AI StrikeSight: Wildlife Risk Intelligence for Aviation

• 

  Wildlife Strike

✈️ Inspiration
Wildlife strikes are a growing threat to aviation safety, often resulting in costly damage, operational delays, and even loss of life. The inspiration for this project began with a deeply hurtful moment: the Ahmedabad plane crash. Initially suspected to be caused by a bird strike — a theory later ruled out — it left me wondering, how could a bird bring down such a massive aircraft? That question triggered a deeper investigation into aviation strike data, species behavior, and risk patterns.

As I explored further, I realized how fragmented and reactive current systems are. Airport managers lack real-time intelligence, and wildlife specialists often work in silos. This disconnect can delay response and prevention efforts. Inspired by real-world aviation data and the need for proactive risk mitigation, we set out to build a solution that empowers both airport managers and wildlife specialists to collaborate using intelligent, real-time insights.

Our goal: bridge the gap between ecological expertise and operational decision-making — turning raw incident data into actionable intelligence that saves lives, protects aircraft, and strengthens aviation safety.

🤖 What it does
AI StrikeSight** is a Tableau-powered platform that predicts, analyzes, and helps prevent wildlife strikes. It features:

• Reusable, drag-and-drop dashboards
  
• Chatbot-powered analytics for quick queries
  
• Semantic modeling for intuitive data exploration
  
• Einstein Model Builder integration to predict aircraft damage probability
  
• Dual persona views for airport managers and wildlife specialists
  
• Slack integration for team collaboration
  
• Tableau Next interoperability for seamless cloud and desktop access

🛠️ How we built it
Data Prep: Cleaned and structured wildlife strike data (location, species, aircraft mass, time, damage level)
-Semantic Layer: Mapped technical fields to business-friendly terms, use AI to write calculations -Predictive Modeling: Built a classification model using Einstein Model Builder:
-Dashboard Design: Created persona-based views with filters, metrics, and drilldowns
-Integration: Embedded chatbot, Slack workflows, and Tableau Pulse alerts

⚠️ Challenges we ran into

• Sparse or inconsistent species data required enrichment
  
• Designing dashboards that served both operational and ecological roles
  
• Embedding real-time collaboration tools without compromising performance
• A lot of data with unknown species, locations and airports

🏆 Accomplishments that we're proud of

• Built a fully functional AI-powered dashboard with predictive capabilities
  
• Agentforce Chatbot is here to provide fully interactive, real-time answers to your questions
• Enabled seamless collaboration between airport and wildlife teams
  
• Created a semantic model that makes complex data accessible
  
• Delivered a scalable, reusable solution with Tableau Next interoperability

📚 What we learned

• How to integrate semantic layers and chatbot analytics for intuitive exploration
  
• How to integrate Slack with Tableau Next
• How to build predictive models using Einstein Model Builder
  
• How to design persona-driven dashboards that support real-world decision-making
  
• How to align ecological and operational data for cross-functional collaboration

🚀 What's next for AI StrikeSight: Wildlife Risk Intelligence for Aviation

• Integrate geospatial alerts for runway-specific risk zones
  
• Deploy predictive insights into live airport operations
  
• Collaborate with aviation authorities to scale across regions
  
• Enhance chatbot capabilities with natural language understanding and proactive alerts

Built With

• agentic
• chatbot
• data
• einstein
• prediction
• tableau
• tableaupulse