"Leveraging Microsoft Azure and the AI capabilities of Copilot in the development of AGRIST Ai has enabled high-speed, high-accuracy analysis of large-scale agricultural data, significantly enhancing the precision of yield forecasting. In addition, Azure Machine Learning has allowed us to automate the entire pipeline from model training to deployment, ensuring consistently high performance. As agriculture holds immense potential for digital transformation, we will continue to harness cutting-edge technologies to realize sustainable and efficient farming."  - Hideki Shimizu, Vice President of Engineering, AGRIST 

Co-Innovation Challenge

AGRIST, a Japanese agritech startup, set out to solve a critical problem facing greenhouse farmers: unpredictable crop yields and market prices. With the average age of Japanese farmers around 67 and a shrinking workforce, labor shortages and uncertainty in production and sales threatened the stability and profitability of farms.

AGRIST’s innovative solutions, including AI-powered harvest robots and smart farming packages, were already helping mitigate some of the ongoing challenges in farming, and they saw another opportunity to add an intelligent solution to yield prediction and market conditions. Traditional methods left farmers exposed to weather and market fluctuations, making it difficult to plan production or optimize sales strategies and stabilize earnings.

To achieve a working solution, AGRIST partnered with the Microsoft AI Co-Innovation Lab in Kobe, taking advantage of the lab’s expertise in AI and close relationship to Microsoft technology and engineering support.

In the Lab

Working closely with Microsoft AI Co-Innovation Lab engineers, AGRIST developed “AGRIST Ai,” an advanced forecasting model using Azure Machine Learning Studio. The solution organized and analyzed over 50,000 data points from environmental sensors and harvest records to deliver highly accurate yield predictions.

Additionally, a machine learning operations (ML Ops) system updates, trains, and streamlined the models so that prediction and forecasting is agile and relevant. With the Microsoft AI Co-Innovation Lab team’s support, AGRIST has a system that allows models to be retrained and updated, creating a stable and sustainable environment where AI can support agriculture.

arch

Solution Impact

The deployment of AGRIST Ai in February 2025 has delivered measurable business results:

  • Revenue is increasing by more than 28.6% per harvest through optimized production and sales.
  • Accurate yield forecasting  is enabling better management decisions and reduced.
  • Inexperienced farmers are achieving 1.5 times the yield of experienced farmers, demonstrating that AI-driven insights can lower barriers to entry and expand opportunities in agriculture.
arch

AGRIST’s partnership with the AI Co-Innovation Lab not only improved technical capabilities but also delivered sustainable, scalable business value—helping farmers facing cultural and economic uncertainty achieve greater stability, profitability, and growth in their critical sector.

立即加入

微软人工智能和物联网实验室

即刻申请，获得与技术专家共同设计解决方案的机会。