Supply chain execution is undergoing fundamental transformation. As complexity accelerates and disruptions become the norm, reactive rule-based systems can no longer keep pace. In this SupplyChainBrain webinar, leaders from EPG and NVIDIA examine how agentic artificial intelligence is transforming logistics from insight-driven visibility to autonomous, intelligent execution. From a supply chain perspective, EPG (Ehrhardt Partner Group) competes in the European WMS market against SAP EWM, Manhattan, Blue Yonder, and Körber with its LFS WMS, Greenstone WCS (Warehouse Control System), and Lydia Voice voice-based picking solutions. NVIDIA's Omniverse, Isaac Sim, cuOpt, and Metropolis platforms form the infrastructure layer for supply chain and logistics AI.
From a supply chain perspective, we can understand how the transition from reactive logistics to autonomous supply chains is developing and what this means for supply chain leaders planning the next era of performance and resilience. Agentic AI represents the next evolution of generative AI, incorporating capabilities for tool use, multi-step reasoning, autonomous decision-making, and self-correction. Anthropic's Claude, OpenAI's o1 and o3, Google DeepMind's Gemini, and Meta's Llama series form the foundation model layer. Orchestration frameworks such as LangChain, CrewAI, AutoGen, and LangGraph enable multiple agents to work collaboratively across logistics workflows.
From a supply chain perspective, specific logistics applications of agentic AI focus on areas such as dynamic routing for last-mile delivery via route optimization; automated exception resolution for warehouse exception management; multi-DC inventory balancing for inventory rebalancing; spot market negotiations for carrier procurement; and alternative supplier discovery for supply risk monitoring. EPG's Greenstone platform is positioned to deliver agentic decision mechanisms between the automation hardware layer and the WMS. The orchestration layer for conveyor, sorter, AS/RS (Automated Storage and Retrieval System), AMR (Autonomous Mobile Robot), and AGV (Automated Guided Vehicle) systems becomes the natural deployment ground for agentic AI.
From a supply chain perspective, NVIDIA's role in this segment extends beyond GPU hardware to encompassing capabilities delivered to logistics leaders through its CUDA software ecosystem: cuOpt for VRP (Vehicle Routing Problem), Isaac Sim for robot training simulation, and Omniverse for digital twin creation. The NVIDIA H100, H200, B200, and GH200 Grace Hopper superchips serve as foundational hardware for foundation model training and inference. Edge AI through the NVIDIA Jetson Orin platform provides local decision mechanisms for warehouse robots, autonomous forklifts, and AMRs. In essence, the vision reflected in the joint webinar between EPG and NVIDIA is concrete evidence that supply chain execution will be fundamentally reshaped by agentic AI in the years ahead.
Key Takeaways:
1. The webinar brings together leaders from EPG and NVIDIA.
2. Reactive rule-based systems fall short in the face of complexity.
3. Agentic AI transforms logistics from visibility to autonomous execution.
4. NVIDIA cuOpt accelerates routing optimization with VRP.
5. EPG Greenstone serves as the automation hardware orchestration layer.
