, and improving service punctuality. This model contributes to the development of multi-objective optimisation and practical application in sustainable logistics sys- tems.