Developed during the TUM.ai Makeathon 2023 for the Genistat challenge by team Too Good To Code.
The future of energy is green! To accelerate the clean energy revolution, we must use the resources we have in the most effective manner. Last year, 50% more solar panels were installed on German roofs than the year before. While this is great news, it does not paint the full picture. Due to supply chain limitations and skill shortage, most home owners in Germany have to wait up to 1 year for a consultation appointment with solar roof experts. For an installation, the waiting periods are even crazier: Waiting times of 2-3 years before a solar roof can be installed have become the new normal. The current solar market does not scale with the demand!
That's why we must ensure that the limited solar panel resources we have are used in the most efficient way possible! Sadly, the current system is quite the opposite. Solar construction companies serve their customers on a first-come-first-serve basis: Whoever requests the solar panel first gets the panel installed first. This inherently means that solar installations on large roofs at locations with optimal sun conditions are often postponed in favor of installation locations that are less well-suited, simply due to the fact that one request came in before another.
- Go to too-good-to-code.streamlit.app
- Insert an address in Berlin, Hamburg, or Bremen
- Select the budget using the slider
- Pin point the exact building (optional)
- Our model will predict the needed amount of solar panels for the residence
- Play around with our visualizations and see how we predicted the roof type, sun radiation, etc.
The AI model and calculations can be found in the roof_classification.zip file.
- Alexander Deutschbauer
- Hari Kesavan
- Tobias Morocutti
- Nathanya Queby