Skip to content

meso03/Hoo_Hacks

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

15 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

WasteWatch: Predicting and Reducing Food Waste

Inspiration

Food waste is a massive problem - $254 billion is lost annually across restaurants, grocery stores, and schools. The largest portion, $62 billion, comes from restaurants, while many people still struggle with food insecurity. Restaurants alone generate between 22 to 33 billion pounds of food waste each year. In fact, 4–10% of food purchased by restaurants is wasted before it ever reaches the customer.

We decided to change that. We created a web application powered by machine learning to help restaurants predict and reduce food waste. By leveraging technology, we provide custom solutions that help restaurants save money and increase profits.

What Does Our Application Do?

WasteWatch is a predictive tool that takes into account:

  • Location and environmental factors
  • Food type and quantity
  • Number of guests
  • Storage conditions
  • Historical sales data

Our interactive dashboard allows restaurants to view the generated analysis and receive solution recommendations tailored to their unique needs.

Key Features

  • Predictive Analytics: Machine learning model with 78% accuracy
  • Interactive Dashboard: Real-time waste metrics and visualization
  • Actionable Recommendations: Custom waste reduction strategies
  • Environmental Impact Tracking: CO2 emissions and water usage metrics

Financial Analytics

  • ROI Projections: Data-driven 5-year financial forecasting
  • Cost Breakdown Analytics: Statistical distribution of waste-related expenses
  • Savings Projection Models: Comparative analysis of current vs. optimized costs payback periods

Technical Architecture

  • Machine Learning Core: Random Forest Regressor trained on 7,500+ restaurant waste records
  • Feature Engineering Pipeline:
    • Shelf life data extraction and normalization
    • Food category standardization with custom mapping algorithm
    • Statistical derivation of utilization metrics
  • Model Performance: Mean Absolute Error of 0.04
  • Cross-validation: K-fold validation to ensure model robustness

How We Built It (Data Science)

We built a predictive food waste management platform by combining machine learning, data integration, and interactive visualizations. Using over 7,500 restaurant waste records, regional statistics, shelf-life data, and real-time inputs, we engineered features like utilization rate and waste-per-guest ratio.

After cleaning and preprocessing the data, we trained a Random Forest Regressor achieving 78% prediction accuracy. We optimized the models, connected them to real-time data feeds, and built a user-friendly dashboard.

Challenges We Ran Into

  • Data Quality: Handling messy data cleaning and organizing datasets with missing or inconsistent values
  • Seasonal Fluctuations: Designing an adaptive model to handle unpredictable demand patterns
  • User Experience: Creating a powerful yet easy-to-use tool for non-technical users

Accomplishments We're Proud Of

  • Building a user-friendly interface that enables non-technical users to make data-driven decisions
  • Developing a system that integrates with real-time data for continuous optimization
  • Creating a solution with measurable environmental and financial impact

What We Learned

  • Data Quality Matters: The accuracy of the model heavily depends on the variety and reliability of data sources
  • Simplicity is Key: Clear, actionable insights are just as important as accurate predictions
  • Technology Can Drive Sustainability: Small optimizations in inventory management can lead to major reductions in food waste

What's Next for WasteWatch?

  • Smart Alerts: Notifications when food is nearing expiration
  • Grocery Store Partnerships: Integrations with suppliers to adjust inventory in real time
  • AI-Powered Learning: Models that improve continuously based on restaurant feedback and usage patterns

Getting Started

  1. Clone the repository:

    git clone https://github.com/yourusername/WasteWatch.git
    cd WasteWatch
  2. Set up the backend:

    cd backend
    pip install -r requirements.txt
    uvicorn main:app --reload
  3. Set up the frontend:

    cd frontend
    npm install
    npm run dev
  4. Access the application:

Impact Potential

  • Data-Driven Savings: Statistical model projects average ROI of 280% in the first year
  • Environmental Impact: Quantitative analysis shows potential to reduce CO2 emissions by up to 2.5kg per kg of food saved
  • Operational Efficiency: Data analysis indicates payback period of approximately 3 months for implementation costs

Final Thoughts

WasteWatch isn't just about predicting food waste - it's about empowering restaurants to make smarter, more sustainable choices. With machine learning and real-time data, we're transforming food waste into optimized inventory management, lower costs, and a greener planet!

Less waste. More savings. A smarter future.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 95.6%
  • Cython 2.8%
  • C 0.7%
  • C++ 0.7%
  • TypeScript 0.2%
  • JavaScript 0.0%