Designing an AI-Driven Intelligent Vehicle Routing Model for Real-Time Order Picking and Delivery Operations in Food Service Logistics

DOI: https://doie.org/10.10399/JBSE.2025363880

Gourab Dutta, Debabrata Sarddar, Rahul Kumar Ghosh

Keywords:

AI-driven logistics; Vehicle routing optimization; Order picking; Food service delivery; Machine learning

Abstract:

With the advent of modern food service logistics, intelligent and dynamic routing systems have become necessary to efficiently tackle complex order picking and last-mile delivery scenarios. The design and implementation of a smart AI-enabled vehicle routing model with real-time data integration, IoT connectivity, and communication networks are proposed to maximize delivery efficiency in food service logistics. According to the model, it uses historical order data, weather data, and time features to predict demand patterns using advanced regression methods while duly incorporating the predictions into a vehicle routing heuristic to fulfil orders dynamically. Exploratory data analysis and feature engineering were applied, architecture was proposed, and several machine learning models were checked for prediction, including linear regression, ridge regression, and support vector regression. Although drawbacks are observed with predictive R² scores pointing to model limitations, the proposed integrated framework has shown useful applications in aiding routing decisions through real-time update communications. The study gives a pragmatic perspective on deploying AI- and IoT-enabled communication systems for the optimization of logistics operations for rapidly evolving urban food-service environments.

PDF