TITLE

Crop Selection and Rotation in Ranchi District using Artificial Intelligence and
Machine Learning

A SYNOPSIS PRESENTED TO SARALA BIRLA UNIVERSITY IN PARTIAL FULFILMENT OF

THE THESIS REQUIREMENT
FOR THE DEGREE OF
DOCTOR OF PHILLOSOPHY IN ENGINEERING AND APPLIED SCIENCE

UNDER THE GUIDANCE OF

Dr. Sanjeev Kumar

By
Abhinav Pathak
 INTRODUCTION

• The world population is growing at an alarming rate causing a significant increase in food consumption. Data driven agriculture is
one of the fastest adopted technologies in the developed countries that tends to increase production in the most efficient manner.

• Digital Farming or Smart Farming helps in seeding all the information such as crop pattern, soil health, weather patterns etc. into a
database and then analyses it through AI and data analytics.

• In a country like India, where nearly 50% of the population is associated with farming and contributes to a fifth of the economy,
Smart Farming can be game changer. It can improve production, boost rural incomes, reduce imports, and can help India compete
with developed nations.

• While the central government is seeking to ensure food security in the world’s second-most populous nation by signing
preliminary agreements with multinationals and a slew of local businesses, each state must try to adopt the technologies
associated with Digital farming to improve productivity with minimal resource usage.

• By gathering the data with respect to the quality of soil, fertile quotient, temperature and moisture levels etc. around the year, and
then examining them using Supervised Learning techniques, one can determine which crop can be cultivated in a particular area
with maximum output.

• It can help in determining if monocropping or multi-cropping/serial-cropping will increase the overall yield.

• Using the techniques associated with Digital Farming, the main idea of research is “Crop Selection and Rotation in the state of
Jharkhand using Artificial Intelligence and Machine Learning”.
 PROBLEM STATEMENT

• As per the Jharkhand state government report in the assembly in the year 2020, Jharkhand is the 4th poorest state in India with
agriculture being the most pronounced source of income for nearly 50% of the population.

• Digital farming is still in its nascent stage in Jharkhand

• Same set of crops are grown in a cyclic manner by the farmers that were cultivated by their forefathers

• For crop selection and rotation most of the farmers still use indigenous techniques or follow the current trend of adoption

To improve the overall yield of crops, Supervised Machine Learning must be applied on dataset compiled using the data of various
agencies in the state of Jharkhand to determine the correct crop selection and rotation.
 REVIEW OF LITERATURE

• Kumar, Spandana, Vaishnavi, Neha, Devi, 2020 highlighted that Supervised Learning techniques when applied on historical data
related to temperature, humidity, pH, rainfall, crop name, etc., can help in predicting the crop yield.

• Suganya, Dayana, Revathi, 2020 proposed that Crop yield prediction can be accurately calculated using Precision, Recall and F1
score.

• Brewster, Roussaki, Kalatzis, Doolin, Ellis, 2017 proposed that with the development of Internet of Things (IoT), data driven
technologies may become an important part of the future of farming.

• Gandhi, Armstrong, Petkar, 2016 proposed that the Rice crop yield is best classified by Bayesnet when the dataset related to
precipitation, minimum temperature, average temperature, maximum temperature, reference crop evapotranspiration, area,
production and yield are available.

• Foley et al, 2011 proposed that data driven farming will help increasing the production while decreasing the overall environmental
impact.
• Goldfray et al, 2010 highlighted that the increase in population will have proportional increase in the global demand of food and
the climate change will also complicate the sustainability associated with food security.
• Singh, Prakash, 2017 highlighted how the cropping pattern has shifted in Jharkhand from lower value crop (paddy) to higher value
crop (wheat)

• Lakra, Singh, Sinha, Kudada, 2010 captured how the farmers of Jharkhand grow several crops simultaneously in one field (multi-
cropping) using indigenous technologies.
 SIGNIFICANCE OF THE RESEARCH

• While the adoption of Smart/Digital Agriculture is still in its initial stages in Jharkhand, it may not get accelerated immediately due
to the operational costs that it brings along with it. The installation of various IoT devices and their maintenance will be an
overhead to start with and will be far-fetched for farmers that have few acres of land to take care of their livelihood.

• Keeping this thought in mind, the best alternative is to mine the data that is there with the state/central government and come up
with a proposal that can help farmers to take conscious decision that is backed by scientific data.

• The research will help the farmers in Crop selection and Rotation thereby increasing the overall production.

• As and when the installation of IoT devices happen, over a period of time with State government’s assistance or through private
firms’ aid, the quantity of data captured will be large and will give better predictions for a particular area.
 OBJECTIVES OF THE RESEARCH

The main objectives of Research are:

• To access and adopt the concept of Digital faming /Smart Agriculture for improving the overall yield of crop production in
Jharkhand

• To determine the correct/appropriate crop selection process and rotation strategy in the various districts of Jharkhand.

• To understand and apply the existing Supervised Learning techniques incorporated in making decisions wrt Digital Farm

• To become familiar with Major and Minor crops cultivated in the state of Jharkhand along with soil types and pattern of farming
(monocropping or multi-cropping or serial-cropping)

• To find the most thriving conditions for these primary and secondary crops.
 HYPOTHESIS

The prime focus of the research it to determine the selection of primary and secondary crop and the rotation strategy in the state of
Jharkhand based on Machine Learning and Artificial Intelligence.

The Hypotheses associated with the research are as follows:

Hypothesis 1
H0 : Usage of Machine Learning and Artificial Intelligence has no significant impact on the overall selection process of primary and
secondary crop and rotation strategy in the state of Jharkhand
H1 : Usage of Machine Learning and Artificial Intelligence has significant impact on the overall selection process of primary and
secondary crop in the state of Jharkhand

Hypothesis 2
H0 : Digital farming has no significant impact on optimising the overall yield crop production in the state of Jharkhand
H1 : Digital farming has significant impact on optimising the overall yeid crop production in the state of Jharkhand
RESEARCH METHODOLOGY

Define the Research Problem

Literature Review / Previous Findings

Formulate Hypothesis

Prepare the Research Design

Collect the data from private/government

agencies wrt type/quality of soil,
temperature and moisture levels around the
year, primary and secondary crop and rainfall

Analyze data using various Supervised

Learning techniques, like, Bayesnet,
Precision, Recall and F1 score etc.

Generalise and Report

 EXPECTED CONTRIBUTION FROM RESEARCH

• The main aim of research is to form a framework that will help in analyzing the historical data and the data obtained using Smart
framing techniques for the state of Jharkhand and then accurately predict the right crop selection and rotation strategy to improve
the overall throughput of the farmlands.

• The contribution will form a steppingstone for all future work that will happen in the state of Jharkhand wrt Digital Farming and
will give a platform to future researchers to fine tune the Supervised Machine Learning techniques to provide better prediction in
future when improved IoT devices are integrated in the Agriculture domain.
• After the research , ground level implementation and mobile app can be developed for farmers.
 SCOPE OF THE RESEARCH

• The scope of the research is to establish a predictable model for Crop Selection and Rotation in the state of
Jharkhand using historical data associated with soil, humidity, temperature, fertile quotient etc.

• The research will depend on two databases

• A generic database that captures the ideal conditions under which crops thrive and their rotation policy
• An additional database capturing historical data associated with soil, humidity, temperature, fertile quotient
etc. for various districts in the state of Jharkhand.

• Using these two databases, a prediction algorithm will be developed that will use Supervised Learning Techniques
to determine the most appropriate Crop selection and rotation policy.

• The research will also provide scope of how integration of IoT devices in future will help in better prediction based
on the live data feed that will be generated and processed on the go to further improve the overall prediction with
respect to Crop Selection and Rotation.
 TENTATIVE CHAPTERISATION OF THE RESEARCH

• INTRODUCTION

• BACKGROUND

• PROCEDURE

• DATA COLLECTION AND ANALYSIS

• RESULTS AND DISCUSSION

• LIMITATIONS AND FUTURE SCOPE OF THE STUDY

• REFERENCES/BIBLIOGRAPHY

• ANNEXURE/APPENDICES
 CONCLUSION

• As per the Jharkhand state government report in the assembly in the year 2020, Jharkhand is the 4 th poorest state in India with
agriculture being the most pronounced source of income for nearly 50% of the population.

• The whole idea of the research is to predict the crop selection and rotation to improve the productivity of the farming lands by
using Smart/Digital Agriculture techniques, thereby improving the economic condition of the farmers.

• The research will focus on prediction algorithm that will use the historical data associated with soil, humidity, temperature, fertile
quotient etc. for various districts in the state of Jharkhand to determine the best possible crop selection and rotation strategies.

• This will have positive impact wrt food availability for the ever-growing population of the state.
 REFERENCES / BIBLIOGRAPHY

• A. Patil, S. Kokate, P. Patil, V. Panpatil and R. Sapkal, "Crop Prediction using Machine Learning Algorithms", International Journal of
Advancements in Engineering & Technology, vol. 1, no. 1, pp. 1-8, 2020.

• Brewster, C., Roussaki, I., Kalatzis, N., Doolin, K., & Ellis, K.,(2017). IoT in agriculture: Designing a europe-wide large-scale pilot. IEEE
Communications Magazine, 55(9), 26-33. https://doi.org/10.1109/MCOM.2017.1600528

• Foley, J. A., Ramankutty, N., Brauman, K. A., Cassidy, E. S., Gerber, J. S., Johnstan, M., et al. (2011). Solutions for a cultivated planet.
Nature, 478(7369), 337-342. https://doi.org/10.1038/nature10452

• Godfray, H. C. J., Beddington, J. R., Crute I. R., Haddad L., Lawrence, D., Muir, J. F., et al. (2010). Food Security : The Challenge of
feeding 9 billion people. Science, 327(5967), 812-818. http://doi.org/10.1126/science.1185383

• K. G. Liakos, P. Busato, D. Moshou, S. Pearson and D. Bochtis, "Machine learning in agriculture: A review", Sensors, vol. 18, no. 8,
pp. 2674, 2018.

• Lakra, V., M. K. Singh, Rekha Sinha & N. Kudada (2010), Indigenous Technology of Tribal Farmers in Jharkhand, Indian Journal of
Traditional Knowledge, Vol. 9, No. 2, 261-263.

• Manishkumar Dholu, K.A. Ghodinde, "Internet of Things (IoT) for Precision Agriculture Application", Trends in Electronics and
Informatics (ICOEI) 2018 2nd International Conference on, pp. 339-342, 2018.

• N. Gandhi, L. Armstrong and O. Petkar, "Rice Crop Yield Prediction using Artificial Neural Networks", proceedings of International
Conference on 2016 IEEE Technological Innovations in ICT for Agriculture and Rural Development (TIAR), 2016.
 REFERENCES / BIBLIOGRAPHY

• Olakunle Elijah, Tharek Abdul Rahman, Igbafe Orikumhi, Chee Yen Leow, MHD Nour Hindia, "An Overview of Internet of Things
(IoT) and Data Analytics in Agriculture: Benefits and Challenges", Internet of Things Journal IEEE, vol. 5, no. 5, pp. 3758-3773, 2018.

• R. Yadav, S. Yadav, N. Gunjal and S. Mandal, "Agricultural Crop Yield Prediction Using Deep Learning Approaches SVM Mutiple
Regression Random forest Regression", 2008.

• Singh, R. P., and Prakash, J. (2017). Transformation of Agriculture in Jharkhand. Agro-Economist, 4(1), 45- 54.

• Suganya, M., Dayana, R., Revathi, R.: Crop yield prediction using supervised learning techniques. Int. J. Comput. Eng. Technol.
(IJCET) 11(2), 9–20 (2020)

• Thomas GW, Blevins RL & Philips SH (1984) No-tillage in the tropics. In: Philips RE & Philips SH (eds) No-Tillage Agriculture:
Principles and Practice. pp 270–301. New York, NY, USA: Van Nostrand Reinhold Co Inc

• T. van Klompenburg, A. Kassahun and C. Catal, "Crop yield prediction using machine learning: A systematic literature
review", Computers and Electronics in Agriculture, vol. 177, pp. 105709, 2020.

• Y. J. N. Kumar, V. Spandana, V. S. Vaishnavi, K. Neha and V. G. R. R. Devi, "Supervised Machine learning Approach for Crop Yield
Prediction in Agriculture Sector", 2020 5th International Conference on Communication and Electronics Systems (ICCES), pp. 736-
741, 2020, June.