Leveraging Generative AI for Data Analysis in Farm Management

Highlights Generative AI successfully analyzed data files related to machinery maintenance, financial health, and soil/yield. The complexities of APIs and custom GPT integration were demonstrated with live public and private data sources. With constrained prompts, Generative AI was able to analyze live weather data, field records, and IoT sensor readings. Abstract. Generative AI is an emerging field with transformative potential for farm data analysis that could reduce knowledge barriers, enhance digital solutions, and enable seamless interaction with multiple data sources. In this work, ChatGPT Plus with OpenAI’s GPT-4o model was used to demonstrate applications with dissociated data (imported CSV, Excel, and PDF documents) regarding machinery maintenance records, financial statements, and combined yield and soil spatial data. ChatGPT successfully analyzed these files to identify trends in fuel costs and equipment servicing, interpret financial health using university extension benchmarks, and rank soil types by productivity based on multi-year yield data. Visualizations were generated, and performance was generally strong, though assumptions and formatting varied slightly based on prompt structure. For integrated data, custom GPTs were developed for ChatGPT Plus to interact with public weather data, a private field records database, and a private IoT sensor SQL database to enable real-time insights utilizing APIs. These integrations supported use cases such as weather-informed decision-making for cover crop termination, retrieving field operation records with schema-aware querying in Airtable, and comparing real-time soil moisture levels across farm fields. Generative AI produced accurate responses to complex queries in these demonstrations, but careful prompting was necessary, and some custom coding or schema familiarity was required in live/integrated data scenarios. As these technologies evolve, they could streamline data workflows, reduce AgTech development costs, and lessen the need for highly specialized tools, making advanced analytics more accessible and affordable for farmers.