Disrupting Supply Chain with Data Science
summary
Company builds a B2B product that crawls company websites to reverse‑engineer client–supplier links, using NLP transformers to detect customer pages and extract names, then assembles a global supplier graph; from 200 Tier‑1 inputs it finds several thousand Tier‑2/3 nodes. Risk assessment combines bankruptcy risk models, financials and newly computed ESG climate and labor scores; a similarity model ranks alternative suppliers from millions of firms so customers can proactively replace high‑risk suppliers. Only ~6% of firms have deep Tier‑2/3 visibility, so this method materially expands coverage.
AI-written summary · about
chapters (19)
- 0:00Supply chain visibility intro
- 1:10Supply chain disruption example
- 2:58Traditional visibility limits
- 3:40Three-step risk mitigation
- 5:50Supply chain networks bottom-up
- 9:48Car manufacturer use case
- 11:32Visualizing complex networks
- 12:34Risky suppliers assessment
- 14:24Risk scores for supply chain
- 16:49Tier 2 and 3 suppliers importance
- 17:12ML for new suppliers
- 19:39Proactive supplier replacement
- 20:00Three-step process summary
- 20:44Q&A risk assessment
- 21:48Online data for real-time risk
- 22:32Weather and climate risk data
- 22:56Q&A ESG metrics
- 23:17ESG metrics explained
- 24:45Q&A data complexity
related talks
28mCombining Third-Party Data, Entity Resolution, and Graph Analytics for ESG Risk Mgmt
27mLeveraging AI for Resilient Supply Chains
27mDisrupting Lead Generation through Artificial Intelligence
25mAI-Driven Supply Chain Security: Using AI-BOM to Detect Hidden Vulnerabilities
43mDomain-driven Data Science
31mKeynote Panel : AI in Supply chain
23mDemand Sensing Decision Support System
23mUsing Software, AI, DS to Source Growth Opportunities