Impacts of Urban Consolidation Centres for Sustainable City Logistics Using Adaptive Dynamic Programming Based Multi-Agent Simulation

Abstract

This paper aims at evaluating the impacts of Urban Consolidation Centers (UCC) for sustainable city logistics using Adaptive Dynamic Programming (ADP) based multi-agent simulation (MAS-ADP). Economic efficiency and environment friendliness criteria were used to evaluate the sustainability of UCC. The results proved that the implementation of UCC as a sustainable city logistics scheme is efficient in reducing 8% of the total delivery cost for freight carrier and reducing 36% of the total emissions released to the environment. It is also showed that the use of learning agents is essential to demonstrate the successful implementation of the UCC, as it is only in the learning-based simulation, UCC operator could get a profit. Our simulation analysis also confirmed that compared to widely used reinforced learning algorithms (Q-learning), MAS-ADP brought increased accuracy to the evaluations’ outcomes of UCC.