Delivery Route Optimization Using Dijkstra’s Algorithm for the Travelling Salesman Problem

Abstract

This paper presents an optimized delivery route planning system using a hybrid approach of Dijkstra’s algorithm and heuristic methods to address the Travelling Salesman Problem (TSP). The objective is to find the most efficient route for a delivery agent in a food delivery scenario using the Zomato delivery analytics dataset. The system integrates Dijkstra's algorithm with the Greedy Nearest Neighbor strategy and compares it with Ant Colony Optimization (ACO) and Genetic Algorithm (GA). A comparative analysis of route accuracy, performance time, computational complexity, and scalability is discussed. The proposed system demonstrates significant improvement in route optimization, particularly for dynamically changing environments with multiple constraints. Visualization of route paths and performance metrics confirms the benefits of integrating classical and nature-inspired algorithms in a hybrid framework.