ABSTRACT
Chapter 5 presents multi-agent simulation with machine learning, which allows multiple stakeholders to understand the behaviour and interaction among them. It encompasses reinforcement learning including Q-learning and adaptive dynamic programming (ADP) for supporting multi-agent simulation. Application of these models in urban delivery using urban consolidation centre (UCC) is discussed. It also highlights a decision support system using multi-agent simulation including multi-actor multi-criteria analysis (MAMCA).
