Data Science and Digital Farming- Data Science and AI for Agriculture

"Without access to modern farming techniques or machinery, let alone science-based climate and weather data, farmers' livelihoods hinge precariously on a changing environment that they're struggling to understand" - U.S. Agency for International Development

Agriculture in India: Growth %, Revenue, GDP contribution

In today’s world, ‘digital farming’ is at the forefront of modern agriculture and is in use by an increasing number of farmers and data scientists across the globe. The percentage of the gross value added (GVA) of agriculture and allied sector to the total economy of India since 2018 has increased from 17.6% in 2018-19 to 20.2% for the year 2020-21 as per the data released by the National Statistical Office and Ministry of Agriculture and Farmers Welfare (1).These increasing values are a positive sign for developing relevant technologies and instilling further resources in the agricultural sector.

Setbacks for Indian Farmers and how Data Science can help:

From plowing land to farming, from harvesting to selling off to retailers: all cash based transactions go unrecorded, needless to mention the farmer's unaccounted time and energy. Crop yields, changes made in soil by selecting fertilizers and seeds, weather predictions, etc. earlier used to take place by observation. But, with the aid of AI tools such as remote sensing, soil fertility indicators, aerial/drone monitoring, agricultural efficiency will be furthermore increased, benefiting farmers.

Tech equipped farms would be able to get analyzed soil fertility levels, forecast crop yields, and more - with the help of Big Data and technologies such as Data Science and IoT. IoT connected sensors help in monitoring soil health while Deep Learning algorithms help in detecting crop diseases. With the likes of such data-backed technology, farmers and agribusinesses can surely grow without any efforts gone to waste.

Applications of Data Science in Agriculture:

Some components of data science which are at play here are:

  • MIS (Management Information Systems): a database where data such as crop yield, crop nutritional status, weed growth, crop count, soil fertility, soil texture, weather predictions, etc. is gathered, stored and analyzed using multiple sensors and similar technologies.

  • Neural Networks and advanced analytics: to detect crop diseases and manage pests. There are advanced algorithms which help in identifying changes in crop health and forecast pest infestation and crop disease spread.

  • IoT: sensors to scale fields, map satellite images of the same, monitor soil health

  • Automated irrigation systems responding as per weather prediction

Such data is assimilated by using a variety of sources and techniques such as 'random sampling' (for detecting soil fertility by taking small samples of the soil and test it in labs), predictive analysis to calculate soil and crop fertility rates, and using third party apps for weather forecasting : all in all help farmers and agribusinesses in increasing their efficiency and planning their logistics.

Companies working in this sector :

Surely there are a lot of companies, agribusiness and ‘agri-tech’ startups working for the benefit of the agricultural sector. One thing which many of them have in common is that they take care of the logistics and supply chain from farm to table, and help in providing on-call manual expert advice on soil health, crop fertility and more to the farmers. However, what we need is: to introduce and further incorporate smart, data-backed, scientific ways and technologies to make our farmers and agribusiness self-sufficient and more efficient.

Some companies making such remarkable changes in the agricultural sector are:

  • CropIn Technology, Aibono, SatSure, etc.: offering real-time weather updates, ability to manage farm activities and predict crop yields, and more.

About the Author:

Saumya Bharadwaj is an undergraduate student majoring in Computer Science. With a keen interest in Data Science and Business Analytics, she enjoys creating technical content in the form of site blogs and articles. She has undertaken several projects in Python and Database Management Systems and continues to develop further skills in this field.

7 views0 comments