Extending Incentive-Based Port Logistics Optimization Through AI-Driven Multimodal Integration

Abstract

This paper proposes an AI-based extension of an incentive-compatible logistics coordination model originally developed for port operations. Building upon the smart contract–driven framework introduced by Hu and Zuo, the study integrates real-time adaptive mechanisms across maritime, port, and urban transport systems. The conceptual architecture incorporates decentralized AI modules to enhance routing, traffic signal control, and operational responsiveness under dynamic conditions. To evaluate the proposed model, a simulation was conducted using MATSim, comparing fixed-time and adaptive signal scenarios within a high-density urban network. The adaptive system employed a decentralized signal controller capable of adjusting traffic light phases in response to local demand. Results showed a significant reduction in average travel time and delay under adaptive control, supporting the effectiveness of real-time coordination in logistics networks. While the simulation outcomes are based on ideal conditions, the underlying logic and comparative insights remain valid and practically relevant. The findings suggest that AI-based control strategies can complement incentive mechanisms by enhancing short-term adaptability and system resilience. This hybrid approach represents a promising direction for managing complexity in multi-modal, port-connected urban logistics chains.