Optimizing Delivery Routes in Postal Logistics Using Mobile Apps and AI

This article presents an analysis and systematization of contemporary methods for optimizing delivery routes based on the use of mobile applications and AI-based solutions. The aim of the study is to develop a conceptual hybrid model that combines predictive analytics of machine learning with dynamic real-time route adjustment, relying on data obtained from couriers’ mobile devices. The methodology encompasses a comprehensive systematic review of scientific publications dedicated to solving the vehicle routing problem (VRP), as well as a comparative analysis of the performance of various algorithmic approaches in the field of AI. As a result, the hybrid adaptive model for route optimization (HAMRO) is described, in which genetic algorithms are employed for the initial generation of an optimal plan, and reinforcement learning methods ensure operational route adaptation under changing road and logistical conditions. The results presented in this article will be of interest to researchers in the field of transport logistics, developers of software for delivery services, and managers of postal and courier companies focused on enhancing the efficiency of operational processes.