Global Journal of Science Frontier Research, A: Physics and Space Science, Volume 23 Issue 1

Data-Driven Knowledge Agriculture: A Paradigm Shift for Enhancing Farm Productivity & Global Food Security Aprajita Srivastava α & Dr. H. O. Srivastava σ Keywords: precision agriculture, knowledge agriculture, data-driven agriculture, SDG2030, farm data visualization, agriculture analytics, agriculture business intelligence. I. I ntroduction nowledge Agriculture TM is a new way of farming that uses technology and tools such as IOT (Internet of Things), Robotics, AI (Artificial Intelligence), UAV (Unmanned Aerial Vehicles), Cloud, Greenhouse, BA (Business Analytics)and specialized software for weather modeling, smart zone seeding, fertilizer modeling, (Srivastava, 2018; Definitions,2022) , to address issues relating to food requirements of current 8.0 billion world population, estimated to grow to 9.8 billion in 2050 (PRB, 2016) . The current practice of farming in most of the countries is based on traditional methods of sowing, harvesting, storage and marketing which results in poor productivity and profit. We define agricultural productivity as the ratio of outputs to inputs, expressed either in volumes or in physical quantities (kg, tons, etc.) (FAO, 2018) . TFP (Total Factor Productivity) is a measure of the efficiency of the contribution of all the significant inputs into production. TFP provides a complete picture of productivity and is more closely connected to unit production costs and market prices than partial productivity indicators (Fuglie, 2015) . There is a close relationship between agricultural productivity and farm incomes. Considering this, increasing farm income is at the center of measures related to food security, rural livelihoods, ending hunger in Africa and other parts of the Globe by 2030 and fulfilling SDG2030 targets on food and nutrition security. Despite the importance of agricultural productivity, data on various parameters required to analyze and deduce the current productivity and forecast the model for improved productivity is scarce and of poor quality. There is a need for new and enhanced data collection frameworks to better measure agricultural production and the amounts of inputs used in the production processes. Data is becoming an ever-important factor in the world, and businesses are developing the skill to use data analysis as a tool for enhanced profit. Agriculture is not immune to this, and data-driven decisions are essential for improved agricultural productivity. Data analysis focuses on cleaning, modeling and visualizing data to provide descriptive, diagnostic, predictive, prescriptive and cognitive analytics. Descriptive analytics based on historical data may be used to generate reports to give a view of a farmer’s production, sales and financial data. Diagnostic analytics helps in answering questions about why certain events happened for example anomalies that might be due to unexpected changes in a metric or a particular market. Predictive analytics techniques use historical data to identify trends, for instance changes in demand or consumption and determine if they're likely to recur. Prescriptive analytics is an application of(ML) (Machine Learning)that prescribes optimal actions to achieve a goal or target, for example use of lesser pesticides due to consumer’s behavior of changing to organic food. It may also help farmers provide historical crop yield record with a forecast reducing risk management. Machine learning uses statistical algorithms that involve statistical and functional analysis of existing data to learn, a process called training. Patterns and relationships in the data identified during training are used to build a model. The model makes intelligent decisions about data it hasn't encountered before. A process called inference is used to make decisions K 1 Year 2023 17 © 2023 Global Journals Global Journal of Science Frontier Research Volume XXIII Issue ersion I VI ( A ) Corresponding Author α : Cloud support engineer in analytics, and Hon. Director of World Development Foundation undertook R&D projects for use of ICT for agriculture in India and Ethiopia for increased productivity and social projects for citizens to express themselves socially, culturally, politically and spiritually, Krishi Apts. Vikaspuri-D, New Delhi- 110018, India. e-mail: aprajitaedu1311@gmail.com Author σ : Ph. D., FIETE, Former Addl. Director General & Head of AIR Resources of All India Radio and Doordarshan (Level of Addl. Secretary to the Govt. of India), is the President and CEO of World Development Foundation. Abstract- Data-driven Knowledge agriculture using mechanized intelligent computer-based monitoring and control systems and complex Software for machine learning and visualization for predicting a variety of parameters such as future food requirements, resource planning for higher yield, and supply chain is the future of farming. This needs to be urgently adopted by the world farming community to provide food to the growing world population, remove hunger, and at the same time sustain planet resources by judicious uses of input such as water, fertilizer, pesticide etc., as envisioned by Sustainable Development Goals 2030. This paper discusses data-driven technology for identifying trends and other insights for making informed decisions for enhanced productivity and profitability, through market research and evaluating customer needs and sentiments.