AuroraSense:– AI Inventory Health Dashboard

AuroraSense is a real-time inventory monitoring and forecasting dashboard built using Snowflake Streamlit, providing instant visibility into stock levels, health status, consumption trends, and risk alerts across multiple locations.

The dashboard identifies critical shortages, predicts stockout dates, and provides actionable insights to support fast and informed supply-chain decisions.

---

Live Working Prototype (Snowflake Streamlit)

Click below to open the hosted working model:

https://app.snowflake.com/nnvfrzc/vl98936/#/streamlit-apps/AURORA_INVENTORY.PUBLIC.HS83MDTAYG8W3429

---

Key Features

1. Real-Time Inventory Health

• Highlights Critical, Warning, and Healthy items
• KPI cards summarizing global stock status
• Automatic risk categorization using Days of Cover

2. High-Risk Alerts

• Dedicated alert section for items nearing stockout
• Filterable by location and item

3. Forecasted Stockout Dates

• Predicts estimated stockout based on:

  • Days of cover
  • Daily consumption
  • Current movement trends

4. Multi-Layered Visual Analytics

• Heatmap for stock availability across locations
• Bar charts for comparing Days of Cover
• Trend line charts for historical stock movement

5. Interactive Filters

• Filter by location, item, and chart type
• Dashboard auto-updates dynamically

---

Tech Stack

Backend & Data Layer

• Snowflake Data Cloud

• Snowflake Tables:

  • DAILY_STOCK
  • STOCK_HEALTH
  • STOCK_ALERTS

Frontend / App Layer

• Streamlit for Snowflake
• Python (Pandas, Altair, datetime)
• Snowpark for data access
• Altair for interactive charts

Screenshots

How to Use

1. Open the Snowflake app using the link above

2. Select a location and item from the sidebar

3. Navigate through:

   • Stock health KPIs
   • Top-risk items
   • Stockout forecast table
   • Alerts
   • Charts

4. Click Generate Summary for AI insights

🛠️ Setup (If Running Locally)

pip install streamlit pandas altair snowflake-snowpark-python
streamlit run app.py

Snowflake credentials required for live data.