Inspiration
Urban areas face rising temperatures, poor air quality, and inefficient building surfaces that contribute to climate stress. I wanted to explore how structured environmental data could be used to turn ordinary buildings into climate-responsive green systems without requiring complex AI models.
What it does
UrbanBloom is a rule-based decision engine that analyzes building inputs and recommends optimal green infrastructure solutions. It matches buildings with suitable vegetation systems, plant species, and installation methods, then calculates compatibility scores and environmental impact estimates such as cooling potential, air quality improvement, and stormwater retention.
How I built it
I built UrbanBloom using a structured knowledge-base approach with JSON datasets for plants, buildings, materials, installation systems, and decision rules. A JavaScript engine processes user inputs, applies rule-based scoring logic, filters compatible options, and generates ranked recommendations with confidence scores and explanations. The frontend displays results through a responsive dashboard interface.
Challenges I ran into
One of the main challenges was designing a system that behaves like intelligent AI without relying on machine learning models. I had to carefully structure the data relationships between plants, materials, and environmental constraints, and ensure the rule engine remained consistent, scalable, and explainable.
Accomplishments that I am proud of
I successfully built a fully functional decision system that simulates AI-like reasoning using only structured data and deterministic logic. The system dynamically generates recommendations, explains its reasoning, and adapts results based on environmental and structural constraints.
What I learned
I learned how to structure real-world environmental knowledge into JSON datasets, including plant traits such as drought tolerance, growth rate, sunlight needs, and pollution resistance, and how these factors influence suitability in urban environments. I gained a deeper understanding of urban greening concepts by studying how different plants interact with building materials, climate conditions, and installation systems like green walls and green roofs. On the technical side, I built a rule-based decision engine in JavaScript that processes this data, applies scoring logic, and generates ranked recommendations. I also connected HTML user inputs directly to this logic to create a fully interactive system where user actions drive the outputs. Overall, I combined environmental science and programming by translating ecological principles into a structured, functional decision-making system.
What's next for UrbanBloom
Next, I plan to integrate real environmental APIs, improve visualization with city-scale simulation tools, and potentially add a hybrid AI layer using machine learning or LLMs to enhance recommendations. I also aim to expand the plant and system database for broader global applicability.
Log in or sign up for Devpost to join the conversation.