Enabling and Empowering Farmers through AI and Game Theory

The agriculture sector contributes to 16% of India’s GDP employing almost 50% of the total work-force. 85% of the farmers are small and marginal, holding less than 5 acres of land and they do not have access to technology that is accessible to the more prosperous farmers.  Agriculture in India is largely dependent on nature; variable climate and more recently global warming make farming both challenging and volatile. Small and marginal farmers continue to be trapped in a vicious circle of low growth, low income, and high debt.

We are working with NABARD (National Bank for Agriculture and Rural Development), Government of India, to develop algorithms to help farmers. These algorithms will be powered by Artificial Intelligence and Game Theory principles.

The following are the problems that we are currently looking into.

Crop Price Prediction

Agriculture crop prediction, especially for commodities with volatile prices such as tomato and onion,  is a grand challenge problem since it depends on so many factors. If accurate predictions are available, it will help farmers to plan when to sow, when to harvest, etc. and will help the policy makers to plan farmer-centric interventions. We are trying to advance the state-of-the-art in this research problem through use of contemporary machine learning and deep learning models in conjunction with advanced data imputation techniques. We are also trying to model demand side factors besides supply side factors.

Crop Recommendation

Farmers often suffer terrible losses due to wrong choice of crops. We are using machine learning, deep learning, and risk modelling techniques to suggest optimal portfolios of crops for an entire year (encompassing Rabi, Kharif, and intervening seasons). Cultural factors, demand/export uncertainty, and climate variability play an important role here and our objective is to build explainable models to convince the farmers to adopt the portfolios recommended.

Agri-Market Design

Farmers are not geared towards a sophisticated use of markets and often end up with unfairly low prices. Our objective is to design social welfare maximizing markets  that are attractive to farmers as well as traders and consumers. We are trying to use techniques from mechanism design to suggest markets that are best suited to agriculture domain.

Famer Collectives

A farmer producer organisation (FPO) or farmer collective is a cooperative collective of farmers which enables economies of scale to be exploited by a large community of farmers coming together under one umbrella.   There are more than 6000 FPOs are in existence in India currently and these are  formed under various initiatives of the Government of India (including SFAC- Small Farmers Agri-Business Consortium), State Governments, NABARD (National Bank for Agriculture and Rural development) over the past decade. Though there are some success stories, FPOs, by and large, are still in the expectation hype cycle. There is a critical need to modernize their operations and provide technical advice and technology support to enable them to better realize the above FPO objectives. In this problem, we intend to use a data-driven, Artificial Intelligence based approach for solving key challenges faced by FPOs. In particular, we have identified the following problems: (1) Reduce input costs to farmers (2) Provide advisories to farmers (3) Increase revenue to farmers from farm output.

AIML Pipeline for Agriculture Domain

There are many generic problems in the agriculture domain for solving which we need sophisticated computation pipelines involving a variety of AIML techniques. These problems have certain unique characteristics that merit a separate workflow and pipeline to be defined to architect a computational platform. We are putting together such an AIML pipeline called AGRI-VAHAN.

Our team consists of:

  • Y. Narahari
  • Mayank Ratan Bhardwaj (Ph.D. Student)
  • Inavamsi Enaganti (Project Associate)
  • Deepanshu (M.Tech. (CSE) Student)
  • Azal Fatima (M.Tech. (CSE) Student)
  • Meghnath Singh (M.Tech. (AI) Student)
  • Rohit Patel (M.Tech. (CSE) Student)
  • P. Sowjanya (M.Tech. (AI) Student)
  • Vaidehi Bhaskara (Research Intern)