The concept of an AI project promises a higher yield for agribusiness

Farmers may be able to make more informed decisions and earn more money with the help of a novel, AI-driven method for crop planning and purchasing agricultural inputs.

Based on deep learning and game theory, GRAMA (Game Theory, Random Processes, Artificial Intelligence, and Machine Learning for Agriculture) is a NABARD-backed project.

Experts from the University of Agricultural Sciences, GKVK, are involved in the project, which is directed by Y Narahari, an honorary professor at the IISc’s Department of Computer Science and Automation.

Through crop planning, crop recommendation, procurement processes, market mechanisms, and price prediction, GRAMA seeks to maximize farmers’ earnings.

The initiative makes use of the data already available on soil types, crop varieties specific to a district, demand, yield, past prices, seasonal fluctuations, and sociocultural influences. If farmers used the proper changes in crop combinations, researchers predicted that their income would increase by 60%.

It indicates that we aren’t even correctly planning the crops. Prof. Narahari told DH that people who have farmed a certain crop variety for a long period might not want to switch to a new crop because they think they already have the necessary knowledge and are resistant to change.

A feature of the model allows one to ascertain how changes in crop combinations affect revenue. “The revenue can increase by 10% even with a 50% compliance rate,” he said.

Enhancing Farmer Producer Organizations (FPOs) is deemed essential by the researchers for expanding the study. They started experimental projects at FPOs in Rajaghatta and Anekal earlier this year, and the FPOs were very enthusiastic, even if the farmers needed to be persuaded about the technology. Prof. Narahari thinks that GRAMA could eventually become an essential part of farming methods, nevertheless, if progressive farmers get more involved and the government issues a mandate.

Bharat Electronics Limited (BEL) developed a mobile app that was used at the two FPOs to evaluate a pilot version of the procurement system. The farmers’ cooperatives gather the necessary information from the farmers. Based on these requirements and current information on several parameters including weather and soil type, the app advises aggregate amounts of seeds, fertilizers, and pesticides to be acquired. The concept suggests a framework that ensures farmer procurement at a reasonable price with consistent quality.

According to Prof. Narahari, the implementation of GRAMA required support from initiatives to reduce transportation and maximize the utilization of available warehouses and other resources. He also mentioned several real-world difficulties.

According to Prakash Kammaradi, an agricultural economist, future farming innovation must offer more specific and financially rewarding price discovery, produce aggregation, and efficient distribution mechanisms.

“Agrarian problems are very complex and challenging. In addition to field experience, the application of cutting-edge technology can give us a deep understanding of these problems. However, solely relying on it for solutions might not be feasible in the long term,” he added.