Skip to content

replit.md

Latest commit

 

History

History
85 lines (65 loc) · 3.86 KB

replit.md

File metadata and controls

85 lines (65 loc) · 3.86 KB

EDUST Smart Waste Management System

Overview

EDUST is a comprehensive IoT-enabled waste management solution designed for municipalities. The system provides real-time monitoring of waste bin fill levels, AI-powered predictions for waste collection optimization, and route planning capabilities. The application features both administrative dashboards for management oversight and mobile interfaces optimized for field workers and drivers.

User Preferences

Preferred communication style: Simple, everyday language.

System Architecture

Frontend Architecture

  • React + TypeScript: Component-based UI with type safety for better maintainability
  • Vite: Modern build tool for fast development and optimized production builds
  • shadcn/ui: Consistent design system built on Radix UI primitives and Tailwind CSS
  • TanStack Query: Robust data fetching, caching, and synchronization
  • Wouter: Lightweight client-side routing
  • WebSocket Integration: Real-time updates for live bin status monitoring

Backend Architecture

  • Node.js + Express: RESTful API server with middleware support
  • In-Memory Storage: Current implementation uses memory-based data storage with full CRUD operations
  • WebSocket Server: Enables real-time communication between server and connected clients
  • Modular Services: Separated concerns for IoT integration and AI service communication

Data Layer

  • Drizzle ORM: Type-safe database toolkit configured for PostgreSQL
  • Schema Design: Structured entities for users, bins, drivers, routes, alerts, and activities
  • Migration Support: Database versioning through Drizzle Kit

IoT Integration

  • Blynk Cloud API: Automated data synchronization every 2 minutes for bin fill levels
  • Real-time Monitoring: Continuous tracking of bin status with automated alert generation
  • GPS Tracking: Location-based services for bin positioning and driver tracking

AI Service Layer

  • Python + Flask: Microservice architecture for machine learning operations
  • Prediction Models: Linear regression for waste fill time predictions
  • Route Optimization: Traveling salesman problem solving using nearest-neighbor algorithm
  • Anomaly Detection: Pattern recognition for unusual waste generation patterns

Authentication & Security

  • Session Management: Express sessions with PostgreSQL store
  • Role-based Access: Admin and driver user roles with appropriate permissions

External Dependencies

Core Frameworks

  • React 18: Frontend UI framework with hooks and modern patterns
  • Express.js: Web application framework for Node.js
  • Flask: Python web framework for AI service microservice

Database & ORM

  • PostgreSQL: Primary database (configured via Drizzle but may need setup)
  • Neon Database: Serverless PostgreSQL provider integration
  • Drizzle ORM: Type-safe database operations and migrations

IoT Platform

  • Blynk Cloud: IoT platform for hardware device communication and data collection

UI Components & Styling

  • Radix UI: Headless component library for accessibility
  • Tailwind CSS: Utility-first CSS framework
  • Recharts: React charting library for data visualization
  • Lucide React: Icon library

Data Management

  • TanStack Query: Server state management and caching
  • React Hook Form: Form handling with validation
  • Zod: Schema validation for type safety

Real-time Communication

  • WebSocket (ws): Real-time bidirectional communication
  • Socket connections: Live updates for bin status and alerts

AI/ML Libraries

  • scikit-learn: Machine learning algorithms for predictions
  • NumPy: Numerical computing for data processing
  • Pandas: Data manipulation and analysis

Development Tools

  • TypeScript: Static type checking
  • ESBuild: Fast JavaScript bundler
  • Vite: Development server and build tool