Smart Agricultural Technology 11 (2025) 100979

Contents lists available at ScienceDirect

Smart Agricultural Technology

journal homepage: [Link]/smart-agricultural-technology

Fuzzy logic-based IoT system for optimizing irrigation with cloud

computing: Enhancing water sustainability in smart agriculture
Abdennabi Morchid a , Zafar Said b,* , Almoataz Y. Abdelaziz c,d , Pierluigi Siano e,f,
Hassan Qjidaa g
a
Department of Physics, Laboratory of Informatics, Signals, Automation and Cognitivism (LISAC), Faculty of Sciences Dhar El Mahraz, Sidi Mohamed Ben Abdellah
University, BP 1796 Fes-Atlas 30003 Fez, Morocco
b
Mechanical and Aerospace Engineering Department, College of Engineering, United Arab Emirates University, Al Ain, 15551, United Arab Emirates
c
Faculty of Engineering, Ain Shams University, Cairo 11517, Egypt.
d
Faculty of Engineering and Technology, Future University in Egypt, Cairo 11835, Egypt.
e
Department of Management & Innovation Systems, University of Salerno, Fisciano 84084, Italy.
f
Department of Electrical and Electronic Engineering Science, University of Johannesburg, Johannesburg 2006, South Africa.
g
Systems and Sustainable Environment Laboratory (SED), Faculty of Engineering Sciences (FSI) Private University of Fez (UPF), Fez, Morocco

A R T I C L E I N F O A B S T R A C T

Keywords: Faced with the growing challenges of water scarcity and the effects of climate change, intelligent irrigation
Cloud computing management is essential to guarantee food security and promote sustainable agriculture. Traditional irrigation
Embedded systems methods are often inefficient, waste water and expose crops to water stress. This paper proposes an innovative
IoT
irrigation system integrating the Internet of Things (IoT), embedded systems, fuzzy logic and cloud computing
Fuzzy logic
Sensor
for optimized water management. The system is based on three layers: IoT devices in the field comprising ESP32
Smart irrigation microcontrollers, temperature and humidity sensors and actuators; a cloud layer using ThingSpeak for real-time
Smart agriculture data collection and analysis via HTTP protocol; and an interactive dashboard for irrigation monitoring and
Sustainable control. Mathematical modeling plays a key role, integrating fuzzy logic to dynamically adjust irrigation ac­
Water security cording to environmental conditions. MATLAB is used to simulate and visualize fuzzy membership functions,
enabling optimized decision-making. Results show that the system reduces water losses by adjusting watering
periods according to soil temperature and humidity. Thanks to ThingSpeak, it effectively maintains soil humidity
close to the target threshold of 62 %, ensuring more precise, weather-responsive irrigation. Evaluation over an
extended period shows a significant reduction in water wastage thanks to strategic management based on well-
defined rules. What’s more, with a cost of just 32 USD, this IoT system proves more affordable than conventional
methods requiring significant investment in infrastructure and maintenance. This intelligent irrigation system
represents a promising solution for sustainable water management in agriculture. By integrating real-time soil
and climate data, it improves water efficiency, optimizes yields and minimizes the risk of water stress, thus
contributing to the resilience of agricultural practices in the face of current and future environmental challenges.

1. Introduction environment [2]. Smart agriculture involves the introduction of

advanced technologies like IoT, AI, and sensors into novel agricultural
Agriculture is central to economic and social development, providing practices [3–5]. These developments enable better and more efficient
much-needed supplies of food for an ever-increasing world population management of process inputs such as water, one of the key variables in
[1]. However, farming nowadays faces unprecedented challenges in its crops. Smart agriculture brings automated irrigation with soil moisture
path since it will be grossly affected by the impacts of climate change, sensors, which worsens due to water scarcity and climate change, which
inefficient management of water resources, and the need to sustain may optimize water use by reducing its waste. Other important chal­
long-term operability. Smart agriculture emerges as a promising solu­ lenges to global agriculture include changes in climate, which have
tion for these problems of food security while still preserving the impacts on crop productivity, water availability, and soil health [6].

* Corresponding author.
E-mail address: zafar@[Link] (Z. Said).

[Link]
Received 28 November 2024; Received in revised form 5 March 2025; Accepted 29 April 2025
Available online 29 April 2025
2772-3755/© 2025 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC license ([Link]
nc/4.0/).
A. Morchid et al. Smart Agricultural Technology 11 (2025) 100979

Smart agriculture provides tools for mitigating the effects of such chal­ power the IoT node. A rectenna is manufactured and validated,
lenges by adapting agricultural practices to changing climatic conditions demonstrating satisfactory performance in an outdoor environment.
and promoting greener methods of production. For example, predictive Article [12] proposes a similar solution by integrating the IoT for fully
models can be used to gauge the risks posed by climate change and automated pump management. Further, the article [13] describes a
adjust planting schedules accordingly to make crops more resilient real-time precision irrigation system, combining a fuzzy logic controller
against climatic hazards. Besides, sustainability in agriculture is with LoRa protocol, allowing advanced optimization of irrigation times
described as a reduction of environmental impacts while sustaining and efficient management of agricultural resources via an IoT platform.
adequate food production to meet the needs of the present and future On the other hand, this research [14] proposed a fuzzy computational
[7]. Smart agriculture executes this process by paving the way for analysis for IoT-based intelligent irrigation systems, integrating fuzzy
environmentally appropriate agricultural practices that include preci­ logic for decision analysis and secure data collection via the Blockchain.
sion farming, diminishing the use of pesticides and fertilizers, and Various sensors were deployed to measure temperature and humidity,
regenerative agriculture for restoring health to soils. The present paper while the position of the valves was determined by a fuzzy Mamdani
will be dedicated to the design of an intelligent irrigation system based model. Another study [15] proposed smart monitoring of agriculture
on some of the newest technologies, namely IoT, embedded systems, with fixed and mobile sensors, providing detailed maps of crop condi­
fuzzy logic, and cloud computing. The high vision is to devise a solution tions. A zoning irrigation system [16] was also developed, optimizing
that can optimize crop irrigation in line with changing climate condi­ plant growth and reducing water and energy consumption through a
tions and soil requirements. This three-layer architecture—comprising fuzzy logic controller and Node-RED interface. Finally, an innovative
IoT field devices, the ThingSpeak cloud platform, and an interactive solution [17] was presented to address the challenges of food produc­
dashboard—is designed to enhance water-use efficiency, effectiveness, tion, using a fuzzy controller in a small greenhouse to adjust environ­
and accuracy in responding to the management of water resources. mental conditions in real-time, improving agricultural efficiency and
Fig. 1 highlight traditional farming practices in Fez region of sustainability. Article [18] presents a smart agriculture framework that
Morocco (North Africa). This rural territory carries representative integrates renewable energy, the IoT for precise irrigation, and an
methods of farming passed down through generations and showcases Android-controlled robotic system. A case study in Sharjah shows the
attachment to its farming culture within the community. The following effectiveness of these technologies via the Blynk platform. Article [19]
images highlight traditional techniques shaped by the region’s topog­ describes a system using smart sensors and fuzzy logic to optimize irri­
raphy and unique climate. It also symbolizes the tool, crop, and method gation and environmental control in mushroom crops. Article [20]
of work peculiarities that characterize this region. These images express proposes a fuzzy irrigation system for cotton fields, improving water
all the resistance of the locals and their important role in the mainte­ efficiency and reducing infestations. The study [21] develops a
nance of agricultural biodiversity in this part of North Africa. cloud-based irrigation system to centralize and optimize water use,
This article [8] introduced an innovative solution combining while the article [22] proposes a tool to improve infrastructure sus­
embedded systems and environmental sensors for a real-time intelligent tainability by reducing CO2 emissions.
irrigation system. The system automatically adjusts irrigation according This article [23] has studied farmers in most regions of India; the
to environmental conditions, improving water use efficiency. Using an majority of those did not have any viable irrigation system. In finding
Arduino Mega 2560 microcontroller and various sensors (temperature, the solution, a solar irrigation system was designed and deployed; an
DHT22 humidity, soil moisture, water level), it optimizes irrigation and innovative system integrated with IoT-based automated systems. The
reduces water waste while increasing agricultural productivity. The IoT system works on the logic of soil moisture and ensures at the right
developed algorithm allows continuous monitoring and adaptive control time the crops get the right amount of water. It is also an advanced
of pumps with data recording for real-time feedback. In addition, article photovoltaic pumping system with all the required capabilities to
[9] presents a real-time fire detection system for smart agriculture, using overcome irrigation challenges throughout various Indian climatic
IoT, embedded systems, and cybersecurity measures. It quickly detects zones, namely composite, hot and humid, cold, and moderate, in addi­
smoke or flames thanks to sensors and a Raspberry Pi 3 B+, with tion to hot and dry climates. This paper [24] also presents a model for
real-time visualization via a Flask interface. Results confirm its effec­ sustainable, smart greenhouse photovoltaic solar energy sources effi­
tiveness. Further, article [10] proposes an intelligent irrigation system, ciently utilized by intelligent control systems for irrigation, temperature,
based on IoT, embedded systems, and cloud, which monitors humidity, and humidity optimization of a closed agricultural atmosphere. This
temperature, and water levels in real-time, and automatically controls system provides autonomous management due to IoT and micro­
water pumps via ThingSpeak and ThingView. This article [11] presents controllers. Moreover, compared with the traditional greenhouse, there
an intelligent IoT-based precision irrigation system integrated with the will be an increase in the efficiency of water consumption by 40 %. This
platform. It details the architecture and operation of the IoT node used. is supplied with condensate from air-conditioning systems that provide
In addition, a radio-frequency energy harvesting technique is applied to 65 % of the water required to irrigate the vegetables for added profit­
ability, more so in the Gulf Countries Council. Article [25] discussed the
smart grid to include IoT and cloud for better sustainability and reli­
ability. It depicts an IoT strategy reducing solar panel fouling and saving
costs of cleaning; thereby, enhancing sustainable energy production.
This system also introduces new IoT-enabled sensors for solar in­
stallations that will extend the life of your equipment while achieving
energy efficiency and accurate information on system operations.
Additionally, CloudIoT uses progressive networks through predictive
analytics, energy optimization, and smart design. It will finally propose
a system based on fog computing in order to reach optimum demand
with advantageous pricing while intelligently managing the grid with
respect to energy sustainability. This paper [26] proposes an "Open Edge
IoT System for Energy Monitoring in Buildings." As deployed within the
Faculty of Electromechanical Engineering of the University of Colima, it
optimizes this already existing network infrastructure by taking
advantage of edge-fog computing, and securely processes sensitive
Fig. 1. Traditional agricultural practices Fez, Morocco (North Africa). electrical data. It is also compatible with various IoT technologies,

2
A. Morchid et al. Smart Agricultural Technology 11 (2025) 100979

ensuring the accuracy of power, voltage, and current measurements. guide to evaluate their intentions [40]. This paper [41] presents the
This describes how the effectiveness of edge computing stands compared awareness of skin diseases and their impact on the implementation of
to other cloud architectures. This article [27] discusses the precision IoT-based intelligent skin monitoring systems. It also explores the
farming technology adapted to home gardening. It gives economic, awareness of IoT systems and their influence on the implementation of
health, physiological, and psychological benefits. It presents an inte­ these technologies, while highlighting the benefits of using an IoT sys­
grated sensor system for smart irrigation by using wireless sensor net­ tem to improve the management and monitoring of skin conditions. In
works for the analysis of soil and water, including accurate pH and addition, this article [42] demonstrates the positive effects of perfor­
humidity measurements. In [28], it presents an enhanced method for mance expectancy, effort expectancy, social influence, and facilitating
fault detection in a wireless irrigation network. Sensor failure is one of conditions on the intention to adopt IoT technology. However, among
the main issues to guarantee real reliability and performance within the older farmers, acceptance of changes in working methods (including the
systems. This approach will provide improvements in maintenance by use of IoT) becomes more difficult with age. In addition, individuals
facilitating the identification of faulty sensors in a faster and more with higher incomes are generally risk-averse and reluctant to accept the
effective manner. Article [29] contributes to this review of intelligent costs of installing modern smart technologies. Further, this study [43]
manufacturing systems in depth, within the context of advanced tech­ provides a review of smart technologies and their impacts on rural
nologies. It goes deep in exploring the analysis of associated technolo­ communities, highlighting the need for empirical research and consid­
gies, AI, and robotics together with challenges related to their eration of socio-cultural contexts. It also explores the contribution of a
implementation, which indeed is an important pre-requisite when socio-technical perspective in this field. On the other hand, this article
considering the integration barriers of such systems within existing [44] examines the adoption of digital technologies in agriculture, a
production environments. Article [30] also deals with the shift toward gradual process influenced by performance expectations and social
sustainable and circular practices, providing nineteen facilitating factors pressure. Farmers’ intentions and the characteristics of their farms play
in the technology adoption of the circular economy. These technolog­ a key role in the decision to adopt these technologies. Well-structured
ical, economic, and regulatory factors are essential for the correct and organized farms are more inclined to incorporate these in­
application of circular economy principles in industries. Still, on the novations. The motivation comes from the increasing challenges
topic of the circular economy, article [31] proposes a possible role Lo­ imposed upon agriculture by both the scarcity of water resources and the
gistics 4.0 could play in supporting the agri-food sector. It highlights impact of climate change. Traditional irrigation methods were quite
best practices in sustainability and how resource management is con­ incapable under those conditions of dealing with such changes in the
nected with the adoption and diffusion of Logistic 4.0 tools and tech­ environment, and even furthering water waste and stress to crops. In the
nologies, enabling the spread of the Circular Economy concept through presence of such scenarios, the present work offers a new efficient
advanced information systems and traceability. Meanwhile, article [32] perspective toward the optimization of irrigation, making use of modern
is about how to contribute to reducing emissions in the agricultural technologies that will bring better sustainability in water and food re­
sector of the countries of South Africa. The existing strategies, such as sources toward resilient and sustainable agricultural development.
sustainable agriculture and clean technologies, were assessed, taking Fig. 2 presents a graphical abstract of this study.
into consideration their actual contribution to carbon footprint reduc­ Although numerous irrigation systems leveraging the IoT have been
tion in this sector. Article [33] reviews the contribution of rural co­ conceived, a significant proportion of these pertains to reliance upon
operatives from the point of view of this question, while Article [34] static thresholds or elementary sensor feedback mechanisms, which may
makes a comparison between digital and green technologies for decar­ incite either an excess or deficiency in irrigation, particularly amid
bonization but shows precisely in what aspects each of these approaches varying environmental circumstances. This paper delineates a unique
contributes to reduced carbon emissions and therefore to the general methodology that propels existing paradigms forward by amalgamating
sustainability of the agricultural sector. The use of real-time monitoring IoT with fuzzy logic control frameworks, which proffer a more sophis­
to boost the efficacy of nature-based flood management techniques is ticated and flexible modality for the administration of irrigation
suggested in this research [35]. This methodology outperforms standard grounded in real-time atmospheric data. In contrast to systems predi­
methods in reducing water levels and optimizing storage, according to a cated upon fixed thresholds, Fuzzy logic enables more flexible decision-
study conducted in the Rangsit district of Thailand. However, this making by interpreting input variables, such as temperature and soil
contribution [36] offers a framework for the implementation of IoT in moisture, in colloquial classifications (e.g., ’low,’ ’medium,’ and ’high’)
agriculture and compares several applications. Four research topics and formulating irrigation determinations that are responsive to real-
were addressed by a systematic review of 82 recent articles. The findings time adjustments for the preservation of optimal soil moisture metrics.
demonstrate that research has shifted toward real-world application and The distinctive contributions engendered by this research can be
is increasingly utilizing machine learning methods. Additionally, an enumerated as follows:
examination of IoT integration for sustainable water management is
provided in this work [37]. It finds 25 elements, uses principal compo­ • Adaptability enhancement: The added flexibility through the use of
nent analysis to group them into 7 enablers, and uses a fuzzy fuzzy logic further increases; hence, the proposed framework can
theory-based multi-criteria technique to rank them. The IoT ecosystem, adapt to a wide range of environmental conditions and contempo­
its setup, and data mobility are the primary enablers. An automated IoT raneously recasting irrigation schedules in order not to over- or
system for calculating environmental river flows using publicly avail­ underapply water resources.
able hydrological data is proposed in this contribution [38]. Being • Improved water efficiency, since it automatically changes the irri­
reusable and expandable, it improves the value of already-existing data, gation time by conducting an in-depth analysis of the temperature
lowers costs, and makes environmental compliance easier. Furthermore, and soil moisture conditions to depict an appropriate time for
by lowering greenhouse gas emissions, innovation and technology releasing water.
support sustainable agriculture, as this research [39] highlights. It shows • Fostering cloud computing integration: Improvement in cloud
how the adoption of technology affects productivity and emissions, and computing integration: Because the Thingspeak cloud continuously
how the uneven adoption of technology in Asian nations has an impact stores real-time data in a single repository, there is true continuous
on environmental sustainability. Applications of machine learning also monitoring and improvement; hence, the system will be more
provide opportunities to advance sustainable farming methods. How­ responsive and accurate than any other manual method.
ever, small-scale production and outdated technology hinder Southeast • Empirical validation in realistic environments: This system was
Asia’s main industry, agriculture [40]. Farmers’ adoption of new tech­ taken further into the field within the real agricultural environment,
nology is influenced by several characteristics, which may be used as a hence corroborating its efficiency to keep soil moisture closer to the

3
A. Morchid et al. Smart Agricultural Technology 11 (2025) 100979

Fig. 2. Graphical abstract of the proposed system.

optimal thresholds irrespective of the variations in the environ­ and efficient management of irrigation. Membership functions are
mental conditions. defined in measured values, giving the specific operating levels for the
irrigation system at certain critical parameters: temperature, humidity,
This demonstrates that a key contribution of this paper is to present and soil moisture. Finally, this section provides an in-depth description
the advantages of fuzzy logic over conventional control systems imple­ of the proposed scheme algorithm, explaining each key step of the
mented in a real-world environment. process in detail to clarify the underlying mechanisms and logic
The paper is organized as follows: Section 2 presents Materials and involved. This approach ensures that irrigation is adapted to actual crop
Methods, Section 3 discusses Results and Discussion, and Section 4 is needs, contributing to more sustainable management of water resources.
devoted to Conclusion and Future Work.
2.1. Framework of the system
2. Materials and methods
IoT technologies, cloud computing, and real-time dashboards
This section presents a rigorous methodological approach for the
contribute something new to intelligent irrigation management to
development and implementation of an intelligent irrigation manage­
optimize water resource use for agriculture. The system is made up of
ment system, integrating IoT technologies, cloud computing, and fuzzy
three well-organized yet interdependent layers that authorize correct
logic. The main objective of this research is to design an architecture
monitoring and effective managing of the environmental conditions of
capable of meeting the growing needs for sustainable water manage­
crops. The collection, analysis, and interpretation of data in real time
ment in agriculture while improving crop resilience in the face of
strengthen not only the process of decision-making but also contribute to
environmental challenges. The three tiers in the proposed architecture
sustainability by way of resource wastage reduction. This system also
are out-of-the-field deployed devices, computations on the cloud
contributes to sustainable development and protection of water re­
through the ThingSpeak platform, and visualization in real-time through
sources by managing water resources in a more balanced and respon­
a dashboard. IoT devices include ESP32 microcontrollers, temperature,
sible way. Fig. 3 depicts the architecture of the system here proposed,
humidity, and soil moisture sensors, and motors like pumps. They shall
whose description is given below, considering peculiar functionalities of
handle the responsibility of gathering environmental data and trans­
each layer composing it-namely, IoT devices deployed in the field, cloud
mitting the same over HTTP to the cloud in near real-time. In turn, this
computing via ThingSpeak, and a real-time dashboard. Each device will
will be elaborated, stored, and analyzed in real-time through Thing­
play its role in developing automation and optimization irrigation pro­
Speak, delivering critical information to support decisions. Finally,
cesses in general for achieving more efficient and resilient agriculture.
displaying the possibility of a real-time dashboard allows for the visu­
alization and optimization of irrigation strategies. The application of
2.1.1. IoT devices (field layer)
fuzzy logic for decision-making is at the core of this research. In relation
The main heart of the IoT system is the integration of intelligent
to the subject, fuzzy logic plays its part in modeling all these complex
components with field condition sensing and real-time actuation. The
and uncertain environmental conditions, thus allowing more flexible
ESP32 microcontrollers are the center of the field layer, as they will be

4
A. Morchid et al. Smart Agricultural Technology 11 (2025) 100979

Fig. 3. Framework of the proposed scheme.

the orchestrator in the entire process based on the data gathered from management. The interface will be of assistance to the end-users because
the sensors installed within the field. Sensors operating due to ventila­ it will transform technical data into understandable, directly applicable
tion measure all the indispensable parameters, such as temperature, air information, hence will contribute to increasing the general efficiency of
humidity, and soil moisture for an effective overall picture of the envi­ the irrigation management system.
ronmental condition. Information gathered, therefore, acts as input to
actuators, which could be irrigation pumps. They water the land auto­
2.2. Mathematical modeling and fuzzy logic rules
matically and precisely, should the need be in that particular area. All
this is powerfully effective through a communication gateway that en­
When developing an intelligent irrigation system based on fuzzy
ables flawless data transfer between the field devices and the cloud.
logic, rigorous mathematical modeling is essential to ensure efficient
Communication using HTTP protocol allows for secure and efficient data
and optimized water management. This modeling is based on a fuzzy
transmission to the ThingSpeak cloud platform. In real practice, this
approach that considers uncertainties and variations in environmental
system - implemented by the field users, the majority of whom are
parameters, such as temperature and soil moisture, to adapt the irriga­
farmers through the friendly user interfaces - returns highly accurate
tion duration dynamically. This process comprises several key steps:
recommendations on irrigation management for decision-making to
input fuzzification, which converts numerical values into fuzzy sets
sustain farming.
using membership functions; fuzzy inference, based on a set of logical
rules formulated from the analysis of environmental conditions and their
2.1.2. Cloud computing (ThingSpeak)
impact on irrigation; and finally, defuzzification, which transforms the
The ThingSpeak cloud computing platform occupies a strategic po­
fuzzy output into a concrete value representing the optimal irrigation
sition within this architecture, ensuring not only the storage of data
duration. The aim is to detail the various stages in the mathematical
collected in the field but also its real-time analysis. Data sent from ESP32
modeling of the system, highlighting the formulas used and their role in
microcontrollers via the HTTP protocol is aggregated and processed on
real-time decision-making for intelligent, sustainable irrigation. Input
this platform. ThingSpeak uses advanced algorithms to analyze envi­
fuzzification is the process of converting precise input values into fuzzy
ronmental conditions, enabling trends and the proactive identification
values using membership functions. This process is used to represent the
of irrigation needs. In addition, the platform offers robust visualization
uncertainty and variability of environmental conditions in smart irri­
capabilities, enabling raw data to be transformed into actionable in­
gation. The system’s input variables are defined as follows:
formation. These analyses are essential to maintain optimal irrigation,
thus reducing water waste and ensuring that crops receive adequate
• Input Variables: Temperature T (◦ C): Low (LT ), Midium (MT ),
water. With ThingSpeak, users can also store large amounts of historical
High (HT )
data, making it easier to make informed decisions based on past and
• Soil Moisture S (%): Low (LS ), Desired (DS ), High (HS )
present patterns.
Each input value xi is associated with a fuzzy set Ai via a membership
2.1.3. Real time dashboard
function μAi (xi ) which expresses the degree to which this value belongs
The real-time dashboard is the system’s user interface, providing a
to a given set, as shown in Eq. (1) :
dynamic overview of field operations. The dashboard shall be enabled
with ThingSpeak, showing continuous data and analytics on the critical μAi (xi ) ∈ [0, 1] (1)
operational parameters related to farmers like soil moisture, ambient
humidity, and temperature, among others. Intuitive visualization of For example, soil moisture fuzzification is represented by the
complex information in simple ways will enable users to reach quick following membership functions, as illustrated in Eq. (2):
decisions. Moreover, the dashboard should be interactive where the user μLs (S), μDs (S), μHs (S) (2)
will be able to change the system settings in light of the received rec­
ommendations for better precision and timeliness of irrigation where μLS (S) represents the degree to which soil moisture belongs to the

5
A. Morchid et al. Smart Agricultural Technology 11 (2025) 100979

Low category. intelligent irrigation system are listed below. Some of these are shown in
Defining membership functions, the membership functions used to Table 1. These will ensure that this system makes optimized decisions on
represent fuzzy variables are generally triangular and trapezoidal. These the duration of the irrigation from the prevailing ambient temperature
functions ensure a smooth transition between the different fuzzy cate­ and soil moisture content. This table tabulates nine rules with respect to
gories of input variables. Triangular function: defined by three param­ all the possible variations that might arise from two main inputs: tem­
eters (a, b, c), it is given by Eq. (3) : perature and soil moisture. Each of these inputs is categorized into three
⎧ levels: low, medium, or high-desired in the case of soil moisture. Ac­
0, x ≤ a or x ≥ c
⎪
⎪
⎪ cording to the combination of these levels, the corresponding rule de­
⎨ x− a
⎪
termines for how long-that is, short, medium, or long-the irrigation
μA (x) = b − a, a ≤ x ≤ b (3)
⎪
⎪ c− x should be effected. It proposes a long irrigation time with high tem­
⎪
⎪
⎩ , b≤x≤c peratures, adding the factor of low soil moisture to compensate for high
c− b evaporation and maintain adequate soil moisture for crops. In cases of
Trapezoidal function: defined by four parameters (a, b, c, d), it is high humidity in the soil at high temperatures, the duration of irrigation
given by Eq. (4) : is reduced to avoid the excess water that would hurt plant growth, not to
⎧ mention waste of such a valuable resource. Fuzzy logic allows the system
⎪
⎪ 0, x ≤ a or x ≥ d
⎪ to change dynamically with field conditions, hence improving water
⎪
⎪
⎪
⎪
⎪ x− a usage efficiency and sustainable agriculture. This intelligent model re­
⎨ b − a, a ≤ x ≤ b
⎪
⎪ duces overdependence on continuous human interference and assures
μA (x) = (4) that irrigation is precisely and optimally done for the attainment of food
⎪ 1, b ≤ x ≤ c
⎪
⎪
⎪
⎪ security and environmental sustainability. The use of fuzzy logic in this
⎪
⎪ d− x study provides a robust approach to intelligent irrigation management,
⎩ d − c, c ≤ x ≤ d
⎪
⎪
⎪
contributing to optimized and adaptive decision-making in the face of
environmental variations. Precise adjustment of membership functions
A rule-based inference system, fuzzy inference relies on a set of improves system reliability and ensures better water security by regu­
logical rules that describe the relationship between inputs (soil tem­ lating water supply efficiently.
perature and moisture) and output (irrigation duration). The fuzzy rules Fig. 4 illustrates the membership functions associated with the
are formulated based on the provided rule Table 2. The general structure “temperature” input variable in the context of the intelligent irrigation
of a fuzzy rule is given by Eq. (5): system. The temperature interval considered ranges from 15 ◦ C to 45 ◦ C,
Ri : IFTis AT ANDSis AS THENIis AI (5) covering three linguistic categories: Low, Medium, and High. The Low
membership function has a trapezoidal shape, reaching a maximum
Where: value between 15 ◦ C and 25 ◦ C, then decreasing linearly until it reaches
Ai : Temperature fuzzy set ( LT , MT , HT ) zero at 25 ◦ C. Above average, the triangular membership function peaks
As : Soil Moisture fuzzy set ( LS , MS , HS ) at 30 ◦ C. the ideal temperature. From this optimum, the membership
AI : Irrigation fuzzy set (Long, Medium, Short) value decreases linearly with deviation in either direction from this
Each rule is applied using the minimum operator to represent logical optimum. The High membership function, which is trapezoidal, starts an
conjunction (AND). The degree of activation of each rule is given by Eq. increase at 35 ◦ C and reaches full membership at 45 ◦ C. These functions
(6) : provide a basis for assessment for the temperature condition under
(
ωi = min μAT (T), μAS (S)
)
(6) which irrigation decisions are made.
Fig. 5 presents the membership functions of the input variable "soil
Where ωi represents the weight of the rule as a function of the de­ moisture," which becomes necessary for the determination of the irri­
grees to which the inputs belong to the corresponding fuzzy sets. gation needs of the system. The considered classes are Low, Desired, and
Defuzzification is the reverse process of fuzzification. It converts the High. The variations in soil moisture range from 55 % to 70 %. For the
fuzzy values obtained at the output of the inference system into a precise Low class, it was adopted with a trapezoidal curve, signaling low levels
numerical value representing the irrigation time. A classic defuzzifica­ of soil moisture. Its membership becomes complete within the interval
tion method is the center of gravity, where the output value I is calcu­ from 55 % to 59 %, then declines. The conventional triangular mem­
lated according to Eq. (7): bership function for class Desired, with its peak at 62 %, can be inter­
∑ preted as an optimum soil moisture condition. The maximum
ωi ci
I = ∑i (7) membership will be attained only at 62 %, and it represents the ideal
i ωi ci
state where no further irrigation is needed for the crop. In the same way,
Where ωi is the rule active weight. On the thar hand, ωi is crisp the High membership function has its peak above 63 % in the same way
irrigation value (e.g., Long = 10 min, Medium = 5 min, Short = 2 min). as in the Low category, designating excessive levels of the soil moisture
In this way, defuzzification makes it possible to determine an irrigation where irrigation should be minimized or stopped. Fig. 6 Membership
time suited to the environmental conditions measured, guaranteeing functions corresponding to the output variable "irrigation duration" sets
optimized management of water resources.
This part presents the fuzzy logic rules and membership functions
Table 1
used in our intelligent irrigation management system. Fuzzy Logic Rules:
Fuzzy logic rules.
Fuzzy logic rules have been very important in determining, concerning
temperature variation and soil moisture, the ideal duration of irrigation. [Link] Temperature Soil Moisture Irrigation Duration
The membership functions model such parameters as linguistic variables 1 Low Low Long
that enable automatic decision-making. It becomes critical for the 2 Low Desired Medium
realization of better sustainability and efficiency in agricultural prac­ 3 Low High Short
4 Medium Low Long
tices, as it gives an exact regulation of crop water requirements, 5 Medium Desired Medium
considering fluctuating environmental conditions. This is represented by 6 Medium High Short
the figures below, showing the input and output variables membership 7 High Low Long
functions. The table of rules gives an insight into the decisions taken by 8 High Desired Medium
9 High High Short
the system. Some of the fuzzy rules that can be used to control this

6
A. Morchid et al. Smart Agricultural Technology 11 (2025) 100979

Fig. 4. Temperature membership functions. Fig. 6. Irrigation duration membership functions.

2.3. Algorithm of the proposed scheme

Table 2: The pseudo-code of the suggested intelligent irrigation using

fuzzy logic explains the way real-time intelligence is involved in an
irrigation system to optimally manage water based on with prevailing
environmental conditions, taking into consideration soil moisture,
ambient temperature, and air humidity. It shall be constituted of a
simple ESP32 platform, sensors, a water pump, and a fuzzy logic-based
controller. It initializes the soil moisture sensors, temperature, and air
humidity, beside the water pump. Then, the ESP32 connects to the Wi-Fi
network to enable communication with the ThingSpeak cloud, which is
a platform used for data monitoring and recording in real-time. Input
variables will be defined from sensor measures: soil moisture, temper­
ature, and air humidity, and only one output variable will be defined,
namely irrigation duration. In this way, a membership function is
generated using MATLAB that can classify data into classes such as High,
Low, Short, Medium, Long, and Desired such that the sensor data is
qualitatively interpretable. The fuzzy logic rules are now imposed on the
system to obtain the optimal time of irrigation. For example, in the case
Fig. 5. Soil moisture membership functions. of low soil moisture coupled with high temperature, the irrigation time
is increased. These rules can also be moved depending on what a certain
the time period over which irrigation has to be applied. Irrigation crop needs, and allow flexibility within the process. The sensors will be
duration, varying between 0 to 10 min, is divided into three linguistic interrogated regularly for new measures that will be translated into
terms or categories: Short, Medium, and Long. The trapezoidal shape of fuzzy values (fuzzification). Later on, this decision process will apply
the Short membership function reflects minimal irrigation with its peak fuzzy rules and will perform defuzzification with the purpose of esti­
lying between 0 and 2 min, then it starts to decline. The triangular mating a concrete irrigation duration. If the irrigation is computed, then
membership function of Average irrigation duration peaks at a value of 3 the pump is turned on for the computed time, and data about this
min, reflecting an irrigation requirement at the level of medium. In the operation -time and length- is sent to ThingSpeak; in case irrigation is
same way, the increasing function Long ranges from 5 to 10 min. This not needed, the pump stays off but again the status of the system is
denotes that the conditions may be such that long irrigation is called for. transmitted to the cloud. The acquired data from all sensors is periodi­
With these, the system will be able to fine-tune the irrigation duration in cally sent to ThingSpeak, using HTTP POST requests. This, in turn, en­
response to the environmental parameters assessed. In this intelligent hances the possibility for real-time environmental condition analysis
irrigation system, the specific needs of crops are calculated to optimize and, thus, makes decisions based on irrigation. The system constantly
water use. By customizing humidity and sun temperature, the system monitors for any type of failure; for example, sensor failures or failures
adjusts irrigation to the specific requirements of tea crops, such as rice or on the cloud connection. Any problem detected gets logged, and an
tomatoes. The use of fuzzy logic enables dynamic adaptation of irriga­ automatic attempt to reset or reconnect to Wi-Fi gets launched. If it is
tion duration and frequency, taking into account environmental condi­ doomed with continuous failure, then notification would be provided to
tions and the characteristics of the tea crop. In addition, by exploiting the user through ThingSpeak itself. MATLAB has been used in the design
historical and temporal data, the system optimizes irrigation using cli­ for simulation and fine-tuning of membership functions and fuzzy rules
matic and local conditions, which also guarantees efficient water man­ for optimization of the onboard controller to perform efficient irrigation
agement. The user interface enables farmers to customize irrigation in terms of the right need at any instance. Lastly, shutdown can be done
parameters for tea cultivation, guaranteeing a fluid and robust either manually or automatically based on preset conditions. It will
approach. switch off the pump at the time of shutdown and disconnect the Wi-Fi
connection while saving data for further analysis.

7
A. Morchid et al. Smart Agricultural Technology 11 (2025) 100979

Table 2
Algorithm 1: Pseudo-code of the proposed smart irrigation-based fuzzy logic.
// Initialization of sensors and devices
Initialize soil moisture sensor
Initialize temperature sensor
Initialize air humidity sensor
Initialize water pump
Initialize ESP32 microcontroller
// Network connection configuration
Configure Wi-Fi connection
Configure ThingSpeak connection using API keys for data transmission (via HTTP)
// Initialization of fuzzy logic controller
// Define input fuzzy variables (temperature, soil moisture)
// Define output fuzzy variable (irrigation duration)
Define fuzzy sets for temperature: [Low, Medium, High]
Define fuzzy sets for soil moisture: [Low, Desired, High]
Define fuzzy sets for irrigation duration: [Short, Medium, Long]
// Use MATLAB to create membership functions for classifying sensor data into fuzzy levels:
For temperature: Low, Medium, High
For soil moisture: Low, Desired, High
For irrigation duration: Short, Medium, Long
// Define fuzzy logic rules based on the following table:
Rule 1: If temperature is Low AND soil moisture is Low, THEN irrigation duration is Long
Rule 2: If temperature is Low AND soil moisture is Desired, THEN irrigation duration is Medium
Rule 3: If temperature is Low AND soil moisture is High, THEN irrigation duration is Short
Rule 4: If temperature is Medium AND soil moisture is Low, THEN irrigation duration is Long
Rule 5: If temperature is Medium AND soil moisture is Desired, THEN irrigation duration is Medium
Rule 6: If temperature is Medium AND soil moisture is High, THEN irrigation duration is Short
Rule 7: If temperature is High AND soil moisture is Low, THEN irrigation duration is Long
Rule 8: If temperature is High AND soil moisture is Desired, THEN irrigation duration is Medium
Rule 9: If temperature is High AND soil moisture is High, THEN irrigation duration is Short
// Main loop
Infinite loop:
// Read sensor data
soilMoisture = Read soil moisture sensor
temperature = Read temperature sensor
// Fuzzification: Convert sensor values into fuzzy sets (Low, Desired, High for soil moisture; Low, Medium, High for temperature)
fuzzyMoisture = Fuzzify(soilMoisture)
fuzzyTemperature = Fuzzify(temperature)
// Apply fuzzy logic rules to determine irrigation duration
irrigationDuration = ApplyFuzzyRules(fuzzyTemperature, fuzzyMoisture)
// Defuzzification: Convert fuzzy output (irrigation duration) to a concrete value
concreteDuration = Defuzzify(irrigationDuration)
// If irrigation is needed, activate pump for a calculated duration
If concreteDuration > 0:
Activate pump for concreteDuration seconds
Send data to ThingSpeak (soil moisture, temperature, irrigation duration)
Else:
Send data to ThingSpeak (soil moisture, temperature, no irrigation)
// Transmit all system data to ThingSpeak cloud for remote monitoring
Send soilMoisture, temperature, system status to ThingSpeak via HTTP POST request
// Monitor for potential errors
If sensor failure or Wi-Fi disconnection is detected:
Log the error and send to ThingSpeak
Attempt to reconnect Wi-Fi or reset sensors
// Wait 2 seconds before the next sensor reading
Wait(2000 ms)
// Shutdown sequence
If the system is manually or automatically stopped:
Deactivate pump
Disconnect from Wi-Fi
Save data and logs for future analysis

3. Results and discussion HTTP protocol. It is in this section of the paper where the results of how
effective fuzzy logic algorithms are when applied to irrigation man­
It gives a detailed analysis of the obtained results and presents a deep agement in relation to variations in temperature and humidity are
discussion on the developed intelligent irrigation management system. simulated and experimentally dealt with. Hence, this analysis is done on
The design of the system was performed by combining modeling and the basis of the accuracy of the data collected, its impact on irrigation
analysis using a simulation tool represented by MATLAB. The execution decisions, and how the system works efficiently in an agricultural
of the algorithm, together with its control of hardware peripherals, has setting. This discussion will cover the performance evaluation of the
been accomplished using C programming, also called Arduino IDE. This system, focusing on the positive points and making some recommen­
ESP32 module integrates several sensors for data gathering, while the dations for future improvements to achieve maximum efficiency and
management and visualization are done through the ThingSpeak plat­ reliability from the system. To guarantee long-term operational reli­
form. To ensure effective and timely communication between sensors ability, our system has been subjected to several rigorous tests under
and the ThingSpeak platform, communication is implemented using various environmental conditions, including extreme situations. These

8
A. Morchid et al. Smart Agricultural Technology 11 (2025) 100979

tests evaluated system stability, sensor accuracy and the consistency of between 57.5 % and 65 %. Soil moisture was below 62 % at several
fuzzy logic-based irrigation decisions. Particular attention was paid to points, including 12:00 (59.5 %), 13:00 (58.5 %) and 17:15 (60 %).
sensor maintenance, essential to ensure system durability. The algo­ However, humidity levels exceeded 62 % at certain times, such as 12:15
rithm developed includes self-diagnostic mechanisms to detect any pm (62.5 %), 2:30 pm (64 %) and 3:15 pm (61 %). The dotted line at 62
sensor drift or failure in real time. This approach enables the system to % is the target moisture; it is convenient to compare with the actual soil
dynamically adjust its parameters or signal the need for intervention, moisture values throughout the day. Such variability in this case is
thus guaranteeing optimal long-term performance. The tests carried out controlled by fuzzy logic rules that measure how much irrigation is
and the maintenance strategies implemented reinforce the system’s deemed necessary by prevailing humidity levels as well as temperature
robustness, guaranteeing efficient, reliable irrigation, even under conditions. Fig. 10 represents temperature variation throughout the day.
demanding agricultural conditions. The temperature data varies from 15.25 ◦ C at 01:45 to 40 ◦ C at 11:00.
In this study, we paid particular attention to the implementation of Notable temperature points include 39.1 ◦ C at 13:00 and 26 ◦ C at 15:30.
robust security measures to guarantee the confidentiality and integrity The figure captures temperature fluctuations and their extremities,
of the data collected and transmitted by the system. The ThingSpeak which are of essence for the analysis that this serves in showing how
cloud platform used for data management offers advanced security temperature controls soil moisture and irrigation need. In the daily
features that protect sensitive information throughout its lifecycle. The performance evaluation, it was observed that the fuzzy logic controller
system uses a login authentication mechanism to secure access to data effectively prevents system operation during periods of high heat, as
stored on the ThingSpeak platform. This restricts access to authorized shown in Fig. 9. In addition, it favors short irrigation times when the
users only, ensuring strict control over data operations. This measure input parameters are moderate. This method significantly reduces the
protects the integrity of information by preventing unauthorized access losses due to evapotranspiration throughout the day. In addition, the
that could compromise system reliability. In addition, to guarantee the controller determines precisely the optimal times and durations for
confidentiality of data exchanged between IoT sensors in the field and system activation through carefully defined fuzzy logic rules. Further,
the cloud platform, we have integrated the use of the HTTP protocol. Fig. 10 illustrates the correlation between temperature and irrigation
This protocol encrypts transmitted information, offering protection duration throughout the day, i.e. irrigation duration as a function of
against attempts to intercept or alter data. Encryption enhances temperature evolution. The scatter plot reveals significant variation in
communication security, ensuring that sensitive information, such as irrigation times in response to temperature changes. For example, when
environmental parameters and irrigation decisions, is protected the temperature reached 40 ◦ C at 11:00, the irrigation time was 6 min.
throughout the transmission process. Conversely, at a lower temperature of 15.25 ◦ C at 01:45, irrigation time
Fig. 7 shows in detail a geospatial mapping of the location of the was minimal. The rules of fuzzy logic play a decisive role in extending
LISAC-Laboratory of Informatics, Signals, Automation, and Cognitivism, irrigation periods when temperatures are high, to ensure that the soil
being part of Faculté des Sciences Dhar El Mahraz - FSDM of the Uni­ retains a target humidity of 62 %. This pedoclimatic irrigation method
versité Sidi Mohamed Ben Abdellah - USMBA, in Fez, Morocco. It is by was validated during the summer season. The fuzzy logic controller
this Google Earth picture that the scene will be set for the current paper, avoids watering during times of significant weather fluctuations, as
which is going to introduce a new intelligent irrigation system. shown in Fig. 10, which minimizes water losses by evaporation while
Advanced technologies of fuzzy logic, IoT, and embedded systems were maintaining plants in optimal condition, that is, without water stress. In
integrated in the laboratory to allow the realization of the solution for addition, the duration of irrigation is adjusted according to data pro­
irrigation management with high performance. In fact, the proposed vided by sensors such as soil moisture, temperature, and ambient
system shall contribute to achieving sustainable agriculture that ensures humidity.
food security by introducing new agricultural practices that make op­ Fig. 11 depicts instantaneous values of measured temperature pre­
timum use of available water resources. This figure highlights the ex­ sented in the ThingSpeak platform. This graph shows the continuous
pected global impact of this research. In the same fashion, this collocates evolution of the measured temperatures; such a graph provides a clearer
the work in the particular geographical and academic context of the snapshot about the current climatic situation. In such a graph, one can
USMBA, a benchmark institution in the field of agricultural study real-time temperature variations and thus study the climatic ten­
technologies. dencies that might come out to be of great relevance regarding antici­
Over the specified observation period, the fuzzy logic-based irriga­ pating when irrigation should be done. Fig. 12 shows, in turn, the real-
tion system demonstrated its ability to adjust irrigation periods ac­ time values of air humidity as captured and visualized on ThingSpeak.
cording to slight variations in weather conditions, as illustrated in the This graph provides an accurate real-time view of the outdoor humidity
following figures and rules. Fig. 8 illustrates the variations in soil condition that is highly important for the reactive adjustment of irri­
moisture levels throughout the day, compared with the desired moisture gation management practices based on data acquired. Indeed, to
level of 62 %. The line graph shows that soil moisture averages varied monitor such humidity variations in real time will be an advantage in
optimizing the effectiveness of irrigation interventions. In addition,
Fig. 13 illustrates soil moisture levels in real-time, as monitored on
ThingSpeak. This graph shows soil saturation at different times,
enabling continuous and accurate monitoring of soil moisture status.
Real-time access to this data is crucial for assessing irrigation re­
quirements with a high degree of accuracy, guaranteeing optimal
management of water resources, and contributing to the sustainability of
agricultural practices.
In order to assess the affordability of the proposed system compared
with traditional irrigation methods, we present a detailed cost analysis
of the main components used in this study. Table 3 below summarizes
the prices of essential components, including the on-board microcon­
troller (ESP32), sensors (soil moisture sensor, DHT22 sensor), irrigation
water pump, as well as other necessary components (test plate, resistors,
connection cables, etc.). In addition, we take into account data man­
Fig. 7. Geospatial mapping of research Laboratory at FSDM, USMBA - Fez on agement solutions via the cloud (ThingSpeak) and the software tools
Google Earth. used for data processing (MATLAB in trial version). The total cost for

9
A. Morchid et al. Smart Agricultural Technology 11 (2025) 100979

Fig. 8. Soil moisture evolution and desired moisture.

Fig. 9. Temperature value.

Fig. 10. Irrigation duration in the function of temperature evolution.

implementing the IoT-based smart irrigation system, integrating all the typically require high investments in infrastructure installation and
components mentioned, comes to 32.00 USD. This is significantly lower maintenance (pumps, pipe networks) as well as manual monitoring of
than the costs associated with traditional irrigation methods, which soil moisture levels. While traditional methods involve high initial

10
A. Morchid et al. Smart Agricultural Technology 11 (2025) 100979

Table 3
Detailed cost analysis.
Component Description Price
(USD)

ESP32 Embedded development board ensuring 6.00

Microcontroller communication and management of sensors
and actuators.
Moisture Sensor Measures soil moisture to optimize irrigation 3.00
decisions.
DHT22 Sensor Detects ambient temperature and humidity to 4.00
adapt irrigation to climatic conditions.
Water Pump Ensures irrigation of the soil based on data 10.00
collected by sensors.
Other components Allows the prototyping and secure connection 9.00
of different electronic components. Stabilize
electrical signals and ensure proper circuit
operation
ThingSpeak (Cloud Provides real-time collection, storage and 0.00
Platform) visualization of sensor data. (Free)
Fig. 11. Temperature value in real time on ThongSpeak cloud. MATLAB (Trial Utilisé pour l’analyse et la visualisation 0.00
Version) avancée des données du système. (Free-
Trial)
Total Cost ​ 32

practices of vital importance for those regions that have big problems
with water scarcity and totally unpredictable weather conditions. The
system will be based on a fuzzy logic controller that will dynamically
adjust the irrigation schedule by taking real-time data input from a
network of sensors deployed in the area. These sensors are interfaced
using ESP32 microcontrollers that monitor critical environmental pa­
rameters: ambient temperature, soil moisture, and humidity in the air.
Contrary to most irrigation systems, which water on rigid schedules or
are activated by fixed threshold triggers, our system interprets sensor
data in real-time using fuzzy logic to make informed decisions on plant
water needs. It, in turn, enables more precise and appropriate man­
agement, hence assuring that water is only made available when it is
needed, reducing the risk of under or over-irrigation. The system com­
prises three interoperable layers: IoT devices-ESP32 controllers, sensors,
pumps, and gateways-cloud services, ThingSpeak; stores and processes
Fig. 12. Humidity value in real time on ThongSpeak cloud.
data, and there is a real-time dashboard used for the control and
monitoring of the device. The collected data is sent over the HTTP
protocol to the ThingSpeak platform for analysis and visualization. On
the other side, performance requirements made it essential to simulate
and visualize, by using MATLAB, membership functions of fuzzy logic in
order to tune them with precision in the development of control algo­
rithms. In this way, fuzzy logic rules would best correspond to specific
environmental conditions of the study area, thus optimizing the system’s
efficiency. Real-life test results demonstrated clear advantages
compared to conventional irrigation methods. A fuzzy logic controller
that could adjust irrigation time with the systematic maintenance of soil
moisture close to the target of 62 % was able to optimize water use.
Our system was performed by saving 70 % water compared to con­
ventional irrigation methods while optimum soil moisture was main­
tained. This level of savings confirms the system’s effectiveness in
optimizing of its water resources, particularly in those short periods of
supply. That way, this approach enables farmers to act in a more
considerate manner when modifying irrigation, thus contributing to
resource conservation. Another value addition is that the adaptive na­
Fig. 13. Moisture value in real time on ThongSpeak cloud. ture of fuzzy logic in irrigation management has been highly effective, as
against the threshold-based conventional systems. This permits flexi­
outlay, the proposed system is based on low-cost, on-board technologies bility in decision-making during variable weather conditions. In fact,
that improve irrigation efficiency, optimize water resource management during temperature peaks, the system delays irrigation during peak
and reduce manual workload. In the long term, this approach offers a hours to minimize losses due to evaporation. Similarly, when tempera­
more economical and sustainable solution for farms. tures fell, irrigation was trimmed to prevent over-irrigation, a factor that
In addition to the advancements in precision agriculture, the present underlined the effectiveness of fuzzy logic in a complex scenario where
paper proposes an intelligent irrigation system based on fuzzy logic with decisions were to be made based on many factors. The benefits derived
novelty in integrating IoT devices, cloud computing, and real-time from this system are not confined to water savings alone. This also en­
monitoring. It highly contributes to enhancing water management sures proper and timely irrigation of crops, increasing their yield and

11
A. Morchid et al. Smart Agricultural Technology 11 (2025) 100979

improving their quality. Real-time and long-term analysis enables subscriptions to cloud platforms such as ThingSpeak is an initial
farmers to make better decisions, which are enabled through the investment that can turn out quite unaffordable for certain farms, at
ThingSpeak cloud platform. Meanwhile, two other important milestones least in developing regions. Further added is the cost of maintenance
in applying IoT in agriculture have been the use of ESP32 micro­ and powering IoT devices. While long-term savings in water and
controllers for IoT applications and the further development of increases in yield may therefore be seen as reasonable rationalization
advanced sensing technologies. The case presented in the paper can for the adoption of such technologies, specific profit-making assess­
show the possible transformation of fuzzy logic in irrigation manage­ ments about the short term are indispensable, particularly in low-
ment function that requires intelligent and adaptive solutions from income areas. Consideration of subsidies or even public-private
contemporary agriculture. This cropping system regulates soil moisture partnerships encouraging the same in regions where water man­
efficiently and resists climatic oscillations; hence, it will play a greater agement is urgent should, therefore, be underlined.
role in the promotion of sustainable agriculture by judicious use of 3. Internet connectivity: The major bottleneck in rural areas pertains to
available water resources in most water-stressed regions. internet connectivity. Therefore, the proposed system relies heavily
(1) The proposed architecture has been designed to guarantee on a stable connection with the cloud for data transmission and real-
modularity and scalability to meet the varied needs of farming opera­ time monitoring. However, internet access in most remote agricul­
tions. The field layer is based on ESP32 microcontrollers, which tural areas is thin or negative, making it a possible hindrance to the
orchestrate the collection and analysis of data from environmental efficiency of the proposed system. This is further limited by the
sensors (temperature, air and soil humidity). Thanks to an efficient constant presence an internet connection requires. This could be
communication gateway and the use of HTTP protocol, this data is overcome using long-range, low-power communication networks,
transmitted securely to the ThingSpeak cloud platform. This architec­ such as LoRaWAN, which does not need an internet connection
ture ensures optimized resource management, reducing network over­ constantly but also allows for transmission over longer distances
load and maintaining operational efficiency even in complex, extended with energy consumption. Hybrid approaches could therefore be
environments. In addition, the ThingSpeak platform provides real-time introduced that would implement local communications and edge
analysis of collected data, facilitating dynamic adaptation of irrigation computing in order to make the full functionality of the system
cycles to environmental conditions. It can also store historical data to possible even without a stable connection to the Internet.
optimize long-term decision-making. The real-time dashboard is an 4. Computational complexity: Although fuzzy logic may be powerful,
essential interface for farmers, offering them a clear and intuitive visu­ the implementation involves high computational costs. Processes
alization of critical system parameters. Its interactivity enables them to such as fuzzification, inference, and defuzzification are hugely
adjust settings according to recommendations, guaranteeing precise and computation-intensive, especially in the cases of complex systems or
efficient irrigation management. To enhance the system’s scalability, we large datasets. The heavy computational burden may reduce per­
plan to integrate advanced strategies such as segmenting the network formance in these kinds of systems when operating in resource-
into intelligent sub-systems, using IoT gateways and optimizing data constrained environments or real-time applications. Whereas
processing algorithms. These improvements will enable the system to be increased computing power has softened some of these constraints,
adapted to large-scale farming operations, while guaranteeing sustain­ the challenge remains very real for systems deployed to rural areas
able management of water resources. Using MATLAB to adjust mem­ lacking ready access to large computing resources.
bership functions can be tedious for some users. To make this task more 5. Membership functions: Membership functions are subjective, and
accessible, user-friendly alternatives can be considered, such as web their selection is one of the most fundamental issues in fuzzy logic
interfaces or mobile applications offering intuitive adjustment, as well systems; this still remains very subjective. These functions, to a great
as No-Code/Low-Code dashboards enabling simplified adjustments. The degree, depend on human expertise or intuition for their definition.
use of Excel with VBA or Python is an accessible solution without the These points become causes for variance in systems design due to
need for MATLAB. In addition, the integration of artificial intelligence- different interpretations by different experts. This may result in
based tools, such as genetic algorithms or particle swarm optimization, biased or inconsistent behavior of the system and hence will affect
enables dynamic adjustment of membership functions according to the accuracy and reliability in the irrigation decision-making process.
data collected. These approaches aim to improve the accessibility and This may be optimized by strict membership function validation and
efficiency of fuzzy models without requiring advanced programming periodic calibration of the system to reduce such errors and give
expertise. reliable results.
The development of the smart irrigation system with the use of fuzzy 6. Socio-economic factors in the adoption of agricultural technologies:
logic and IoT comes with its own advantages in handling agricultural Limited attention to socio-economic factors, such as farmers’ will­
water management. Besides, there are also certain challenges and lim­ ingness to adopt new technologies, can hamper the practical imple­
itations on its scalability, cost, and connectivity among other issues. mentation of these innovations. To remedy this, it is essential to
Thus, this should be taken as a wider diffusion perspective in developing implement awareness-raising and training strategies tailored to
regions. farmers. Targeted educational programs, practical field demonstra­
tions and awareness campaigns can play a key role in promoting
1. Scalable System: One of the main challenges facing large farms is understanding of the benefits of agricultural technologies. In addi­
scalability. The perfect implementation of our system in small and tion, it is important to consider the economic aspects, offering in­
medium-scale farming makes integration over larger areas very centives and business models adapted to local realities, in order to
complicated with sensors and more IoT devices. This would require facilitate adoption and ensure successful integration of technologies
an extended sensor network with considerable communication into farming practices.
infrastructure to manage a large volume of data in real-time. More­
over, the technological change would have to be implemented on 4. Conclusion and future work
many control points and actuators so they could keep up with high
response and elaboration speed. The use of LoRaWAN or mesh This study presented an innovative smart irrigation system,
technologies would prolong their coverage in large environments combining the Internet of Things (IoT), fuzzy logic, cloud computing
and give added value to data management. and real-time sensors to address the challenges of water management in
2. Cost: The initial implementation cost of such a system is rather high agriculture. Thanks to mathematical modeling, the system dynamically
since it relies so much on technological infrastructure. Buying ESP32 adjusts irrigation needs according to soil temperature and humidity,
microcontrollers, sensors, and communication devices alone enabling more precise management of water resources and a significant

12
A. Morchid et al. Smart Agricultural Technology 11 (2025) 100979

reduction in wastage. The integration of fuzzy logic in irrigation Declaration of competing interest
decision-making guarantees fine-tuned adaptation to changing condi­
tions, such as climatic variations and fluctuating soil moisture. The use The authors declare that they have no known competing financial
of MATLAB to simulate and model fuzzy membership functions enables interests or personal relationships that could have appeared to influence
clear visualization of decision mechanisms and dynamic adjustment of the work reported in this paper.
irrigation parameters. The results show that, thanks to intelligent irri­
gation management, the system is able to maintain optimum soil Acknowledgement
moisture around the target threshold of 62 %, while avoiding over-
irrigation, in particular by reducing watering time during moderate The authors used generative AI tools to assist in grammar correction,
temperatures and increasing it during periods of intense heat. This not clarity improvement, and refinement of language throughout this
only reduces water loss through evaporation, but also water stress in manuscript. All content was reviewed and validated by the authors to
crops, ensuring optimal plant growth. In terms of performance, analysis ensure accuracy, originality, and adherence to ethical standards.
over an extended period showed a marked improvement in irrigation
efficiency. The system achieved a substantial reduction in water losses,
Data Availability Statement
using strictly defined fuzzy rules to adjust irrigation to actual crop needs.
These dynamic adjustments also enabled a reduction in water con­
All the data is presented in this paper.
sumption, thus contributing to a more sustainable management of water
resources. The results show a significant reduction in water consump­
References
tion, and the system is more affordable (USD 32.00) than traditional
methods. The use of the ThingSpeak cloud platform enables continuous [1] K. Pawlak, M. Kołodziejczak, The role of agriculture in ensuring food security in
monitoring and rapid decision-making. This offers greater responsive­ developing countries: considerations in the context of the problem of sustainable
ness and faster, more accurate decision-making, which is essential in the food production, Sustainability 12 (2020) 5488, [Link]
su12135488.
context of climate change and increasingly variable weather conditions.
[2] T.R.C. Konfo, A.B.P. Chabi, A. Amoussouga Gero, C. Lagnika, F. Avlessi, G. Biaou,
The integration of climate data and soil parameters also enables better C.K.D. Sohounhloue, Recent climate-smart innovations in agrifood to enhance
prediction of crop water requirements, ensuring optimal irrigation producer incomes through sustainable solutions, J. Agric. Food Res. 15 (2024)
100985, [Link]
management.
[3] A. Morchid, R. El Alami, A.A. Raezah, Y. Sabbar, Applications of internet of things
In addition to the prospects already mentioned, the integration of (IoT) and sensors technology to increase food security and agricultural
renewable energy sources, such as solar panels, is a key avenue for sustainability: benefits and challenges, Ain Shams Eng. J. 15 (2024) 102509,
enhancing the system’s energy sustainability. Although this approach [Link]
[4] A. Morchid, M. Marhoun, R. El Alami, B. Boukili, Intelligent detection for
has been discussed, its actual implementation remains to be explored in sustainable agriculture: a review of IoT-based embedded systems, cloud platforms,
future work. The use of solar energy would enable sensors and IoT de­ DL, and ML for plant disease detection, Multimed. Tools Appl. 83 (2024)
vices to be powered autonomously, reducing dependence on traditional 70961–71000, [Link]
[5] A. Morchid, R. El Alami, H. Qjidaa, M. El Alaoui, H. Touati, K. El Khadiri, M.
energy sources and improving system viability in isolated agricultural Ouazzani-Jamil, Fire safety system implementation for controlling and monitoring
environments. In addition, advanced data analysis using artificial in­ a siren in smart farm using gas sensor and flame sensor, in: S. Motahhir, B. Bossoufi
telligence and machine learning technologies will be further developed (Eds.), Digital Technologies and Applications, Springer International Publishing,
Cham, 2022: pp. 733–742. [Link]
to optimize irrigation scheduling. The application of artificial intelli­ [6] L. Kumar, N. Chhogyel, T. Gopalakrishnan, M.K. Hasan, S.L. Jayasinghe, C.
gence algorithms will make it possible to refine the assessment of crop S. Kariyawasam, B.K. Kogo, S. Ratnayake, Chapter 4 - Climate change and future of
water requirements based on the environmental parameters collected, agri-food production, in: R. Bhat (Ed.), Future Foods, Academic Press, 2022,
pp. 49–79, [Link]
thus ensuring more precise and adaptive irrigation. In addition, the
[7] S.H. Muhie, Novel approaches and practices to sustainable agriculture, J. Agric.
system will be extended to other types of crops, soils, and climates. This Food Res. 10 (2022) 100446, [Link]
adaptation enables the system to meet the specific water requirements of [8] A. Morchid, R. Jebabra, R.E. Alami, M. Charqi, B. Boukili, Smart agriculture for
sustainability: the implementation of smart irrigation using real-time embedded
different crops and optimize irrigation based on local conditions,
system technology, in: 2024 4th International Conference on Innovative Research
thereby improving overall efficiency. Finally, particular attention will in Applied Science, Engineering and Technology (IRASET), 2024, pp. 1–6, https://
be paid to data security in the IoT. Future research will focus on [Link]/10.1109/IRASET60544.2024.10548972.
strengthening secure transmission protocols and implementing [9] A. Morchid, R. Jebabra, A. Ismail, H.M. Khalid, R. El Alami, H. Qjidaa, M. Ouazzani
Jamil, IoT-enabled fire detection for sustainable agriculture: a real-time system
advanced encryption mechanisms to ensure the protection of sensitive using flask and embedded technologies, Results Eng. 23 (2024) 102705, https://
information and the integrity of communications between the various [Link]/10.1016/[Link].2024.102705.
system components. Additionaly, plan to focus our future work on [10] A. Morchid, I.G. Muhammad Alblushi, H.M. Khalid, R. El Alami, S.R. Sitaramanan,
S.M. Muyeen, High-technology agriculture system to enhance food security: a
developing a stand-alone website for monitoring farm system data. concept of smart irrigation system using Internet of Things and cloud computing,
These lines of research will not only improve the efficiency and security J. Saudi Soc. Agric. Sci. (2024), [Link]
of water resource management, but also increase the system’s resilience [11] A.D. Boursianis, M.S. Papadopoulou, A. Gotsis, S. Wan, P. Sarigiannidis,
S. Nikolaidis, S.K. Goudos, Smart irrigation system for precision agriculture—The
in the face of future environmental and technological challenges. AREThOU5A IoT Platform, IEEE Sens. J. 21 (2021) 17539–17547, [Link]
10.1109/JSEN.2020.3033526.
CRediT authorship contribution statement [12] M.J. Hoque, M.S. Islam, M. Khaliluzzaman, A fuzzy logic- and internet of things-
based smart irrigation system, Eng. Proc. 58 (2023) 93, [Link]
ecsa-10-16243.
Abdennabi Morchid: Writing – original draft, Visualization, Vali­ [13] M. Benzaouia, B. Hajji, A. Mellit, A. Rabhi, Fuzzy-IoT smart irrigation system for
dation, Supervision, Software, Resources, Project administration, precision scheduling and monitoring, Comput. Electron. Agric. 215 (2023) 108407,
[Link]
Methodology, Investigation, Formal analysis, Data curation, Conceptu­ [14] I. Jdey, Trusted smart irrigation system based on fuzzy IoT and blockchain, in: J.
alization. Zafar Said: Writing – review & editing, Visualization, Vali­ Troya, R. Mirandola, E. Navarro, A. Delgado, S. Segura, G. Ortiz, C. Pautasso, C.
dation, Project administration, Methodology, Funding acquisition, Zirpins, P. Fernández, A. Ruiz-Cortés (Eds.), Service-Oriented Computing – ICSOC
2022 Workshops, Springer Nature Switzerland, Cham, 2023: pp. 154–165. https
Formal analysis, Data curation. Almoataz Y. Abdelaziz: Writing – re­
://[Link]/10.1007/978-3-031-26507-5_13.
view & editing, Visualization, Validation, Investigation, Formal anal­ [15] D. Cavaliere, S. Senatore, V. Loia, Crop health assessment through hierarchical
ysis, Data curation. Pierluigi Siano: Visualization, Validation, fuzzy rule-based status maps, Knowl. Inf. Syst. (2024), [Link]
Resources, Investigation, Formal analysis, Data curation. Hassan Qji­ s10115-024-02180-w.
[16] H. Benyezza, M. Bouhedda, S. Rebouh, Zoning irrigation smart system based on
daa: Writing – review & editing, Visualization, Validation, Supervision, fuzzy control technology and IoT for water and energy saving, J. Clean. Prod. 302
Project administration, Investigation, Formal analysis, Data curation. (2021) 127001, [Link]

13
A. Morchid et al. Smart Agricultural Technology 11 (2025) 100979

[17] V. Thomopoulos, F. Tolis, T.-F. Blounas, D. Tsipianitis, A. Kavga, Application of [31] M. Krstić, G.P. Agnusdei, P.P. Miglietta, S. Tadić, Logistics 4.0 toward circular
fuzzy logic and IoT in a small-scale smart greenhouse system, Smart Agric. economy in the agri-food sector, Sustain. Futures 4 (2022) 100097, [Link]
Technol. 8 (2024) 100446, [Link] org/10.1016/[Link].2022.100097.
[18] A.U. Rehman, Y. Alamoudi, H.M. Khalid, A. Morchid, S.M. Muyeen, A.Y. Abdelaziz, [32] L. Hayo, H. Hasegawa, Enhancing emission reductions in South African agriculture:
Smart agriculture technology: an integrated framework of renewable energy the crucial role of carbon credits in incentivizing climate-smart farming practices,
resources, IoT-based energy management, and precision robotics, Clean. Energy Sustain. Futures 8 (2024) 100260, [Link]
Syst. (2024) 100132, [Link] [33] K. Rostami, L. Salehi, Rural cooperatives social responsibility in promoting
[19] F. Irwanto, U. Hasan, E.S. Lays, N.J. De La Croix, D. Mukanyiligira, L. Sibomana, sustainability-oriented activities in the agricultural sector: nexus of community,
T. Ahmad, IoT and fuzzy logic integration for improved substrate environment enterprise, and government, Sustain. Futures 7 (2024) 100150, [Link]
management in mushroom cultivation, Smart Agric. Technol. 7 (2024) 100427, 10.1016/[Link].2023.100150.
[Link] [34] J. Brinken, F. Behrendt, S. Trojahn, Comparing decarbonization potential of digital
[20] L. Bin, M. Shahzad, H. Khan, M.M. Bashir, A. Ullah, M. Siddique, Sustainable smart and green technologies, Sustain. Futures 6 (2023) 100125, [Link]
agriculture farming for cotton crop: a fuzzy logic rule based methodology, 10.1016/[Link].2023.100125.
Sustainability 15 (2023) 13874, [Link] [35] L. Ruangpan, M. Mahgoub, Y.A. Abebe, Z. Vojinovic, S. Boonya-aroonnet, A.
[21] B. Et-taibi, M.R. Abid, E.-M. Boufounas, A. Morchid, S. Bourhnane, T. Abu Hamed, S. Torres, S. Weesakul, Real time control of nature-based solutions: towards smart
D. Benhaddou, Enhancing water management in smart agriculture: a cloud and solutions and digital twins in Rangsit Area, Thailand, J. Environ. Manag. 344
IoT-Based smart irrigation system, Results Eng. 22 (2024) 102283, [Link] (2023) 118389, [Link]
10.1016/[Link].2024.102283. [36] A. Maroli, V.S. Narwane, B.B. Gardas, Applications of IoT for achieving
[22] M.A. Garcia-Espinal, F.-J. Sanchez-Romero, M. Perez-Sanchez, P.A. Lopez-Jimenez, sustainability in agricultural sector: a comprehensive review, J. Environ. Manag.
Enhancing sustainability in irrigation networks: a multicriteria method for 298 (2021) 113488, [Link]
optimizing flow distribution and reducing environmental impact, Results Eng. [37] D. Narang, J. Madaan, F.T.S. Chan, E. Chungcharoen, Managing open loop water
(2024) 102609, [Link] resource value chain through IoT focused decision and information integration
[23] A.K. Yadav, V. Yadav, H. Malik, R. Khargotra, T. Singh, Design of novel IoT-based (DII) modelling using fuzzy MCDM approach, J. Environ. Manag. 350 (2024)
solar powered PV pumping systems for agricultural applications in diverse climatic 119609, [Link]
zones of India, Results Eng. 23 (2024) 102584, [Link] [38] L. Miasayedava, K. McBride, J.A. Tuhtan, Automated environmental compliance
rineng.2024.102584. monitoring of rivers with IoT and open government data, J. Environ. Manag. 303
[24] S. Al-Naemi, A. Al-Otoom, Smart sustainable greenhouses utilizing microcontroller (2022) 114283, [Link]
and IOT in the GCC countries; energy requirements & economical analyses study [39] M. Qayyum, Y. Zhang, M. Wang, Y. Yu, S. Li, W. Ahmad, S.N. Maodaa, S.R.
for a concept model in the state of Qatar, Results Eng. 17 (2023) 100889, https:// M. Sayed, J. Gan, Advancements in technology and innovation for sustainable
[Link]/10.1016/[Link].2023.100889. agriculture: understanding and mitigating greenhouse gas emissions from
[25] A. Rajagopalan, et al., Empowering power distribution: unleashing the synergy of agricultural soils, J. Environ. Manag. 347 (2023) 119147, [Link]
IoT and cloud computing for sustainable and efficient energy systems, Results Eng. 10.1016/[Link].2023.119147.
21 (2024) 101949, [Link] [40] M.I. Nor Diana, N.A. Zulkepli, L.K. Ern, M.R. Zainol, Factors affecting behavioral
[26] D.A. Verde Romero, E. Villalvazo Laureano, R.O. Jim´enez Betancourt, E. Navarro intentions of farmers in Southeast Asia to technology adoption: a systematic review
Alvarez, An open source IoT edge-computing system for monitoring energy analysis, J. Environ. Manag. 367 (2024) 122045, [Link]
consumption in buildings, Results Eng. 21 (2024) 101875, [Link] jenvman.2024.122045.
10.1016/[Link].2024.101875. [41] A. Johri, S. Bhadula, S. Sharma, A. Shankar Shukla, Assessment of factors affecting
[27] A. Rabak, et al., Sensor system for precision agriculture smart watering can, Results implementation of IoT based smart skin monitoring systems, Technol. Soc. 68
Eng. 19 (2023) 101297, [Link] (2022) 101908, [Link]
[28] N. Jihani, M.N. Kabbaj, M. Benbrahim, Kalman filter based sensor fault detection in [42] M.H. Ronaghi, A. Forouharfar, A contextualized study of the usage of the Internet
wireless sensor network for smart irrigation, Results Eng. 20 (2023) 101395, of things (IoTs) in smart farming in a typical Middle Eastern country within the
[Link] context of Unified Theory of Acceptance and Use of Technology model (UTAUT),
[29] S. Phuyal, D. Bista, R. Bista, Challenges, opportunities and future directions of Technol. Soc. 63 (2020) 101415, [Link]
smart manufacturing: a state of art review, Sustain. Futures 2 (2020) 100023, [43] K. Jakobsen, M. Mikalsen, G. Lilleng, A literature review of smart technology
[Link] domains with implications for research on smart rural communities, Technol. Soc.
[30] A. Rejeb, S. Zailani, K. Rejeb, H. Treiblmaier, J.G. Keogh, Modeling enablers for 75 (2023) 102397, [Link]
blockchain adoption in the circular economy, Sustain. Futures 4 (2022) 100095, [44] C. Giua, V.C. Materia, L. Camanzi, Smart farming technologies adoption: which
[Link] factors play a role in the digital transition? Technol. Soc. 68 (2022) 101869
[Link]

14