Sprout to Flourish

By Magda Mojsiejuk & Odd Data & Design Studio

VIDEO

Concept & Prototype description

The Algorithmic Simulator for Designing Regenerative Farming Practices will focus on companion planting, offering a practical tool to optimize small to medium scale sustainable farming systems. This simulator will enable users to design planting layouts that enhance biodiversity, improve soil health, and support natural pest control by suggesting intelligent crop pairings based on ecological principles. It will generate dynamic layouts by incorporating data on plant relationships—such as nitrogen fixation, pest resistance, and growth patterns.
Grounded in regenerative agriculture and permaculture, this design tool will provide a user-friendly interface to plan farm layouts that maximize ecological benefits, helping farmers make informed, sustainable decisions within a streamlined development timeline.

Connection to the scenario
Future scenario: Patterns that Persist

The Algorithmic Permacultural Planning Simulator addresses key issues raised by the “Patterns that Persist” scenario, including complicated design practices, ideological polarization, and farmers feeling disenfranchised. The tool simplifies complex planting and harvesting decisions, enabling farmers to design biodiverse, climate-resilient systems, similar to the thriving food forests mentioned in the scenario. By combining traditional agroecological practices with advanced AI and environmental analysis, it supports the transition from extractive to regenerative agricultural models while promoting collaboration. The simulator aligns with the scenario’s vision of algorithmic planting and transforming bioregions, helping farmers adapt to changing climate conditions and work together towards sustainable agroecological practices.

Technology

The system is a closed-loop, algorithm-assisted design and planning framework that emphasizes adaptability and collaborative intelligence. It evolves by refining its understanding through human input and algorithmic analysis, and works with incomplete data. The system is powered by four key databases: a Macro Context Database (large-scale environmental data), a Micro Context Database (localized measurements), a Species Database (plant and organism information), and a Relationship Graph Database (species interactions). Using design guidelines and user inputs, it runs a simulation to optimize species selection, evaluate species relationships, and generate a spatial layout. The results are displayed as a layered design map, allowing users to adjust inputs and refine the design iteratively. The system operates on a Google Cloud instance with a web-based user interface.