COMPUTER SCIENCE SCHOOL

GRADUATION PROJECT
Fall 2023/2024

Smart Agriculture Platform: A Data-Driven System for

Project title
Optimized Irrigation in Agriculture

Submitted By:

# Name ID Major Email

1 George Isacc Ezzat 202100202 Cloud [email protected]
2 Zeyad Osama Mohamed 202100393 Cloud [email protected]
3 Eyad Amr Mohamed 202100309 Cloud [email protected]
4 Seif El-Deen Mohamed 202100139 Cloud [email protected]
5 Yousef Ahmed Abdelmohsen 202100553 Cloud [email protected]
6 Shahd Hassan Mohamed 202100762 DS [email protected]
7 Zeiad Haitham Mohamed 202100140 DS [email protected]

Supervised By:

Supervisor Professor: Dina Salem

Table of Contents

ACKNOWLRDGEMENTS………………………………………………………………

ABSTRACT………………………………………………………………………………..

LIST OF ABBREVIATIONS…………………………………………………………….

LIST OF TABLES………………………………………………………………………...

LIST OF FIGURES……………………………………………………………………….

CHAPTER 1: INTRODUCTION………………………………………...........................

CHAPTER 2: MARKET RESEARCH AND PLANNING…………………………….

CHAPTER 3: PROPOSED DESIGN AND NEXT PLAN……………………………...

REFRENCES……………………………………………………………………………...
 ACKNOWLEDGEMENTS

We would like to express our sincere gratitude to all those who have supported and guided
us throughout the development of this project.

First and foremost, we extend our deepest appreciation to our supervisor, [Dr/Dina Salem],
for their valuable guidance, constructive feedback, and continuous encouragement. Their
expertise and insights have been instrumental in shaping this project.

We also grateful to our family for their unwavering support, patience, and motivation
during this journey. Their encouragement has been a constant source of strength.

We would like to acknowledge corelia company specially Eng/Noor the team leader, who
have provided invaluable advice, technical discussions, and moral support throughout this
process and this wouldn’t happen without the opportunity of computer science department

Finally, we extend our appreciation to anyone who has directly or indirectly contributed to
this project. Your support has been truly invaluable.

Thank you all.

 ABSTRACT

Agriculture plays a crucial role in food production and economic stability, yet inefficient
irrigation practices lead to water waste, soil degradation, and reduced crop yields. To
address these challenges, this project proposes IRRIGO, a smart agriculture system that
assists farmers in managing irrigation efficiently. The system integrates real-time weather
and soil monitoring through APIs and provides intelligent decisions for optimal irrigation
using neural network

The proposed system consists of a mobile application that allows farmers to input their
land locations using GPS and specify the crop types they are cultivating. The system
collects weather data, including temperature, humidity, precipitation, wind speed, wind
direction, and air pressure, alongside soil parameters such as surface temperature,
subsurface temperature (at 10 cm depth), and moisture levels. These updates occur twice
daily at 12 AM and 12 PM.

By analyzing the collected data, the system determines whether it is an appropriate time
to irrigate based on weather conditions. Additionally, it calculates the exact amount of
water needed by considering soil type (clay or sandy or loamy), crop water requirements,
and current soil moisture levels. If the soil does not have sufficient water, the system
calculate the remaining water required to achieve optimal soil hydration for healthy crop
growth and take the decision for irrigation

The system is powered by a neural network model that continuously learns and improves.
During the shadow phase, if the model's accuracy drops below 90%, expert farmers
testing the system are prompted for feedback to enhance decision-making.

This smart irrigation system enhances water conservation, improves crop productivity,
and reduces the risks of over- or under-irrigation. The platform empowers farmers with
data-driven insights, making irrigation decisions more efficient and sustainable.
5

LIST OF ABBREVIATIONS

API Application Programming Interface

GPS Global Positioning System

AI Artificial Intelligence

ML Machine Learning

DB Database
vi

LIST OF TABLES

Table Page

Table 1: Comparison of Traditional vs. Smart Irrigation ……………………….….Page 9

Table 2: Irrigation Recommendations Based on Soil Moisture …………………….Page 16

Table 3: Weather and Soil Data Collected …………………………………………Page 17

Table 4: Weather-Based Irrigation Scheduling …………………………………….Page 18

Table 5: Soil Type and Water Retention Capacity …………………………………Page 18

Table 6: Crop Water Requirements Based on Soil Type …………………………..Page 19

Table 7: Weather API Dataset……………………………………………………...Page 20

Table 8: Soil API Dataset…………………………………………………………..Page 21

 LIST OF FIGURES

Figure Page

Figure 1: Traditional vs. Smart Irrigation Water Usage Comparison…………….. Page 15

Figure 2: Soil Moisture vs. Irrigation Need Graph………………………………… Page 16

Figure 3: Difference between good and bad irrigation……………………………..Page 16

Figure 3: Data Flow Diagram DFD………………………………………………...Page 22

Figure 4: ER Diagram………………………………………………………………Page 23

Figure 5: Use Case Diagram………………………………………………………..Page 24

Figure 6: Sequence Diagram……………………………………………………….Page 25

Figure 7: Activity Diagram…………………………………………………………Page 26

Figure 8: Gantt Chart……………………………………………………………..,,.Page 27

 8

CHAPTER 1

INTRODUCTION

In the modern era, agriculture faces significant challenges due to climate change, water
scarcity, and inefficient irrigation methods. Farmers often struggle to determine the
optimal time and amount of water required for irrigation, leading to water wastage or
insufficient irrigation, both of which negatively impact crop yield. To address these
challenges, smart agriculture technologies have emerged as a crucial solution for
enhancing water management and optimizing crop growth.

This project proposes IRRIGO, a smart agriculture system that helps farmers optimize
irrigation through a mobile application integrated with weather and soil monitoring APIs.
The system provides real-time updates on weather conditions and soil moisture levels,
analyzing this data to determine and take decision whether irrigation is necessary and the
precise amount of water needed.

Using a neural network model, IRRIGO ensures efficient water use by considering crop
type, soil type, and current moisture levels. During its shadow phase, expert farmers
provide feedback if accuracy drops below 90%. This system aims to reduce water waste,
enhance irrigation efficiency, and improve crop productivity.

This chapter provides a comprehensive overview of the project, including its objectives,
purpose, scope, and constraints.

1.1 Overview
This project focuses on the development of a smart irrigation assistant for farmers,
incorporating weather and soil monitoring APIs into a mobile application. The system collects
and analyzes real-time data on weather conditions (temperature, humidity, precipitation, wind
speed, etc.) and soil parameters (moisture levels, surface and subsurface temperature). Based
on these inputs, the system provides precise irrigation decision , ensuring optimal water usage
for different crops and soil types.
 9

1.2 Objectives
The aim of this project is to develop an intelligent irrigation system using a Neural Network
model that provides real-time decisions based on weather and soil data. The system focuses on
optimizing irrigation processes to reduce water wastage by determining the precise amount of
water required based on soil moisture levels and crop water needs. By ensuring appropriate
water distribution, the project aims to improve crop yield and promote healthier plant growth.
Modern technologies, such as APIs and mobile applications, will be integrated to assist farmers
in their decision-making process, offering real-time insights and recommendations.
Additionally, the system will automate data collection and analysis, significantly reducing the
need for manual monitoring and enabling farmers to focus more on crop management and
productivity.

1.3 Purpose
The purpose of this project is to provide an efficient, automated, and data-driven irrigation
management solution that helps farmers make informed decisions regarding irrigation. By
using real-time weather and soil data, the system ensures optimal water usage, thereby
addressing water conservation issues and improving agricultural productivity.

And here you are a small comparison between Traditional Irrigation & Smart Irrigation
Factor Traditional Irrigation Smart Irrigation (Our System)
Water Usage High, often excessive Optimized based on real-time data
Cost Higher due to water waste Lower due to efficiency
Decision Making Manual guesswork Data-driven, automated
Crop Health Risk of over/under-irrigation Optimal hydration for better growth
Environmental Impact Wastes water Supports sustainable farming
 10

1.4 Scope
• Planning and research on smart irrigation methodologies and agricultural water
requirements.
• Designing and developing an user friendly mobile application that integrates
weather and soil monitoring APIs.
• Implementing a data analysis AI model to evaluate soil moisture, weather
conditions, and crop needs.
• Providing irrigation decisions based on analyzed data.
• Testing and evaluating the system’s performance and accuracy.
• Documenting the development process and findings.

1.5 General constraints

• Time constraints: The project required extensive research, development, and
testing within a limited timeframe.
• Data availability: Accessing real-time and accurate weather and soil data posed a
challenge in some regions.
• Complexity of irrigation models: Designing an algorithm that accurately
determines irrigation needs for various crops and soil types required rigorous
testing and refinement.
• Hardware limitations: The system’s efficiency depends on reliable API
integration, and limited sensor availability may affect real-time data accuracy.
• User adaptability: Farmers may require training and adaptation to use the mobile
application effectively
 11

CHAPTER 2

MARKET RESEARCH AND PLANNING

2.1 Project planning

2.1.1 Feasibility Study
This project is feasible due to the increasing adoption of smart farming technologies, the
availability of weather and soil APIs, and the need to reduce water wastage in agriculture.
The system is cost-effective, scalable, and beneficial to farmers by enhancing productivity
and sustainability.

2.1.2 Market and Buisness View

❖ Global and Regional Trends in Smart Agriculture

• Global Smart Agriculture Market Growth: The smart agriculture market is
growing rapidly, driven by the increasing need to optimize water usage and improve
crop yield. By 2025, the global smart agriculture market is projected to reach $20+
billion, with a compound annual growth rate (CAGR) of over 9%.

• Regional Adoption (MENA Region): The Middle East and North Africa (MENA)
region faces serious challenges related to water scarcity and food security.
Countries like Egypt, Saudi Arabia, and the UAE are heavily investing in smart
agriculture solutions, including IoT-based irrigation systems, precision farming,
and AI-driven agriculture platforms.

• Government Initiatives in Egypt: Egypt’s government has launched multiple

projects to promote smart irrigation systems and digital transformation in
agriculture.
- The National Strategy for Water Resources 2037 focuses on sustainable
water management.
- The 100,000 Feddan Project is a large-scale initiative to reclaim and
cultivate desert areas using modern technologies.
12

❖ Current Issues in Traditional Agriculture in Egypt and Arab Nations

• Water Scarcity: Egypt relies heavily on the Nile River for agriculture. The average
water share per person is declining and now falls below 500 m³ per year, classifying
Egypt as a water-scarce country. Efficient irrigation is crucial.
• Inefficient Water Usage: Traditional irrigation practices such as flood irrigation
result in water waste. Smart irrigation systems can reduce water consumption by
up to 30%, saving millions of cubic meters annually.
• Climate Change Impact: Increasing temperatures, unpredictable weather patterns,
and desertification are major challenges for farmers.
• Limited Access to Real-Time Data: Most small and medium-sized farms lack
access to weather and soil data, making it difficult to manage irrigation effectively.

❖ Business Opportunity and Target Market’

• Target Customers
1. Individual Farmers: Especially those with medium to large-scale farms.
2. Agricultural Cooperatives and Associations: Supporting farmers with
shared services.
3. Government and NGOs: Partnering on projects to promote sustainable
agriculture.
4. Agro-Tech Startups and Investors: Offering a white-label version of the
platform or integration with other smart systems.

• Market Size and Demand in Egypt

- Agricultural Land in Egypt: ~9.6 million feddans. Most farms still use
traditional irrigation methods, creating a huge demand for smart solutions.
- Key Crops in Egypt: Wheat, corn, rice, potatoes. Your system can
initially target these high-demand crops.
- Potential Cost Savings for Farmers: By optimizing water usage and
reducing crop failure due to weather, farmers can increase profit margins by
10-20%.
13

❖ Competitive Advantages that Let us Compete

• Real-Time Weather and Soil Data Integration by using APIs and
limitation of using sensors and hardware which is cost effective
• Providing farmers with localized, accurate weather and soil updates
twice daily (12 pm and 12 am).
• AI-Powered Analysis & Decision-Making
• Water-Level Analysis: Model calculating the current moisture level to
know how much water is needed for optimal irrigation.
• Crop-Specific Insights: Irrigation descions based on the type of crop and
soil (sandy vs. clay vs. loamy), which many systems don’t offer that.
• User-Friendly Mobile Application specific for farmers which they can
understand easily
• Allows farmers to easily input their farm’s location, crop type, receiving
automated alerts.
• Focus on Water saving to be really helpful for farmers to catch up them to
buy the system
• Low-Cost
• Offering a more affordable solution compared to expensive imported
systems to make it easy on start up farmers and small farms.
• Scalable to different crops specially with trendy ones in Egypt and arab
nations to compete in this region
14

❖ Competitive Landscape in Egypt and Arab Nations

• Direct Competitors
1. International Smart Irrigation Systems: Expensive and not tailored to
local crops and climate.
2. Local IoT Agriculture Startups: Still in the early stages, offering niche
solutions without comprehensive weather-soil integration.

❖ Key Challenges and Risks

• Farmer Awareness and Adoption: Many farmers may resist adopting new
technology. Awareness campaigns and training will be essential.
• Initial Costs: Even though the platform is affordable, some farmers may find it
costly compared to traditional methods.
• Data Accuracy and Reliability: Ensuring accurate data from weather and soil
APIs is critical.
• Internet Connectivity: Some rural areas may lack consistent internet access,
affecting real-time updates.

2.1.3 Estimated Cost

The estimated cost of the project will depend on several factors, including the development
of the software components such as the mobile application and API integration. A
significant portion of the budget will be allocated to data acquisition costs, particularly for
weather and soil data API subscriptions. Additionally, expenses will include server and
cloud storage services required for data processing and storage. The cost of testing and
deployment is also an essential consideration to ensure the system's reliability and
performance. Finally, ongoing maintenance and updates will be necessary to keep the
system up to date and functioning effectively over time.
 15

2.2 Analysis and Limitation of existing system & Need for the new system
Traditional irrigation systems are often manual, inefficient, and waste significant amounts of
water. Farmers rely on experience rather than real-time data, leading to over-irrigation or
under-irrigation. Current smart irrigation solutions may have limited automation, lack accurate
weather data integration, or be expensive.

And here we can provide you some analysis as a prove the importance of improving to the new
system instead of the traditional and old one:-

1) Crop waste due to inefficient irrigation can range from 10-35%, depending on the
crop type, climate, and the irrigation methods used.
2) Over 80% of water consumption in Egypt is for agriculture, and irrigation
inefficiencies contribute significantly to water wastage. Traditional irrigation methods
like surface or flood irrigation result in wastage of 30-50% of water and worldwide
there is wastage from 20-30%.
3) In Egypt, employing around 20-25% of the total workforce (~5-6 million workers as
of 2024). For irrigation and crop monitoring specifically.

Figure 1 traditional vs. Smart Irrigation Water Usage

Comparison
 16

2.3 Analysis of the new system and Software requirements

2.3.1 User requirements
• Users (farmers) should be able to input land location and crop type.
• The system should :-
- provide real-time weather and soil conditions.
- It should decide optimal irrigation times.
- It should calulate optimal water quantities depending on moisture level which
from it we derive the water level in the soil as following
Soil Moisture Level (%) Irrigation Water Amount (Liters per Square
Needed? Meter)
80% – 100% No 0L
50% – 79% Maybe Small irrigation needed
30% – 49% Yes Medium irrigation needed

And this is a simple graph explain the relation between moisture level and irrigation
amount needed

Figure 2 Soil Moisture vs. Irrigation Need Graph Comparison Figure 3 Difference between good and bad irrigation
 17

2.3.2 System Requirements

• Mobile application with a user-friendly interface.
• Cloud-based storage for real-time data processing.
• Neural Network model for accurate decisions
• API integration for weather and soil monitoring to provide the system with several
data/parameters as following

Parameter Description API type(source)

Temperature Measures air temperature in °C Weather API
Humidity Measures moisture in the air (%) Weather API
Wind Speed Measures speed of wind (m/s or km/h) Weather API
Wind Direction Determines the direction of wind Weather API
Precipitation Checks if it's raining and how much (mm) Weather API
Snowfall Detects if there is snowfall (mm) Weather API
Air Pressure Measures atmospheric pressure (hPa) Weather API
Surface Temperature Temperature of the soil surface (°C) Soil API
Subsurface Temperature Temperature 10 cm below soil surface (°C) Soil API
Soil Moisture Measures water content in soil (%) Soil API
 18

2.3.3 Domain Requirements

• The system must be agriculture-focused.
• It should adapt to different climate conditions to take accurate decisions as
following
Weather Condition Irrigation Decision
Sunny & Hot Increase irrigation frequency
Cloudy & Mild Moderate irrigation
Rainy No irrigation needed
Windy Reduce surface irrigation, use drip systems

• It should support various soil types that will differs in the decisions as following
Soil Type Water Retention Capacity Drainage Irrigation Frequency
Speed
Clay Soil High Slow Less frequent, but deeper watering
Sandy Soil Low Fast More frequent, smaller amounts
Loamy Soil Moderate Balanced Medium frequency, best for farming
 19

• It should support various crop types as following

Crop Type Water Requirement (Clay Water Requirement (Loamy Water Requirement (Sandy
Soil) (mm per season) Soil) (mm per season) Soil) (mm per season)
Wheat 500 mm 550 mm 600 mm
Corn 700 mm 780 mm 850 mm
Tomatoes 600 mm 680 mm 750 mm
Potatoes 550 mm 600 mm 650 mm
Rice 900 mm 1000 mm 1100 mm
Note for more clarify:
Understanding mm in Irrigation
• 1 mm of irrigation = 1 liter of water per square meter (L/m²)
• If a crop needs 500 mm per season, that means it requires 500 liters of water per
square meter over the growing season.
Example for Calculation
If a farmer has 1 hectare (10,000 m²) of wheat that requires 500 mm of water per season:
• Total water needed = 500 mm × 10,000 m²
= 5,000,000 liters (or 5,000 cubic meters)

2.3.4 Functional Requirements

• Weather API integration for real-time data.
• Soil API integration to monitor moisture levels.
• The data should be updated twice daily (12 PM & 12 AM).
• Neural Network model for irrigation decisions
• User input interface for land and crop management.
• Notification system for irrigation alerts.
2.3.5 Non- Functional Requirements
• The system should be highly reliable and scalable.
• The mobile application should be user-friendly and responsive.
• The system should operate with minimal downtime.
 20

CHAPTER 3

PROPOSED DESIGN AND NEXT PLAN

3.1 Dataset Description for APIs

In our Smart Agriculture Platform, we utilize two primary APIs for real-time weather
and soil data acquisition:

1. Open-Meteo API – Provides detailed weather forecasts.

This API provides hourly weather data for the specified location. The following
parameters are collected and analyzed:
Parameter Description Unit Data Type

latitude Latitude of the location Degrees (°) Float

longitude Longitude of the location Degrees (°) Float
temperature_2m Hourly air temperature at 2 meters °C Float
above ground
relative_humidity_2m Hourly relative humidity at 2 meters % Float
above ground
wind_speed_10m Hourly wind speed at 10 meters m/s Float
above ground
precipitation Hourly precipitation amount mm Float
timestamp Time of the forecasted data point ISO 8601 (UTC) String
21

2. AgroMonitoring API – Supplies essential soil information.

The Soil API provides real-time soil data for a given location, which helps monitor
soil moisture and temperature, crucial for irrigation management. Below are the parameters
collected:
Parameter Description Unit Data Type
lat Latitude of the location Degrees (°) Float
lon Longitude of the location Degrees (°) Float
soil_temp Soil temperature at surface level °C Float
soil_moisture Volumetric water content in soil m³/m³ Float
t0 Start time of the requested period Timestamp Integer
t1 End time of the requested period Timestamp Integer
elevation Elevation of the location above sea level Meters (m) Float
 22

3.2 Data Flow Diagrams (DFD)

This diagram is a graphical representation of how data flows within a system. It shows how
data enters, moves through, and exits the system, highlighting the processes, data stores

Figure 4 Data Flow Diagram DFD

 23

3.3 ER Diagram
This diagram is a visual representation of the data model for a system. It illustrates the
relationships between entities (tables or objects) and their attributes, helping to design and
organize the database structure.

Figure 5 ER Diagram
 24

3.4 Use-case diagram

This diagram is a visual representation of how users (actors) interact with a system. It shows
the system's functionality and the different use cases (tasks or actions) that users can perform.
It’s useful for understanding system requirements and user interactions.

Figure 6 Use Case Diagram

 25

3.5 Sequence Diagram

This diagram that illustrates how objects interact in a particular scenario by showing the
sequence of messages exchanged between them.

Figure 7 Sequence Diagram

 26

3.6 Activity diagram

This diagram that represents the workflow or processes within a system. It shows how activities
are performed, the flow of control, and decisions made along the way.

Figure 8 Activity Diagram

 27

3.7 Gantt Chart

It represents a timeline of tasks or activities. It shows the start and end dates of each task, their
durations, and how they overlap, helping track project progress and deadlines.

Figure 9 Gantt Chart

3.8 Next and Future Plan
1. Enhancing Data Accuracy and Real-time Monitoring
• Increase Data Update Frequency: Instead of updating every 12 hours, allow users
to customize the update frequency or integrate real-time monitoring for better
accuracy.
• Use IoT Sensors: Integrate physical IoT sensors for soil moisture, temperature, and
other parameters in addition to API data.
• Satellite & Drone Integration: Use satellite and drones to monitor crop health,
soil conditions, and irrigation efficiency.
2. More Advanced AI and Machine Learning
• AI-based Irrigation Forecasting: Implement machine learning models that predict
future irrigation needs based on weather patterns, soil moisture history, and crop
growth cycles.
• Automated Crop Disease Detection: Train AI models to analyze crop health using
image recognition and suggest how to prevent it.
• Adaptive Irrigation Strategies: Let the system learn over time and suggest better
irrigation strategies based on historical data.
 28

3. Smart Automation and IoT Integration

• Automated Irrigation Control: Connect our system to smart irrigation systems that
can automatically turn water pumps on/off based on the recommendations.
4. Expand Crop and Soil Database
• More Crop-Specific Recommendations: Improve our database by studying a wider
variety of crops, their water requirements.
• Soil Nutrient Analysis: Expand the system to suggest fertilizers or treatments based
on soil health analysis.
5. User Experience & Accessibility
• Multi-language Support: Make the application accessible to farmers from different
regions.
• Voice Commands & Alerts: Implement voice command features for easier access and
send SMS or push notifications for critical updates.
• User Dashboard & Reports: Provide visual analytics and detailed irrigation history
for better decision-making.
6. Weather & Climate Change Adaptation
• Climate Change Prediction: Use long-term climate models to help farmers adjust
irrigation practices based on climate shifts.
• Drought and Flood Warnings: Integrate extreme weather alerts to help farmers take
preventive actions.
29

REFRENCES

Research Papers:

1. "Smart Irrigation Systems in Agriculture: A Systematic Review"

This paper provides a comprehensive review of smart irrigation systems utilizing
artificial intelligence techniques in both urban and rural agriculture. It offers
insights into current technologies and methodologies that can inform your
system's design.

mdpi.com

2. "A Study On Smart Irrigation Systems For Agriculture Using IoT"

This research article explores sensor-based automated irrigation systems,
discussing various sensors and technologies that can be integrated into smart
agriculture platforms.

ijstr.org

3. "Design and Implementation of a Smart Irrigation System"

This paper details the development of a smart irrigation system using embedded
systems and environmental sensors, which could provide practical guidance for
your project's implementation.

questjournals.org

APIs for Weather and Soil Data:

1. Open-Meteo API – Provides detailed weather forecasts.

2. AgroMonitoring API – Supplies essential soil information.

Thank You For Your Reading

See You Next Disscussion