Investigating the Potential of Causal Reinforcement Learning in Collaborative Urban Logistics: A Systematic Literature Review

Causal reinforcement learning (causal RL) represents an innovative approach that combines the principles of causality with reinforcement learning (RL), facilitating more effective and interpretable decision-making in dynamic environments. This systematic literature review investigates the potential application of causal RL in the domain of collaborative urban logistics, a critical component of modern urban infrastructure involving the efficient transportation and delivery of goods within cities. The study aims to identify current applications, methodologies, and frameworks of RL in urban logistics and evaluate its impact on collaboration among logistics agents. By synthesizing existing research, the review highlights key findings, trends, and proposes future research directions to advance the integration of causal RL in urban logistics. This study is essential for understanding the importance of causal RL in solving the complexity of urban logistics challenges by enhancing the explainability for better solutions quality.