International Journal of Computational and Experimental

Science and ENgineering

(IJCESEN)
Vol. 10-No.4 (2024) pp. 1158-1168
http://www.ijcesen.com
Copyright © IJCESEN ISSN: 2149-9144
Research Article

A Smart Irrigation System Using the IoT and Advanced Machine Learning
Model- A Systematic Literature Review
Ponugoti Kalpana1*, L. Smitha2, Dasari Madhavi 3, Shaik Abdul Nabi 4, G. Kalpana5, Sarangam
Kodati 6
1
Department of Computer Science and Engineering, AVN Institute of Engineering and Technology, Hyderabad,
Telangana, 501510, India.
* Corresponding Author Email: [email protected] - ORCID:0000-0002-4014-8566
2
Department of Information Technology, G Narayanamma Institute of Technology and Science, Hyderabad, Telangana,
India.
Email: [email protected] - ORCID: 0000-0002-6966-7642
3
Department of Computer Science and Engineering (AIML), Sridevi Women's Engineering College, Hyderabad,
Telangana, India,
Email: [email protected] - ORCID: 0000-0003-1083-8096
4
Department of Computer Science and Engineering, AVN Institute of Engineering and Technology, Hyderabad,
Telangana, India 501510,
Email: [email protected] - ORCID: 0000-0003-2755-3097
5
Department of CSE-AI&ML, Malla Reddy Engineering College for Women (Autonomous), Hyderabad, Telangana,
India.
Email: [email protected] - ORCID: 0000-0002-2152-6138
4
Department of Information Technology, CVR College of Engineering, Hyderabad, Telangana, 501510, India.
Email: [email protected] - ORCID: https://orcid.org/0000-0001-9196-3774

Article Info: Abstract:

DOI: 10.22399/ijcesen.526 The rapid advancement of IoT (Internet of Things) technologies and sophisticated
Received : 18 October 2024 machine learning models is driving innovation in irrigation systems, laying the
Accepted : 22 November 2024 foundation for more effective and eco-friendly smart agricultural procedures. This
systematic literature review strives to uncover the advancements and challenges in the
Keywords: advancement and implementation of IoT-based smart irrigation systems integrated with
advanced machine learning techniques. By analyzing 43 relevant studies published
Smart Irrigation Systems, between 2017 and 2024, the research focuses on the ability of these technologies have
IoT in Agriculture, evolved to meet the challenges of modern agriculture irrigation system. Predictive
Machine Learning Models, analytics, anomaly detection, and adaptive control—that enhance irrigation precision
PRISMA methodology. and decision-making processes. Employing the PRISMA methodology, this review
uncovers the strengths and limitations of current systems, highlighting significant
achievements in real-time data utilization and system responsiveness. However, it also
brings attention to unresolved issues, including the complexities of data integration,
network reliability, and the scalability of IoT-based frameworks. Additionally, the study
identifies crucial gaps in standardization and the need for flexible solutions that can
adapt to diverse environmental conditions. By offering a comprehensive analysis, this
review provides key insights for advancing smart irrigation technologies, emphasizing
the importance of continued research in overcoming the existing barriers to wider
adoption and effectiveness in various agricultural settings.

1. Introduction security and sustainable farming practices [1,2].

Traditional irrigation methods, heavily reliant on
The global agricultural sector faces significant inconsistent monsoon rains and manual
challenges in the 21st century, with water scarcity intervention, often result in inefficient water usage
emerging as a critical issue threatening food and suboptimal crop yields [3]. In response to these
 Ponugoti Kalpana, L. Smitha, Dasari Madhavi, Shaik Abdul Nabi, G. Kalpana, Sarangam Kodati/ IJCESEN 10-4(2024)1158-1168

challenges, the integration of IoT technology and optimizing resources, aligning with UN Sustainable
advanced ML models has given rise to smart Development Goal 6. Advances in IoT sensors and
irrigation systems, heralding a new era in precision machine learning, particularly deep learning and
agriculture [4,5]. ensemble methods, have improved the accuracy and
Smart irrigation systems signify a revolution in efficiency of these systems.
agricultural water management, presenting an Cloud computing and big data analytics perform a
advanced method to maximize resource efficiency primary place in the scalability and effectiveness of
and boost crop yield. These systems utilize a smart irrigation approach. These technologies
network of IoT sensors strategically positioned enable the processing and analysis of huge
across fields to gather real-time data on critical quantities of data collected from multiple sources,
factors like soil moisture, temperature, humidity, including historical weather patterns, soil maps, and
and other environmental conditions [6,7]. This crop-specific information. The integration of this
continuous stream of data forms the foundation for diverse data set allows for more comprehensive and
intelligent decision-making in irrigation accurate modeling of irrigation needs among
management [8]. At the heart of these smart diverse crop varieties and geographical positions.
systems lie advanced machine learning algorithms Regardless of the several benefits, tages, the
that assess the collected information to predict widespread adoption of smart irrigation systems
irrigation requirements with remarkable accuracy faces several challenges. The first-time expenditure
[9]. Through the examination of past data alongside for implementation, particularly for minor farmers,
live sensor readings and weather forecasts, these can be a significant barrier. Furthermore, the
ML models can decide when, where, and how requirement for stable online connectivity in remote
much to irrigate effectively [10]. This degree of areas and the technical expertise required to operate
accuracy guarantees that crops get the ideal amount and maintain these systems pose challenges in some
of water at the appropriate time, reducing waste and regions. As the technology continues to mature,
enhancing efficiency [11]. researchers and industry professionals are working
The adoption of smart irrigation systems offers on addressing these challenges. Efforts are being
multifaceted benefits to farmers and the agricultural made to develop more affordable and user-friendly
sector as a whole. Primarily, these systems systems, as well as to provide training and support
significantly reduce water consumption, addressing to farmers transitioning to smart irrigation methods.
the critical issue of water scarcity in many regions Furthermore, ongoing research is focused on
[12]. Smart irrigation systems contribute to water improving the accuracy of ML models, integrating
conservation by supplying it precisely where and additional data sources, and developing more robust
when it's necessary, reducing wastage and ensuring and energy-efficient IoT sensors [16].
efficient use of this valuable resource while This methodical literature review intends to deliver
maintaining or even improving crop health and an extensive summary of the current landscape of
yield [13]. Furthermore, the automation inherent in smart irrigation systems, exploring the various IoT
these systems drastically reduces the labor and ML technologies employed, their benefits,
requirements associated with traditional irrigation challenges, and prospects for further development
methods [14]. Farmers can monitor and control in the field of agriculture. By examining the latest
their irrigation systems remotely through mobile research and Hands-on applications, this report
applications, receiving alerts and recommendations seeks to shed light on the transformative benefits of
based on real-time data analysis. smart irrigation in addressing water deficiencies,
This approach not only conserves time but also increasing agricultural efficiency, and supporting
facilitates more accurate and adaptable oversight of sustainable farming approaches in today’s world.
agricultural resources [15]. The economic
implications of smart irrigation systems are 2. Background and Related Works
significant. Although the upfront expenditure on
IoT sensors and machine learning systems might be 2.1 IoT-based Smart Irrigation Systems
significant, the enduring advantages—such as water Advanced irrigation systems powered by Internet of
conservation, enhanced crop productivity, and Things (IoT) technology have surfaced as an
decreased labor expenses—typically lead to a effective remedy for addressing water shortages in
positive return on investment for agricultural agriculture. These systems employ a range of
producers. Additionally, the optimized use of water sensors to monitor real-time data on soil moisture,
and nutrients can lead to improved crop quality, temperature, and humidity [17]. By integrating IoT
potentially commanding higher market prices. devices with microcontrollers like Arduino or
Smart irrigation systems support sustainable NodeMCU, farmers can remotely monitor and
agriculture by minimizing water waste and control their irrigation systems [18,19]. This system
1159
 Ponugoti Kalpana, L. Smitha, Dasari Madhavi, Shaik Abdul Nabi, G. Kalpana, Sarangam Kodati/ IJCESEN 10-4(2024)1158-1168

not only conserves water but also minimizes labor more precise decision-making [32,33]. Sasikala et
and maximizes crop growth [20,21]. Kumar et al. al. introduced a smart irrigation system that utilizes
proposed an IoT-based smart irrigation system that the Partial Least Squares Regression (PLSR)
uses soil moisture sensors to monitor moisture algorithm to analyze information from diverse
levels in real-time [22]. The system automatically sensors, including humidity, temperature, light
activates a water pump once soil moisture declines intensity, and soil moisture [34]. This approach
beneath a specific limit, ensuring optimal water allows for more nuanced control of irrigation based
supply to crops. Similarly, Sangita Kurundkar et al. on multiple environmental factors. Looking
developed a hybrid method for selecting irrigation towards the future, Reddy et al. discusses the
methods based on climate changes and soil potential of combining logistic regression with IoT
moisture levels, demonstrating the potential of IoT- technology to create a fully automated irrigation
enabled smart irrigation controllers in improving system that considers crop type, weather
agricultural productivity [23]. conditions, and soil moisture to determine watering
needs accurately and cost-effectively [35]. Dhrub et
2.2 Machine Learning in Irrigation Management al. introduced an Adaptive Radial Deep Neural
The integration of machine learning (ML) Network (ARDNN) algorithm for smart farming,
algorithms with IoT-based irrigation systems has which uses IoT and Wireless Sensor Networks
further enhanced the efficiency and accuracy of (WSNs) to monitor plant status and manage water
water management in agriculture. ML models can more accurately [36]. Naveena et al. presented a
assess sensor information and weather forecasts to cloud-based platform that not only monitors soil
identify irrigation requirements and streamline moisture but also incorporates online weather
water usage [24,25]. Bernardo et al. presented a prediction data for more precise watering schedule
smart irrigation system that uses decision tree optimization [37]. Harini et al. covers the
algorithms to predict water requirements based on manufacturing process of a 500 MW turbo
temperature, humidity, and moisture data [26]. The generator stator bar at BHEL Haridwar, vital for
system achieved high accuracy in its predictions, India's electricity production [38]. Also advanced
guiding farmers in effective irrigation management. modeling of IoT and ML-based smart irrigation
In another study, Anitha et al employed an system is proposed to predict and optimize water
Adaptive Neuro-Fuzzy Inference System (ANFIS) usage, improving efficiency and reducing waste
AI algorithm to improve automated irrigation [39]. Where smart irrigation system using AI and
system decision control, reducing computational sensors is designed to monitor and manage water
complexity and enhancing water management [27]. supply, enhancing crop productivity and reducing
costs [40]. Implementing a smart irrigation system
2.3 AI-driven Automated Irrigation Systems with IoT and wireless sensors helps reduce water
Artificial Intelligence (AI) performs a crucial wastage and improve irrigation efficiency through
position in evolving fully automated irrigation real-time monitoring [41]. The integration of solar
systems that can make decisions with minimal power with smart irrigation systems is another
human intervention. These systems not only promising direction. Nanthakumar et al. proposed a
analyze sensor data but also incorporate complex smart solar irrigation system using IoT and Random
decision-making processes to optimize water usage Forest algorithms, which aims to optimize water
[28,29]. Sapaev et al proposed an AI-driven water usage while leveraging renewable energy sources
management system that uses soil moisture sensors, [42]. This approach not only addresses water
pH sensors, light sensors, and weather forecasting scarcity but also promotes sustainable energy use in
data [30]. The system employs a K-Nearest agriculture. Looking towards the future, Krish et al.
Neighbors (KNN) algorithm to achieve a 98.6% discusses the potential of combining AI-driven
recognition rate in predicting irrigation needs. water management systems with comprehensive
Similarly, Pandey et al. developed an intelligent sensor arrays, including buzzers and LCD displays,
irrigation system using deep learning techniques to to provide farmers with real-time feedback and
control the amount of water allocated to different alerts [43]. This holistic approach to irrigation
plant species based on their identification, management promises to further optimize water
demonstrating the potential for highly specialized use, improve crop health, and enhance overall
and efficient irrigation practices [31]. agricultural productivity. Besides these works there
are a number of works done in this fields and
2.4 Advanced Techniques and Future Directions reported in the literature [44-51].
Recent advancements in smart irrigation systems Table 1 presents the summary of different existing
have focused on improving accuracy, reducing techniques handled by the researches. These
costs, and incorporating additional data sources for advancements in IoT, ML, and AI technologies are.
1160
 Ponugoti Kalpana, L. Smitha, Dasari Madhavi, Shaik Abdul Nabi, G. Kalpana, Sarangam Kodati/ IJCESEN 10-4(2024)1158-1168

Table 1 Survey from the Different Authors

Ref Methodology Results Obtained Limitations
[1] IoT-based system with soil moisture sensors and Limited by sensor
Improved automation in irrigation
automated irrigation accuracy
[2] IoT sensors, Node-RED platform, MongoDB, Improved irrigation efficiency Limited to specific
various ML models environmental
conditions
[3] IoT sensors, data analytics Enhanced water conservation Limited by sensor
placement
[4] IoT sensors, OPEN CV and RASPBIAN OS Improved crop monitoring Limited by image
quality
[5] Various ML models (KNN, LR, NN, SVM, NB) Enhanced farming efficiency Limited to specific
dataset and
conditions
[6] IoT sensors, AI algorithms, PLSR model 97.86% prediction accuracy Limited to specific
geographical area
[7] IoT sensors, deep learning for groundwater 90% accuracy in prototype model High Complexity
prediction
[8] IoT sensors, microcontroller, cloud Performs well on collected dataset Limited by hardware
platform,SVM Classifier capabilities
[9] LSTM model is applied to groundwater level Accurate irrigation Complex integration
estimation recommendations with existing systems
[10] Soil Moisture sensor Improved prediction of water needs Limited to specific
environmental
conditions
[11] IoT devices, ML algorithms Enhanced irrigation efficiency Limited by AI model
complexity
[12] Sensors, Big Data and the Cloud in the irrigation Enhanced data-driven irrigation Limited by data
sector processing
capabilities
[13] IoT devices, ML algorithms Enhanced irrigation intelligence Limited by cost
constraints
[14] IoT sensors, ESP WROOM 32 microcontroller Enhanced automation in drip Limited by sensor
irrigation accuracy
[15] IoT sensors, Node-RED platform, MongoDB, KNN best with 98.3% accuracy, Limited to specific
various ML models RMSE of 0.12 environmental
conditions
[16] Sensors for weather forecasting Improved water-use efficiency, Limited to specific
contribution to SDGs systems
[17] Various ML models (KNN, LR, NN, SVM, NB) Random forest achieved 99.98% Limited to specific
accuracy dataset and
conditions
[18] IoT sensors, ESP WROOM 32 microcontroller Real-time monitoring of field Limited by sensor
parameters accuracy
[19] IoT sensors, data analytics Improved sensor modeling Limited by sensor
placement
[20] Node MCU ESP 8266 Wi-Fi Module 80% accuracy in controlled setting Potential issues with
AI decision-making
[21] IoT sensors, ARDNN algorithm Enhanced irrigation prediction Limited by algorithm
complexity
[22] IoT-based system with soil moisture sensors and Reduced water wastage, improved Limited by cloud
automated irrigation efficiency connectivity
[23] IoT sensors, microcontroller, cloud platform Improved water conservation and Limited by solar
crop yield power availability
[24] KNN algorithm, IoT sensors Improved water management by Limited to specific
obtaining 98.4 accuracy rate and geographical area
0.016 root mean squared error
(RMSE).
[25] Node MCU (ESP8266), Soil Moisture sensor Improved crop recommendations Limited by image
(Resistive / Capacitive). quality
[26] Adaptive Neuro-Fuzzy Inference System(AI Improved irrigation control Potential for sensor

1161
 Ponugoti Kalpana, L. Smitha, Dasari Madhavi, Shaik Abdul Nabi, G. Kalpana, Sarangam Kodati/ IJCESEN 10-4(2024)1158-1168

System) failure
[27] Solar panels, IoT sensors, ML algorithms Enhanced energy-efficient irrigation Limited by solar
power availability
[28] IoT sensors, SVM algorithm Reduced water demand by up to Limited by SVM
25% model complexity
[29] light sensors, temperature, humidity sensors, soil Optimized water usage Limited by weather
moisture sensors and KNN algorithm forecast accuracy
[30] Arduino UNO REV3, Arduino IDE, Improved water flow measurement High energy
Temperature/Moisture Sensor consumption
[31] ESP32, ML algorithms Cost-effective irrigation planning Limited by cost
constraints
[32] IoT sensors, AI algorithms, PLSR model Improved efficiency and crop yield Limited by specific
crop types
[33] IoT devices, logistic regression Improved prediction of water needs Limited to specific
crop types
[34] IoT sensors, ARDNN algorithm Enhanced irrigation prediction Limited by algorithm
complexity
[35] Moisture Sensor, DHT11 Sensor, NodeMCU. Improved sustainability in irrigation Limited by cloud
connectivity
[36] Hydro, and turbo generators Enhanced irrigation efficiency Limited by prototype
constraints
[37] NODEMCU ESP8266, SOIL SENSOR, DHT11 Enhanced monitoring capabilities Limited by cloud
SENSOR,Ensemble Mosel(Random Forest, connectivity
Gradient Boosting, and Support Vector
Regression)
[38] Arduino Uno controller,DHT11 sensor,Blynk Improved irrigation control Limited by AI model
(Android) app complexity
[39] IoT sensors, ESP8266NODEMCU Improved soil moisture monitoring Limited by sensor
accuracy
[40] Solar power, IoT sensors, ML algorithms Improved energy-efficient irrigation Limited by solar
power availability
[41] IoT sensors, Blynk cloud platform Improved automated irrigation Limited by cloud
control connectivity
[42] Arduino Uno, Wi-Fi Module, Relay Module Improved water management Potential for sensor
errors
[43] IoT sensors, ESP WROOM 32 microcontroller Enhanced real-time monitoring Limited by
microcontroller
capabilities

paving the way for more efficient, sustainable, and 3.1 Research Questions
precise irrigation practices in agriculture, This review addresses several research questions to
addressing the critical challenge of water scarcity direct the analysis of the chosen literature:
while improving crop yields and reducing  RQ-1: What are the primary factors influencing
operational costs. the adoption of IoT in smart irrigation systems?
 RQ-2: How do advanced machine learning
3. Systematic Literature Review models enhance the precision and efficacy of
agricultural irrigation?
Conducting a systematic literature review (SLR) is  RQ-3: What challenges arise from integrating
crucial for compiling and analyzing existing IoT and machine learning for real-time irrigation
research, offering a detailed perspective on management, and what solutions are emerging?
advancements, challenges, and future directions in  RQ-4: How do existing smart irrigation
a specific field. This review focuses on the frameworks compare in terms of water
intersection of IoT and advanced machine learning conservation, scalability, and ease of use?
models in smart irrigation systems, aiming to map  RQ-5: What best practices are recommended for
the current landscape, identify key challenges, and combining IoT and machine learning to enhance
uncover opportunities for innovation in enhancing irrigation efficiency and sustainability?
agricultural efficiency and sustainability.

1162
 Ponugoti Kalpana, L. Smitha, Dasari Madhavi, Shaik Abdul Nabi, G. Kalpana, Sarangam Kodati/ IJCESEN 10-4(2024)1158-1168

3.2 Data Sources and Search Strategy 3.4 PRISMA Methodology

The review process involved comprehensive The PRISMA framework was used to assure a
searches in major academic databases such as methodical and transparent method of review. The
Research Gate, IEEE Xplore, Springer, and MDPI, PRISMA flowchart which is portrayed in Figure 2
concentrating on studies published between 2017 provides a detailed visualization of each stage in
and 2024. The arrangement of the paper is depicted the review, from the initial search and screening of
in Figure 1. In this systematic literature review, a studies to their final selection for in-depth analysis.
comprehensive search strategy was employed to Search Results and Study Selection
capture relevant studies on IoT-based smart The search process yielded a total of 702 articles
irrigation systems and advanced machine learning across the selected databases. After removing
techniques. Key terms such as "IoT-driven redundancies and applying the criteria for inclusion
irrigation systems," "machine learning in and exclusion, 106 articles earmarked for detailed
agriculture," "precision irrigation technologies," analysis. Following a detailed evaluation of
and "sensor-based smart farming" were used. abstracts and full texts, 43 articles were found to be
Boolean operators (AND, OR) refined the search to relevant and included in the final analysis.
focus on studies combining these technologies. The
review prioritized high-quality sources,
incorporating scholarly journal articles, conference
papers, and research reports. The selection process Search Analysis
began with an abstract review to assess relevance, 25
followed by a full-text analysis to ensure alignment
No of Papers 20
with the research objectives. To ensure that recent
developments were included, citation analysis and 15
manual database searches were also conducted, 10
locating supplementary studies overlooked in the 5
primary search. This exhaustive technique assured a
0
full in-depth comprehension of the latest
Research IEEE Springer MDPI
innovations and challenges in smart irrigation Gate
systems utilizing IoT and machine learning.

3.3 Inclusion and Exclusion Criteria Figure 1: Search analysis from different sources
To guarantee that the review includes high-quality
and pertinent studies, the following criteria were
applied as follows in Table 2.
Table 2 Inclusion and Exclusion criteria
Criteria Included Excluded
Publication Date Studies published between 2017 and 2024 Studies published before 2017 or after
2024
Type of Peer-reviewed journal articles, conference papers, Non-peer-reviewed sources, opinion
Publication and technical reports pieces, and non-technical reports
Focus Area Research focused on IoT, machine learning, and Research not focused on IoT, machine
smart irrigation systems learning, or agriculture
Research Empirical studies, case studies, systematic reviews Theoretical papers without empirical
Methodology validation
Language Publications in English Non-English publications
Geographical Studies with significant global or regional impact Studies with highly localized issues and
Scope limited general relevance
Innovation and Studies presenting novel approaches, major Studies that replicate existing methods
Impact advancements, or impactful findings without new insights
Technological Research involving current or emerging Research focused on outdated
Relevance technologies in IoT and machine learning technologies or methods
Application Scope Studies with practical applications in smart Research with theoretical or non-practical
irrigation and agriculture applications
Data Type Research utilizing diverse data sources, including Studies based on limited or non-diverse
real-time and historical data data sources
System Integration Studies exploring integration of IoT and machine Research that does not address the
learning in smart irrigation systems integration of these technologies

1063
 Ponugoti Kalpana, L. Smitha, Dasari Madhavi, Shaik Abdul Nabi, G. Kalpana, Sarangam Kodati/ IJCESEN 10-4(2024)1158-1168

 Scalability Issues: Scalability continues to be a

critical issue, particularly when dealing with
extensive datasets across diverse environmental
conditions. Leveraging cloud computing and big
data analytics appears essential to overcoming
these limitations.
 Implementation Complexity: While advanced
smart irrigation systems offer significant
benefits, the complexity of integrating IoT and
machine learning technologies can be a barrier,
especially for small-scale farmers.

3.2 Key Insights and Observations

The analysis of the 43 chosen articles offers an in-
depth insight into the present condition of smart
irrigation systems utilizing IoT and sophisticated
machine learning models. It included:
 Real-Time Data Utilization: IoT sensors
provide continuous data on soil moisture,
weather conditions, and crop health, which
Figure 2 PRISMA Flowchart machine learning models use for precise
irrigation management.
3.5 Overview of Selected Studies  Enhanced Efficiency: Machine learning
The selected studies encompass various aspects of techniques optimize water usage by predicting
IoT and machine learning applications in smart irrigation needs more accurately, leading to cost
irrigation systems. Key observations include: reductions and improved resource management.
 IoT Technology Advancements: Recent  Increased Crop Yields: Automated irrigation
studies demonstrate significant progress in based on data-driven insights results in better
developing IoT sensor networks tailored for crop health and higher productivity.
irrigation, focusing on real-time data collection
and the ability to remotely monitor agricultural 4. Findings
fields.
 Machine Learning Integration: The This analysis aims to address the research questions
application of advanced machine learning (RQs) determined for this systematic literature
techniques, like predictive modeling and review. Data extraction was performed on the
adaptive control, has been shown to improve selected research articles (n=43), and the results are
irrigation scheduling and optimize water usage, discussed with respect to the study's RQs. Also, a
leading to better resource management. gap analysis is provided to highlight areas requiring
 Challenges and Emerging Solutions: While
further research and development.
integrating IoT with machine learning presents
challenges, such as ensuring reliable network
4.1 Solutions to RQs:
performance, managing data integration, and
RQ-1: What are the primary factors influencing
achieving scalability, emerging technologies like
the adoption of IoT in smart irrigation systems?
cloud computing and big data analytics are
The adoption of IoT in smart irrigation systems is
offering promising solutions.
primarily influenced by several key factors. Firstly,
the potential for significant water conservation and
3.6 Comparative Analysis improved crop yields has been a major driver. Tace
A comparative analysis of the methodologies, et al. (2022) demonstrated that IoT-based systems
system architectures, and performance metrics can reduce water usage by up to 30% while
across the selected studies reveals the following maintaining or improving crop health. Secondly,
insights: the decreasing cost of IoT sensors and the
 Water Efficiency: IoT-enabled systems have
increasing availability of reliable wireless
proven effective in improving water efficiency connectivity in rural areas have made these systems
by offering precise irrigation based on real-time more accessible to farmers. Additionally, the
environmental data. However, optimizing sensor integration of IoT with existing agricultural
networks for large-scale use remains a practices has become smoother, reducing the barrier
challenge.
1064
 Ponugoti Kalpana, L. Smitha, Dasari Madhavi, Shaik Abdul Nabi, G. Kalpana, Sarangam Kodati/ IJCESEN 10-4(2024)1158-1168

to adoption. However, concerns about data security limiting their ease of use for farmers without
and privacy, as well as the initial investment technical expertise. In terms of scalability, cloud-
required for implementation, remain significant based solutions have demonstrated better
challenges for some farmers. performance in managing large-scale operations,
RQ-2: How do advanced machine learning but they may require significant infrastructure
models enhance the precision and efficacy of investments. On the other hand, simpler, more
agricultural irrigation? localized systems may be easier to implement and
These models can analyze complex datasets from use but might not offer the same level of water
various sources, including soil moisture sensors, conservation or scalability. The trade-offs between
weather forecasts, and historical crop data, to these factors often depend on the specific needs and
optimize irrigation schedules. For instance, Ramya resources of the agricultural operation.
et al. (2020) developed a machine learning RQ-5: What best practices are recommended for
framework that could predict soil moisture levels combining IoT and machine learning to enhance
with 95% accuracy, allowing for precise irrigation irrigation efficiency and sustainability?
timing. Furthermore, machine learning algorithms Firstly, it's crucial to establish a comprehensive
can adapt to changing environmental conditions sensor network that captures a wide range of
and learn from past irrigation cycles, continuously relevant data, including soil moisture, weather
improving their performance over time. This conditions, and crop health indicators. Secondly,
adaptability is crucial in the face of climate change the integration of multiple data sources, including
and variable weather patterns, ensuring that historical data and weather forecasts, can
irrigation remains efficient even as conditions significantly improve the accuracy of irrigation
evolve. predictions. Thirdly, implementing a user-friendly
RQ-3: What challenges arise from integrating interface that provides clear, actionable insights to
IoT and machine learning for real-time farmers is essential for widespread adoption and
irrigation management, and what solutions are effective use of the system. Additionally, regular
emerging? system calibration and maintenance are vital to
The integration of IoT and machine learning for ensure ongoing accuracy and reliability. Finally,
real-time irrigation management faces several incorporating feedback mechanisms that allow the
challenges. One primary issue is the complexity of system to learn from actual irrigation outcomes and
data integration from diverse sources, which can farmer interventions can lead to continuous
lead to inconsistencies and errors in decision- improvement in performance over time.
making. Another challenge is the need for robust
and reliable connectivity in often remote 4.2 Gap Analysis
agricultural areas, where network infrastructure Despite the significant advancements in IoT and
may be limited. Additionally, the scalability of machine learning for smart irrigation, several gaps
these systems to large agricultural operations can be remain in the current research and implementation:
problematic, as it requires managing vast sensor  Limited studies on long-term sustainability and
networks and processing large volumes of data in economic viability of these systems, especially
real-time. To address these challenges, emerging for small-scale farmers.
solutions include edge computing technologies that  Insufficient research on the integration of smart
allow for local data processing, reducing reliance irrigation systems with other aspects of
on constant internet connectivity. Cloud-based precision agriculture, such as pest management
platforms are also being developed to handle large- and fertilization.
scale data integration and analysis more efficiently.  Lack of standardized protocols and frameworks
Furthermore, efforts are being made to standardize for IoT in agriculture, hindering interoperability
IoT protocols in agriculture, which would facilitate and widespread adoption.
better interoperability between different systems  Limited exploration of the potential
and sensors. environmental impacts of large-scale sensor
RQ-4: How do existing smart irrigation deployment in agricultural settings.
frameworks compare in terms of water  Insufficient attention to the social and cultural
conservation, scalability, and ease of use? factors affecting the adoption of smart irrigation
Some systems, like those employing advanced technologies in different regions.
predictive analytics, have shown remarkable water
conservation capabilities, with reports of up to 50% The review highlights difficulties such as complex
reduction in water usage compared to traditional data integration, high implementation costs, and the
methods. However, these highly efficient systems need for robust data security. To handle these
often come with increased complexity, potentially
1065
 Ponugoti Kalpana, L. Smitha, Dasari Madhavi, Shaik Abdul Nabi, G. Kalpana, Sarangam Kodati/ IJCESEN 10-4(2024)1158-1168

problems, forthcoming inquiries should concentrate System. International Journal of Applied

on advancing model precision, reducing technology Engineering Research 12(18);7306-7320.
costs, and integrating emerging technologies like [2]Stephen a, & Anitha, Dr & Lawrence, Dr. L.
edge computing. Arockiam. (2019). A Hybrid Method for Smart
Irrigation System. International Journal of Recent
Technology and Engineering. 8(3);2995. DOI:
5. Conclusions 10.35940/ijrte.C4826.098319.
[3]Rishabh Modi, Madhavan P, Karan Veer Mahajan
This review highlights the transformative role of (2019). Smart Irrigation System. International
IoT and advanced machine learning in smart Journal of Engineering and Advanced Technology
irrigation systems. The incorporation of IoT sensors (IJEAT), 8(4);411.
with machine learning algorithms has led to [4]V R Balaji. Smart Irrigation System using Iot and
Image Processing. International Journal of
significant improvements in water management and Engineering and Advanced Technology (IJEAT)
crop productivity. These advancements offer real- 8(6S);115 DOI:10.35940/ijeat.F1024.0886S19
time data collection and sophisticated analysis, [5]Tefera, H. A. ., Dongjun, H., & Njagi, K. . (2020).
resulting in precise water usage, optimized Implementation of IoT and Machine Learning for
irrigation schedules, and reduced operational costs. Smart Farming Monitoring System. International
Despite these advancements, challenges such as Journal of Sciences: Basic and Applied Research
integrating data from diverse sources, managing (IJSBAR), 52(1);67–77.
high implementation costs, and addressing data [6]Blasi, Anas & Abbadi, Mohammad & Al-Huweimel,
security concerns remain significant. Future Rufaydah. (2020). Machine Learning Approach for
research should focus on refining machine learning an Automatic Irrigation System in Southern Jordan
Valley.
models for better accuracy, reducing system costs,
[7] Ramya, S., Swetha, A. & Doraipandian, M. (2020).
and exploring innovative solutions like edge IoT Framework for Smart Irrigation using Machine
computing to enhance system scalability and Learning Technique. Journal of Computer Science,
reliability. By overcoming these challenges, IoT 16(3); 355-363.
and machine learning can continue to drive the https://doi.org/10.3844/jcssp.2020.355.363
transformation of smart irrigation, contributing to [8] Bhoi, Ashutosh & Nayak, Rajendra & Bhoi, Sourav
more sustainable and efficient agricultural & Sethi, Srinivas & Panda, Sanjaya & Sahoo, Kshira
practices. & Nayyar, Anand. (2021). IoT-IIRS: Internet of
Things based intelligent-irrigation recommendation
system using machine learning approach for efficient
Author Statements: water usage. PeerJ Computer Science. 7. e578. DOI:
10.7717/peerj-cs.578.
 Ethical approval: The conducted research is [9] Sleem, Ahmed & El-henawy, Ibrahim. (2021). Smart
not related to either human or animal use. Irrigation System with Predictive Analytics using
 Conflict of interest: The authors declare that Machine Learning and IoT. Journal of Intelligent
they have no known competing financial Systems and Internet of Things. 2;77-83. DOI:
10.54216/JISIoT.020204.
interests or personal relationships that could
[10]Dr. Hetal Patela , Dr. Shailesh Khantb , Dr. Atul
have appeared to influence the work reported in Patel (2021).Artificial Intelligence and IoT Based
this paper Smart Irrigation System for Precision Farming.
 Acknowledgement: The authors declare that Turkish Journal of Computer and Mathematics
they have nobody or no-company to Education (TURCOMAT), 12(10);4462-4467, DOI:
acknowledge. 10.17762/turcomat.v12i10.5184
 Author contributions: The authors declare that [11]J. Angelin Blessy and A. kumar (2021). Smart
they have equal right on this paper. Irrigation System Techniques using Artificial
Intelligence and IoT, Third International Conference
 Funding information: The authors declare that
on Intelligent Communication Technologies and
there is no funding to be acknowledged. Virtual Mobile Networks (ICICV), Tirunelveli, India,
 Data availability statement: The data that pp. 1355-1359, doi:
support the findings of this study are available 10.1109/ICICV50876.2021.9388444.
on request from the corresponding author. The [12]Burhanuddin Badrun and Murshal Manaf (2021).
data are not publicly available due to privacy or The Development of Smart Irrigation System With
ethical restrictions. IoT, Cloud, and Big Data. 2021 IOP Conf. Ser.:
Earth Environ. Sci. 830;012009. DOI 10.1088/1755-
References 1315/830/1/012009.
[13]Shrikant Kokate , Durga Late , Mrunali Padwal ,
[1]Yuthika Shekhar, Ekta Dagur, Sourabh Mishra Vivek Khatri and Karthi Dhanasekaran (2021).
(2017). Intelligent IoT Based Automated Irrigation Intelligent Irrigation System Based on ML and IoT.
International Journal of Engineering and
1066
 Ponugoti Kalpana, L. Smitha, Dasari Madhavi, Shaik Abdul Nabi, G. Kalpana, Sarangam Kodati/ IJCESEN 10-4(2024)1158-1168

Management Research 11(6); [25]Aditya Gor , Kandarp Joshi , Rushil Togadiya , Dr.
https://doi.org/10.31033/ijemr.11.6.6 Warish Patel, (2023), Automation in Irrigation using
[14]Jeeni Patel, Ashutosh Dubey, Ashutosh Singh, Bibin IoT and ML based Crop Recommendation System,
Mathew (2021). Automatic Drip Irrigation using IoT International journal of engineering research &
and Machine Learning. International Journal for technology (IJERT) 12(03);171-76 DOI:
Research in Engineering Application & Management 10.17577/IJERTV12IS030112
(IJREAM) 07(2); DOI: 10.35291/2454- [26]Bernardo, Myrtel & Evangelista, Jermyn &
9150.2021.0222 Gatchalian, Shiela Marie & Tejada, Rael. (2023).
[15]Youness Tace, Mohamed Tabaa, Sanaa Elfilali, Development Of Artificial Intelligence Algorithm
Cherkaoui Leghris, Hassna Bensag, Eric Renault, For Smart Irrigation Using Internet Of Things (IOT)
(2022). Smart irrigation system based on IoT and A BASC Automated System for Improved
machine learning, Energy Reports, 8(9);1025-1036, Agricultural Irrigation. 1; 24-36. DOI:
https://doi.org/10.1016/j.egyr.2022.07.088. 10.52631/jemds.v1i1.77.
[16]Khaled Obaideen, Bashria A.A. Yousef, Maryam [27]B., Anitha & Pandi, Jeyakani & V., Mahalakshmi &
Nooman AlMallahi, Yong Chai Tan, Montaser Shalini, S. & R., Senthil. (2023). Design and
Mahmoud, Hadi Jaber, Mohamad Ramadan (2022). Implementation of a Smart Solar Irrigation System
An overview of smart irrigation systems using IoT, Using IoT and Machine Learning. E3S Web of
Energy Nexus, 7;100124, Conferences. 387. DOI:
https://doi.org/10.1016/j.nexus.2022.100124. 10.1051/e3sconf/202338705012.
[17]Iorliam, Aamo & Bum, Sylvester & [28]Guttikonda, Kranthi & Bangare, Manoj & Bangare,
AONDOAKAA, Iember & Iorliam, Iveren & Shehu, Pushpa & Kumar, Chanda & Raj, Roop & Arias-
Yahaya. (2022). Machine Learning Techniques for Gonzáles, José & Omarov, Batyrkhan & Mia, Md.
the Classification of IoT-Enabled Smart Irrigation (2023). Internet of Things Sensors and Support
Data for Agricultural Purposes. Gazi University Vector machine integrated intelligent irrigation
Journal of Science Part A: Engineering and system for agriculture industry. Res Sq DOI:
Innovation. 9; 378-391. DOI: 10.21203/rs.3.rs-3193954/v1.
10.54287/gujsa.1141575. [29]Abo-Zahhad, M. (2023). IoT-Based Automated
[18]Mantesh H Dollin , Meghana Nekar, Meghana , Management Irrigation System Using Soil Moisture
Ashwitha Thomas (2022). IoT Based Monitoring Data and Weather Forecasting Adopting Machine
System in Smart Agriculture. International Journal Learning Technique. Sohag Engineering Journal,
of Advanced Research in Science, Communication 3(2);122-140. doi: 10.21608/sej.2023.209528.1037
and Technology (IJARSCT) 2(2);349. DOI: [30]J. Sapaev, A. Arifjanov , Kh. Ramazonov , and I.B.
10.48175/IJARSCT-2872 Sapaev (2023). Smart technologies for determining
[19] Sami M, Khan SQ, Khurram M, Farooq MU, water flow in irrigation systems. E3S Web of
Anjum R, Aziz S, Qureshi R, Sadak F (2022). A Conferences 383, 02012
Deep Learning-Based Sensor Modeling for Smart [31]Pandey, Prabhat and Agarwal, Sudhir (2023), A Low
Irrigation System. Agronomy. 12(1);212. Cost Smart Irrigation Planning Based on Machine
https://doi.org/10.3390/agronomy12010212 Learning and Internet of Things. Available at
[20]N. Rahul, S. Sumathi, S. Rajaprabu, J. Prawin SSRN: https://ssrn.com/abstract=4414709 or http://d
Kumar, R.Varthish (2022). IoT-Based smart x.doi.org/10.2139/ssrn.4414709
irrigation system using artificial intelligence. [32]Anand Tilagul , Arunkumar, Darshan, Gajendra,
International Journal of Creative Research Thoughts Likith Gowda (2024). Automated Irrigation System
(IJCRT). 7(12); Using Artificial Intelligence (AI). International
[21] Sukhdev Singh, Arvind Kumar Bhardwaj (2022). Journal for Innovative Research in Science and
An efficient machine learning inspired smart Technology. 13(5); DOI:
irrigation system for agriculture. NeuroQuantology 10.15680/IJIRSET.2024.1305102.
20(13);4366-4372 doi: [33]Dr. Rachna, K. Somkunwar, Padma Dev Mishra ().
10.48047/nq.2022.20.13.NQ88531 Logistic Regression and Internet of Things Based
[22]Kumar, G. & Nagaraju, G. & Rohith, D. & Smart Irrigation to Predict Crops Water Need.
Vasudevarao, A.. (2023). Design and Development International Journal on Recent and Innovation
of Smart Irrigation System Using Internet of Things Trends in Computing and Communication 11(10)
(IoT) - A Case Study. Nature Environment and https://doi.org/10.17762/ijritcc.v11i10s.7604
Pollution Technology. 22;523-526. DOI: [34]Sasikala S, Sita Devi Bharatula. (2024). IoT-Based
10.46488/NEPT.2023.v22i01.052. Smart Irrigation System Based Adaptive Radial Deep
[23]Sangita Kurundkar, Vinod Panzade, Sachi Nagdeve, Neural Network (ARDNN) Algorithm Applicable for
Mufaddal Habibi, Mohini Mane (2023). IoT Based Various Agricultural Production. International
Smart Irrigation System. Journal of Engineering Journal of Intelligent Systems and Applications in
Sciences. 14(04). Engineering, 12(13s);351 –. Retrieved from
[24]Tace Y, Elfilali, Tabaa, Leghris (2023). https://ijisae.org/index.php/IJISAE/article/view/4602
Implementation of smart irrigation using IoT and [35]Reddy, V. & Harivardhagini, S. & Sreelakshmi, G..
Artificial Intelligence. Mathematical modeling and (2024). IoT and Cloud Based Sustainable Smart
computing, 10(2);575–582 DOI: Irrigation System. E3S Web of Conferences. 472.
10.23939/mmc2023.02.575 10.1051/e3sconf/202447201026.
1067
 Ponugoti Kalpana, L. Smitha, Dasari Madhavi, Shaik Abdul Nabi, G. Kalpana, Sarangam Kodati/ IJCESEN 10-4(2024)1158-1168

[36]Dhrub Shaw, Anuraag Banerjee, Shashank Saurav, Physiological Data Transmission in Fog-BAN
Priya Jha, Dr. Sangeeta Roy (2024). Smart system for Environment. International Journal of
efficient irrigation methods using IoT. International Computational and Experimental Science and
Research Journal of Modernization in Engineering Engineering, 10(4);671-681.
Technology and Science 06(06). https://doi.org/10.22399/ijcesen.490
https://www.doi.org/10.56726/IRJMETS58949 [48]M, P., B, J., B, B., G, S., & S, P. (2024). Energy-
[37]M. Naveena, B. Srinivas Raja, A.S.P. Gowtham, S. efficient and location-aware IoT and WSN-based
Umar Farooq, M. Vishnu Vardhan (2024). IoT and precision agricultural frameworks. International
machine learning approach for smart irrigation Journal of Computational and Experimental Science
monitoring system. International Journal of Creative and Engineering, 10(4);585-591.
Research Thoughts (IJCRT), 12(3) https://doi.org/10.22399/ijcesen.480
[38]Harini S, Shobana A, Sapna S, Vedhashree R, [49]S, P. S., N. R., W. B., R, R. K., & S, K. (2024).
Prabhakaran M (2024). Smart irrigation powered by Performance Evaluation of Predicting IoT Malicious
artificial intelligence (AI). International Journal of Nodes Using Machine Learning Classification
Creative Research Thoughts (IJCRT), 12(4). Algorithms. International Journal of Computational
[39]Tushar V. Dhurjad , Ankita S. Bhadane , Shruti N. and Experimental Science and Engineering,
Borse , Rohit S. Dhurjad (2024). IoT based Smart 10(3);341-349. https://doi.org/10.22399/ijcesen.395
Irrigation System using soil Moisture sensor and [50]Achuthankutty, S., M, P., K, D., P, K., & R,
ESP8266NODEMCU. International Journal of Novel prathipa. (2024). Deep Learning Empowered Water
Research and Development (IJNRD). 9(2). Quality Assessment: Leveraging IoT Sensor Data
[40]Dasarinki Ramprasad , Dhoni Sai Kumar , Gujju with LSTM Models and Interpretability Techniques.
Siva Reddy , Bevara Ravi Teja , Dunna Varshitha Sri International Journal of Computational and
(2024). Solar-Powered Smart Irrigation System using Experimental Science and Engineering, 10(4);731-
Machine Learning & IoT. International Journal of 743. https://doi.org/10.22399/ijcesen.512
Research Publication and Reviews, 5(3);7610-7616 [51]Alkhatib, A., Albdor , L., Fayyad, S., & Ali, H.
https://doi.org/10.55248/gengpi.5.0324.0926 (2024). Blockchain-Enhanced Multi-Factor
[41]Dr. K. Ramesh Babu, Kola Blessy, Velpula Venkata Authentication for Securing IoT Children’s Toys:
Kasi Eswar Reddy, Nallapalli Khadar Basha, Dadi Securing IoT Children’s Toys. International Journal
Sravanthi, Pitti Chinni (2024). Implementation of of Computational and Experimental Science and
smart irrigation system using IoT. JETIR 11(5). Engineering, 10(4);1041-1049.
[42]Dr. G. Nanthakumar , Mr. K. Pazhanivel , Abinesh https://doi.org/10.22399/ijcesen.417
A , Ganesh K. R, Harish V , Nagendran S (2024).
IoT Based Smart Irrigation System using Artificial
Intelligence. International Journal of Advanced
Research in Science, Communication and
Technology (IJARSCT) , 4(6)
[43]Krish R. Mehta, K. Jayant Naidu, Madhav Baheti,
Dev Parmar and A. Sharmila (2023). Internet of
Things Based Smart Irrigation System Using ESP
WROOM 32. Journal on Internet of Things 5;45-
55. https://doi.org/10.32604/jiot.2023.043102.
[44]Radhi, M., & Tahseen, I. (2024). An Enhancement
for Wireless Body Area Network Using Adaptive
Algorithms. International Journal of Computational
and Experimental Science and Engineering,
10(3);388-396. https://doi.org/10.22399/ijcesen.409
[45]Nagalapuram, J., & S. Samundeeswari. (2024).
Genetic-Based Neural Network for Enhanced Soil
Texture Analysis: Integrating Soil Sensor Data for
Optimized Agricultural Management. International
Journal of Computational and Experimental Science
and Engineering, 10(4);962-970.
https://doi.org/10.22399/ijcesen.572
[46]D, jayasutha. (2024). Remote Monitoring and Early
Detection of Labor Progress Using IoT-Enabled
Smart Health Systems for Rural Healthcare
Accessibility. International Journal of
Computational and Experimental Science and
Engineering, 10(4);1149-1157.
https://doi.org/10.22399/ijcesen.672
[47]S, P., & A, P. (2024). Secured Fog-Body-Torrent : A
Hybrid Symmetric Cryptography with Multi-layer
Feed Forward Networks Tuned Chaotic Maps for
1068