Akademik Veri Yönetim Sistemi
18th International Conference on Agents and Artificial Intelligence, ICAART 2026, Marbella, İspanya, 5 - 08 Mart 2026, cilt.2, ss.1009-1016, (Tam Metin Bildiri)
Automated sortation has emerged as a major trend in logistics, enabling scalable and time-efficient sorting of items through the deployment of robotic agents. This study examines the use of capacity-enhanced agents in automated sorting domain, by evaluating task assignment strategies that improve the Token Passing with Multiple Capacity (TPMC) algorithm for the Multi-Agent Pickup and Delivery with Capacities (MAPDC) problem. In a simulated sorting domain, eight methods leveraging agent waypoint information are evaluated against the commonly used nearest-pickup task selection method. The results show that task assignment heuristics incorporating Closeness Centrality, Hausdorff Distance, and cost-based measures significantly improve solution quality of path planning in sortation systems. Moreover, the gains are greater in the sortation domain with disjoint pickup and delivery areas than in automated warehouse settings without spatial constraints, highlighting the importance of environment structure in MAPDC.