Cognitive Agtech framework: A step towards farmers' prosperity

With monsoon Gods pouring their blessings over the Southern and Western India, it’s time to revisit the ambitious promise of the Indian government to double up farmers’ income by 2022. This is ambitious, especially when the Indian farmers have been blighted by drought and crop prices have collapsed in the last 4 years.

Last week the prime minister had a long round of discussions with the top industrialists and CEOs of the country and Agriculture was again one of the priorities he pushed on the agenda for these leaders. No wonder, agri-tech is the new buzz-word in the Indian startup ecosystem. But where does the opportunity lie in this rather fragmented, unorganized and malpractices-struck sector?

Even though the agricultural exports have shown a significant growth over the last couple of years (CAGR of over 7-8%), the fundamental issues of decreasing landholding size, poor input quality, reduced availability of water and uneven technology adoption/mechanization remain the same, hampering the overall growth of this sector in India. To add to that, we have sheer lack of organized market-linkage for the produce, inadequate facilities for storage (especially for the perishables) which further bring the pace of growth down.

Here comes the opportunity – from farm to fork – of digitalization of this complex value chain, which many of the start-ups are trying to tap into. It’s quite obvious though that solving the challenges as stated above is non-trivial - not only because of the socio-political reasons, but also because Agriculture is a very complex subject and needs to be addressed holistically than in pockets. That’s what we at IBM realized last year and came up with a ‘Cognitive Agribusiness’ framework. I call it a ‘bold’ step because this framework brings together the big data from on-field sensors, aerial drones, remote sensing satellites and weather sources together to generate actionable insights for the farmers, agronomists and several other stakeholders in the Agribusiness value chain to address the challenges of this industry.

This framework dwells on a 3-layered architecture, wherein, the bottommost layer is a data lake in cloud where IBM’s Watson IoT Platform ingests and aggregates data from specialized IoT/local sensors, remote sensing and weather sources; the second layer above that is the Artificial Intelligence or analytics layer where the data underneath is used to derive valuable insights such as ‘Crop vigor analysis (High Definition corrected NDVI)’, ‘Pest/Disease Prediction Maps’, ‘Soil Moisture Insights’ and so on , and the topmost layer is application layer which consumes these analytic insights into full-fledged applications such as Precision Farming, Irrigation Management, Farm Management etc.

Through this framework, we not only demonstrated the technological benefits, but also established the business value such as yield improvement, input optimization and pest/disease management, which we accomplished while working with our partnering clients in India and globally. On top of that, the value and need for this framework was further reinforced with IBM winning one of the most prestigious industry awards – Aegis Graham Bell Awards, where my team and I competed against several great innovations from large enterprises as well as growing start-ups. I hope to scale up these successful pilots into much bigger initiatives, which would then become probably the biggest contribution in fulfilling the dream of farmers’ prosperity in India.