Cognitive Approaches for Dynamic Allocation of Vehicles in Real-Time Logistics
Year: 2026
Type: conference-paper
Abstract: From static, predictable conditions to dynamic, realtime environments shaped by data, traffic, and customer demand fluctuations. Traditional static algorithms struggle to address this complexity, resulting in computational delays and the constant need for real-time updates. To overcome this, we propose integrating Deep Reinforcement Learning (DRL) with these static algorithms, allowing the system to proactively predict and adapt to changing transport scenarios. By treating logistics as an environment and the optimization system as an intelligent agent, our approach anticipates challenges by utilizing agent rewards tied to real-time constraints, such as traffic and time, thereby enabling alternative solutions before issues arise.
Language: English
Subfield: Industrial and Manufacturing Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W7203631329