License Plate Recognition System for Improved Logistics Delivery in a Supply Chain with Solution Validation through Digital Twin Modeling

Technological advancements aim to facilitate more optimal supply chain flows. Despite advancements, delays brought by traffic enforcement contact apprehensions disrupt the supply chain's efficiency, thus facilitating implied losses. Although companies cannot control the traffic, the government can optimize traffic flow and enforce traffic laws with minimal disruptions to supply chain delivery agents. A method of optimizing traffic flow while maintaining an uncompromised traffic law enforcement initiative is through an automated no-contact apprehension policy (NCAP). This enforces an AI-based system to detect traffic violations and link them to the license plate information of a motor vehicle. In these systems, an LPR module is essential. This study investigates using a neural network-based LPR architecture for traffic control and management. A MATLAB digital twin simulation is used to model a supply chain flow. The simulation compares and contrasts the supply chain flow in two cases: one where traffic enforcement is conducted in-person, while the other using NCAP. Simulation results suggest that more deliveries can be made when the government enforces an NCAP system.