Construction_Dashboard_ai

Inspiration

Construction projects have lots of data, but teams still spend a lot of time checking spreadsheets, emails, and making calls just to know what's happening. We wanted to build something that brings everything into one place and helps teams spot problems before they cause delays.

What it does

Construction Dashboard AI is a dashboard that tracks procurement, materials, and equipment in one place. It lets users: Track material requests from request to delivery. Monitor materials and equipment on site. Predict possible delivery delays. Highlight items that need attention. Log requests and updates using an AI chat assistant instead of long forms. The goal is to help project managers know what needs attention without searching through multiple systems.

How we built it

We built the project around two AI features. The first is an AI assistant that helps workers submit material requests and update statuses through chat. It asks follow-up questions and uses RAG with project data like BOQs and vendor information to give accurate responses. The second is a delay prediction model that estimates delivery dates and delay risk using past vendor performance and material lead times. All of this is shown in a single dashboard with procurement tracking, equipment status, an activity log, and a "Needs Attention" section.

Challenges we ran into

Our biggest challenge was keeping the project focused. There are many ideas like IoT sensors and shipment tracking, but we wanted to build something useful within the hackathon. Making the AI reliable was another challenge, so we stored conversation data separately instead of relying on the model's memory.

Accomplishments that we're proud of

Built one dashboard for procurement and equipment tracking. Created an AI assistant for material requests. Added RAG to make responses project-specific. Built a delay prediction feature. Designed a simple dashboard that highlights the most important issues.

What we learned

We learned that AI works best when it's solving a clear problem. We also learned how to combine RAG, prediction models, and dashboard design into one workflow.

What's next for Construction_Dashboard_ai

Next, we'd like to: Connect with ERP systems. Add live shipment tracking. Improve delay predictions with more data. Build a mobile version for site teams. Add notifications and approval workflows. Support multiple construction projects in one dashboard.

Built With

• chromadb
• docker
• fastapi
• figma
• github
• langchain
• node.js
• ollama
• postgresql
• postman
• python
• rag
• react
• recharts-backend:-fastapi
• scikit-learn-tools:-git
• sqlalchemy-ai/ml:-ollama
• tailwind-css
• typescript
• vectordb