AI-powered platform for ice cream formulation and sensory prediction
This AI-driven platform enhances ice cream development by predicting sensory outcomes using a Random Forest model, allowing formulators to optimize ingredients and processing parameters. It features a scalable, secure cloud deployment and interactive dashboard.
Overview
The AI-powered platform for ice cream formulation and sensory prediction offers an innovative approach to optimizing ice cream development. Utilizing a full-stack AI-driven solution, formulators gain an interactive dashboard to select ingredients and adjust key processing parameters, such as freezing temperature and mixing speed. This platform empowers users to predict sensory outcomes like creaminess and sweetness, facilitating rapid virtual prototyping and innovation while reducing development risks.
Technical specifications
Key features:
- Interactive Dashboard: Built with React, allowing dynamic adjustments to ingredients and processing parameters.
- Machine Learning Model: A Random Forest model predicts sensory outcomes based on ingredient data, achieving a Mean Squared Error of ~0.35.
- API and Cloud Deployment: A robust Python API using FastAPI, securely deployed on AWS S3, ensures scalability and security.
- Adaptability: The model can be retrained with proprietary data for enhanced accuracy and customized predictions.
Technology readiness level
The platform is at Technology Readiness Level 6, indicating a fully functional MVP with a modern, scalable, and secure architecture. Future phases include data integration, feature enhancement, and production hardening, aiming to refine and implement the solution in a production environment.
About Teo AgriTech
Teo AgriTech is an agricultural technology company that provides precision insights for the fruit industry, focusing on yield and quality estimation. The company utilizes automated drone technology to scan individual trees, which allows for detailed data collection and analysis. By integrating artificial intelligence, their system generates actionable reports that aim to improve estimation accuracy to approximately 90%, offering a significant improvement over traditional manual observation methods. The technology is designed to be adaptable to different orchard sizes and various types of fruit, with an initial focus on citrus.
This solution addresses challenges related to inaccurate yield forecasting, which can lead to economic losses for farmers and buyers during price negotiations. By providing reliable yield data and detecting issues such as overproduction early, Teo AgriTech helps farmers manage their fruit loads, optimize product quality, and negotiate fair market prices before harvest. The company targets a broad range of stakeholders in the fruit sector, including farmers, buyers, and exporters, with the goal of increasing operational efficiency and maximizing profitability. Implementation of their system is designed to be efficient, typically requiring a few weeks to become operational.
delivered to your inbox.