Sarcouncil Journal of Engineering and Computer Sciences

Abstract: The increasing complexity of modern organizational structures and distributed workforces has created significant challenges in meeting scheduling systems. Traditional centralized scheduling approaches face limitations in scalability, privacy preservation, and adaptation capabilities. The Adaptive Multi-Agent Meeting Scheduling framework leverages Federated Reinforcement Learning to enable decentralized and privacy-preserving optimization. By combining distributed agents with federated learning capabilities, the system maintains scheduling efficiency while protecting individual data privacy. The results demonstrate marked improvements in conflict resolution, resource utilization, and scheduling optimization across large-scale organizational deployments.