SupplyShield 2.0 is an AI-powered logistics intelligence system designed to monitor global shipments in real time, detect risks proactively, and automate decision-making and communication to reduce costly disruptions. Supply chains today face unpredictable events — weather delays, port strikes, roadblocks, and geopolitical risks — yet most systems rely on manual reporting, slow updates, and siloed communication. These gaps lead to delayed actions, misinformed stakeholders, and massive financial losses. SupplyShield 2.0 solves this with a full-stack AI-driven dashboard that uses real-time inputs (shipment logs, weather, news) to identify potential disruptions. It uses Claude LLM via Anthropic to summarize risks, suggest optimal decisions (like rerouting or expediting), explain severity levels, and even draft professional messages or Slack updates for stakeholders — automatically. The system is built using Streamlit for interactive visualization, LangChain + LangGraph for smart logic flows, and ChromaDB to store historical decisions and learn from them. It integrates OpenWeatherMap, GNews, and Slack Webhooks, providing a 360° view of each shipment — including weather at its current location, breaking news alerts, AI-generated risk summaries, and cost comparisons between action choices (e.g., penalty vs air freight vs rerouting). Users can upload new shipment entries or choose from samples. The app allows for one-click PDF and CSV report exports and features an internal chatbot interface that remembers the current session. It is built with modular architecture, real-time visuals (Plotly maps and charts), memory-driven chat, and customizable alerts. By automating insight, communication, and contingency logic, SupplyShield 2.0 helps companies reduce operational losses, speed up decision-making, and improve reliability in their global supply operations. It combines the power of LLMs with logistics domain intelligence to build the next-generation smart supply chain layer.
Category tags:"Nice and original idea, I think it can also be applied for commodity traders to improve accuracy. The UI needs to be improved from auto-generated Streamlit app, but the feature ideas depicted there are great. Presentation is good, lot of clear examples of benefits."
MACIEJ MAZUR