AI Replays Decisions to Improve Supply Chain Performance
summary
o9’s closed-loop AI reconstructs every planning decision, traces links across demand, procurement, production and execution, and quantifies impacts on KPIs like inventory and service level. The system layers four graphs (value chain, decision context, policy learning, solver graph), runs LLMs over them and uses AWS components such as Sagemaker and Lambda. It reports concrete findings—44% of excess inventory cases trace to forecast adjustments, a 4% intersection set explains most overstock—and recommends prioritized actions to drive value leakage stories and achieve up to 90% touchless planning with Post Game Analytics.
AI-written summary · about
chapters (11)
- 0:00Introduction to AI in supply chain
- 1:49Need for decision replay systems
- 3:35Problem of disconnected supply chain insights
- 4:41Limitations of LLMs in supply chain
- 5:46Graph technology layers explained
- 7:10Tracing decisions and their impact
- 8:58Post-game analysis on excess inventory
- 10:25Consolidating root causes over cycles
- 11:50Corrective actions and policy changes
- 13:35Post-game analyzer agents benefits
- 14:41Consistency and touchless planning
related talks
27mLeveraging AI for Resilient Supply Chains
26mRiddles in Supply Chain Management - How AI solves them?
23mAI Here & Now
16mUsing AI to solve complex business problems: optimizing...
10mAI in Procurement: Enhancing Supply Chain Resilience
29mAI & ML in tailored software for decision-making purposes
26mAI-Powered Predictive Analytics: Optimize Performance
32mGlobal Artificial Intelligence Virtual Conf - Webinar Mar 2022