IoT Based Precision Agriculture

1. Bangladesh agriculture sector is plighted by the following problems: i) Inadequate irrigation management ii) Excessive use of resources including fertilizers and pesticides iii) High cost of production for the farmers and iv) Largest greenhouse gas (GHG) emitting sector.

2. We are proposing to design, test and deploy Internet of Things (IoT) based Precision Agriculture System (Prototype). It employs data from multiple sources, including sensors, to improve crop yields and increase the cost-effectiveness of crop production.

3. The project aims to help:

• Farmers: Precision agriculture helps farmers to monitor soil and plant parameters, automated field management, and minimize pesticide, fertilizer and irrigation needs. It can determine peak conditions for plant growth and nutrients need of their crops.
• End Consumers: People will get safer products with acceptable level fertilizer and pesticide residues at an acceptable price.
• Government Agencies: It will enable a more sustainable agriculture sector in line with SDG Goals.

Bangladesh agriculture sector is vulnerable due to climate change, excess resource use and associated pollution, and high production cost.

Climate change will bring greater fluctuation in crop production, food supplies, and market prices will aggravate the situation of food insecurity and poverty in Bangladesh, which will adversely affect the livelihoods of millions of people of this country. Besides, agriculture is the largest user of water, can reach up to 90% of total water consumption, and also the largest GHG emission contributor. It is estimated that climate change will reduce 8-17% rainfall by 2050, which will have an impact of yield decrease by 8-10%.

Availability of agricultural land in Bangladesh is gradually declining at a rate of 1% per year. About 60% of farmers are functionally landless and depend on sharecropping of land owned by the others. The fertility status of soils is extremely variable and it is estimated that more than 100 kg nutrients/ha/year are mining out from the soil system. As a result, there is a sharp increase in fertilizer use and irrigation in Bangladesh agriculture during the last two decades.

All these problems can be mitigated by the use of IOT based precision agriculture techniques.

The solution proposed is geared towards the agriculture sector. In particular, the project will work with the farmers and researchers of this sector. The project team has undertaken a preliminary feasibility study in this regard and have discussed the idea and solutions with researchers and farmers. The farmers will be the most important stakeholder in implementing the project since they will be users of the prototype solution.

Recent research results clearly indicate that the climate change is significantly impacting the overall yield and cost of crop production. Our solution will integrate weather data, information on plant growth, and nutrients, pesticides and irrigation needs. This proposed solution aims i) to reduce water use, ii) to reduce fertilizer and pesticide use, iii) to increase productivity iv) to reduce production cost v) to develop a machine learning decision support tool and vi) to develop a comprehensive crop management database.

We will design and develop a functional IoT enabled scalable precision agriculture system prototype. The prototype will use a common crop (e.g., Brinjal) as use case to test the system. Since the technology has been successfully implemented in developed countries, we are confident that the test case will give us deeper insights to develop and deploy the IoT enabled precision agriculture system to large extent. At the end of the prototyping, we will conduct feasibility analysis and will develop a business model. Finally, we will develop the large scale deployment strategy of the prototype system to Bangladesh agriculture sector.

Our solution is based on the following four stages,

• Sense : Monitoring of critical plant parameters 24x7x365 from the farm and uploading the data to our cloud platform.

• Analyze : This data will then be analyzed and make accessible to the farmer anytime, anywhere on any device for data-driven decision making.

• Predict: Our prediction algorithm will continuously analyze the farm level data to predict the ideal growth conditions, resource requirements including irrigation, sprays and other preventive measures.

• Act: Finally, the Farmer gets notified on his device and acts accordingly.

The major technology and components includes: 

1. IoT sensors (e.g., soil temperature and moisture sensor, leaf wetness sensor, solar radiation sensor, soil nutrient sensor etc.)

2. Analytics software 

3. Network connectivity (available wireless communication technology) 

4. Machine learning platform

5. Crop life cycle database 

6. Mobile application (Android application)

Processes and Technology:

• IOT devices for data collection (weather, leaf moisture etc.)

• AWS Cloud Infrastructure with AWS Lake Formation to keep unstructured data

• Machine Learning model to predict growth conditions, resource requirements and crop disease outbreak so that Farmer can take precautions

• Android app to display modern farming related technical knowhow, farm based analytics and prediction or alert data.
  

The technology used in the project has proven hardware, software and processes, available in the developed countries. However, such IoT based precision agriculture has not been successfully tested as far as we are aware in Bangladesh. Hence, deployment strategies and economic model customized to Bangladeshi context is missing. In addition, the project will develop precision agriculture database analytics software with machine learning capability which is novel.

Thus, the major innovation in this project would be developing machine learning models for various types of crops to predict growth conditions, resource requirements and plant disease outbreak. This predictive model will make a huge impact on how a farmer operates and will help them to optimize the resources and stay informed about any potentially threats.

Precision agriculture technology has been utilized with success in various countries including Europe, USA and Asia and has yielded benefits as mentioned earlier. These are proven technology and adequate science is already documented. Hence, it is perceived that a customized IoT enabled precision agriculture system will successfully mitigate the problems identified.

Our machine learning model will analyze real-time data on plant growth, resource needs, and will optimize the predictive decision making. This overall process will reduce the climate change impacts on crop production and/or management with a greater precision.

1. 0

2. The project aims to develop the IoT enabled precision agriculture prototype system within 18 months from the start date of the project. Thus, the number will be "0" in one year.

3. The results of the project will be transferred to large scale trials in multiple segments of the Bangladesh agriculture sector from Month 19-Month 36, while mass commercial deployment is has been planned from Month 37-Month 60. we plan to serve 10% of Bangladesh agriculture sector during this third phase of commercial deployment.

We have a team involving academics and consultants with expertise in IT, project management, environment and life cycle analysis, brand management and operations management.

Dr. Saad Hasan is an Assistant Professor in the Department of Operations and Supply Chain Management at the American International University-Bangladesh (AIUB). Dr. Hasan have teaching, research, project management, and coordination experience related to operations and supply chain management in academic and industry settings in the UK and Bangladesh.  Some of the major research projects he has been involved with includes CONVERGE (EU funded), Etoile, (EU (funded), Unite (UK) etc. In addition, Dr. Hasan has provided research consultancy on UK construction supply chain during his tenure at the National Energy Foundation (UK).

Mr. Shovan Samaddar wth 8+ years of experience as Agile Project Manager, Product Owner and Scrum Master, I have led and worked across diverse business domains such as insurance, crowdfunding, e-commerce to digital advertising in different product platforms like Web, desktop, mobile and was also involved in digital transformation projects for the enterprises based in Europe and USA throughout the entire software product development life cycle.

Mr. Shafkat Reza Chowdhury is an Assistant Professor in the Department of Marketing at AIUB) since 2017. He has a  Master’s degree in International Business from Brunel University, London, UK. He also holds an LLB degree from University of London, UK. He previously had corporate experiences in various industries in Bangladesh and his areas of expertise are in the fields of Marketing, Sales, Service Marketing and International Business.