Challenge background
Singapore generates over 800,000 tons of food waste each year, a pressing issue for a nation committed to sustainability. A significant portion of this waste comes from supermarkets, restaurants, and events, often comprising surplus food that remains safe and consumable. While this surplus holds immense potential to benefit diverse groups, such as residents, migrant workers, and visitors, much of it goes unused due to the absence of an efficient redistribution system. Addressing this challenge offers not only an opportunity to minimize food waste but also to create a positive social impact by reducing financial burdens, enhancing food security, and fostering a more inclusive community.
The problem
Singapore has a wealth of surplus food that, if efficiently redistributed, can significantly benefit individuals and communities, reduce waste, and promote sustainability. However, the current absence of a centralized solution to connect businesses with surplus food to charities, food banks, and people in need presents a missed opportunity. An effective platform is essential to bridge this gap, enabling surplus food to reach those who need it most and maximizing its social, economic, and environmental value.
Goal of the project
Develop a dashboard platform that:
- Visualizes historic data on food surplus across Singapore.
- Utilizes AI to predict surplus from supermarkets, restaurants, and events.
- Connects businesses, charities, and food banks to redistribute surplus effectively.
Project timeline
- 1
Week 1
Week 1: Project Setup and Data Collection
- Define scope, user personas (businesses, charities, food banks, etc.), and functional requirements.
- Identify and source datasets (e.g., historical food waste data, business inventory data, food bank needs, event data).
- Gather publicly available datasets or establish partnerships with supermarkets, restaurants, and event organizers.
- 2
Week 2
Week 2: Data Cleaning and Preparation
- Clean and preprocess data to remove duplicates, handle missing values, and standardize formats.
- Perform exploratory data analysis (EDA) to understand trends, anomalies, and correlations.
- 3
Week 3
Week 3: Dashboard Wireframe and Historical Data Visualization
- Design wireframes or mockups for the dashboard (e.g., historical surplus data view).
- Use Tableau or a similar tool to create a prototype with historical data visualizations.
- 4
Week 4
Week 4: Build Machine Learning Models
- Develop machine learning models to predict food surplus based on patterns in historical data.
- Train and validate models using a portion of the collected datasets.
- 5
Week 5
Week 5: API Development
- Build an API to fetch real-time predictions from the ML model and connect it to the dashboard.
- Ensure the API can retrieve both historical and predicted data.
- 6
Week 6
Week 6: Dashboard Integration and AI View
- Integrate the predictive AI view into the dashboard.
- Implement features for real-time updates and interactive filtering (e.g., location, type of surplus).
- 7
Week 7
Week 7: Testing and Refinement
- Conduct end-to-end testing of the platform (data pipeline, ML predictions, and dashboard functionality).
- Fix bugs, optimize performance, and improve usability based on feedback.
- 8
Week 8
Week 8: Deployment and Launch
- Deploy the dashboard for public use (host on Tableau Public or a web platform).
- Promote the platform through outreach to businesses, charities, and communities.
- Gather initial feedback from users for further improvement.
What you'll learn
- Master data collection, cleaning, and analysis.
- Build predictive ML models.
- Develop APIs to integrate AI and historical data into a live public dashboard.