International Journal of Advanced Technology and Engineering Exploration, Vol 10(109)

ISSN (Print): 2394-5443 ISSN (Online): 2394-7454

Research Article
http://dx.doi.org/10.19101/IJATEE.2023.10102045

Smart fertigation system with mobile application and fuzzy logic optimization
Nurul Anis Zulaikha Izahar1, Mohd Noor Derahman1*, Mohamad Afendee Mohamed2 and Imas
Sukaesih Sitanggang3
Department of Communication Technology and Network, Universiti Putra, Malaysia1
Department of Informatics and Computing, Universiti Sultan Zainal Abidin, Malaysia 2
Department of Computer Science, IPB University, Bogor, Indonesia3

Received: 01-August-2023; Revised: 13-December-2023; Accepted: 16-December-2023

©2023 Nurul Anis Zulaikha Izahar et al. This is an open access article distributed under the Creative Commons Attribution (CC
BY) License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is
properly cited.

Abstract
Precision farming plays a pivotal role in addressing the dual challenge of increasing crop productivity and reducing the
environmental impact of agriculture. This arises from the critical need to elevate crop productivity to meet the increasing
global demand for food, while actively confronting the economic and environmental consequences linked to suboptimal
conventional farming practices. Conventional agriculture consistently grapples with inefficiencies in resource
management, marked by the impractical and excessive utilization of fertilizers and water. This not only results in
mounting production costs, but also gives rise to substantial threats to the environment, encompassing soil degradation,
water pollution, and the loss of biodiversity. Thus, this study introduces a smart fertigation system, incorporating internet
of things (IoT) technology and fuzzy logic optimization, specifically designed for large-scale bird's eye chili production.
The innovative system integrates IoT technology with fuzzy logic to fine-tune fertilization and irrigation processes.
Notably, a fuzzy inference system, implemented using MATLAB and Arduino UNO, dynamically optimizes nutrient
delivery according to the growth stage of the chili plants. This sophisticated approach ensures that fertilization is precisely
tailored to the specific needs of the crops at each developmental phase. Additionally, the development of the C-farm
mobile application empowers farmers with remote monitoring capabilities, enabling them to oversee and manage the
system from anywhere. This mobile application provides real-time insights into the smart fertigation system , granting
farmers unprecedented control over their agricultural operations. Our findings also highlight the efficacy of fuzzy logic
in enhancing the precision of automated fertigation systems. By dynamically adjusting nutrient delivery in response to the
nuanced growth stages of chili plants, our system demonstrates its adaptability and responsiveness, resulting in optimized
resource utilization and improved crop outcomes. This innovative integration of technology not only holds promise for
large-scale crop production but also addresses the pressing issues of water and fertilizer waste in contemporary
agriculture. Moving towards a more sustainable and efficient agricultural paradigm, the smart fertigation system,
featuring fuzzy logic optimization and remote monitoring capabilities, stands as a beacon of progress in the quest for
precision farming.

Keywords
Smart fertigation, Precision agriculture, Fuzzy logic, Arduino UNO, Bird’s eye chili.

1.Introduction With the projected rise in food demand estimated to

In Malaysia, the agricultural sector plays a vital role reach 59 to 98 percent by 2050 [1] farmers will need
as it provides food, which is the most fundamental to enhance crop production either by adding more
human necessity. Fertigation is a crucial technique agricultural areas or improving productivity on
used to improve crop yields by combining current lands. However, traditional agricultural
fertilization with an irrigation system. It is more methods are insufficient for producing substantial
efficient than traditional fertilization, reduces soil amounts of crops and require a significant amount of
erosion and water consumption, and regulates the human labor, time, and money.
release of fertilizer.
Recent decades have seen significant technological
advancements in agriculture, making it more
industrialized and technology driven. The internet of
*Author for correspondence things (IoT) and mobile applications [24] are now
1580
 International Journal of Advanced Technology and Engineering Exploration, Vol 10(109)

being used in fertigation systems to reduce labor solution that aligns with sustainable and efficient
costs and time. These advancements aim to improve farming practices.
agricultural production in both quality and quantity,
including the production of bird's eye chilies, which The structure of this paper is organized as follows:
is a popular spice in traditional Southeast Asian Section 2 provides an in-depth review of the relevant
cuisine, particularly in Malaysia, Indonesia, and literature, contextualizing the study within the
Thailand [58]. However, despite the presence of existing body of knowledge. Section 3 offers a
commercial bird's eye chili growers in Malaysia, the detailed presentation of the proposed methodology
country still imports a significant amount from other and system design. The results and analyses
countries due to limited local production. Addressing emanating from the study are presented in Section 4.
the challenge of achieving mass production of bird's Section 5 synthesizes the findings and provides a
eye chilies within a technological and industrial discussion. The final section, Section 6, concludes
framework requires a nuanced approach. The fusion the study, summarizing the key takeaways and
of IoT technology, mobile applications, and advanced offering insights into potential future research
agricultural practices, as proposed in this study, directions.
presents a potential solution to this predicament. By
harnessing the power of technology-driven precision 2.Literature review
farming, the system aims to create an environment This study is of four significant substantial for their
where bird's eye chili plants can flourish on a distinct purposes. It starts with the development of an
substantial scale, contributing to the reduction of IoT-based automation fertigation system, mainly to
import dependency and bolstering local production optimize the delivery of water, and nutrients to the
capabilities. plants besides others. The second component
involves the incorporation of AI, enhancing the
A smart fertigation system was proposed in this paper system's decision-making capabilities. The third
with mobile application and fuzzy logic optimization component centers on the practical application of the
designed to address the challenges faced by farmers system, where users can interact with and monitor the
in traditional agricultural practices. The use of fuzzy automation process. And lastly, the bird's eye chili
logic artificial intelligence (AI) in such a system plants care, that focuses on ensuring their growth
allows for accurate determination of water and throughout the process.
fertilizer distribution rate based on data collected
from the embedded sensors. To ensure efficient water 2.1IoT-based automation fertigation system
and fertilizer usage, a mobile application has been The integration of IoT technology into fertigation
developed to display data from the sensors and systems holds great potential for revolutionizing farm
monitor water and fertilizer usage. By using this management and decision-making, thereby assisting
system, farmers can save time and reduce labor costs, farmers in optimizing crop production and resource
while also increasing productivity and efficiency. utilization [9]. However, it's important to
Thus, the main objective of this study is to design an acknowledge the limitations surrounding the adoption
IoT-based automation fertigation system, develop a of IoT technology, as not all farmers have the same
mobile application that can be used to monitor the access to advanced technology due to various reasons
system, and implement fuzzy logic in decision- such as affordability, connectivity, and technical
making processes to increase the accuracy on volume skills. This limitation may delay the widespread
of water and fertilizer that the plant needs based on adoption of such systems, and it's important to
the plant growth phase (root, leafy & fruit phase). consider the digital divide and accessibility issues.
The scope of this study primarily focuses on the
production of the bird’s eye chili plant for large-scale In many existing fertigation systems, a common
agricultural businesses. approach involves the utilization of at least two
sensors as input parameters. For example, a study by
The contribution of this study lies in its holistic Ahmed et al. [10] incorporated soil moisture and pH
approach to optimize crop production through the sensors as input parameters in their automated
integration of IoT technology and fuzzy logic fertigation system. While this approach is effective
optimization. By addressing key challenges in for many crops, it may not be universally applicable,
agriculture, such as resource wastage and as different crops have varying requirements and
environmental impact, the study offers a potential environmental conditions. The system's reliance on
preset values for pH and electrical conductivity (EC)
1581
Nurul Anis Zulaikha Izahar et al.

adjustments may not account for all potential generalizable for other crops or production systems.
variations in soil types and crop species, presenting a Meanwhile, [15] proposed smart IoT system for chili
limitation. production using long range (LoRa) technology. The
system uses IoT devices such as sensors, pumps, and
Another intelligent fertigation system developed by valves to monitor and automate the operations of
Ruan et al. [11] relied on pH and EC sensors. growing chili. The system can be used for real-time
However, a limitation here is that these sensors may data collection, monitoring, and analysis of various
need to be maintained and calibrated regularly. This parameters such as soil moisture, temperature, and
regular maintenance and calibration can be a burden humidity. This is believed to assist farmers in
for farmers, especially for those with limited optimizing their production processes, reduce costs,
technical expertise. Furthermore, these sensors may and increase yields. However, it is worth to note the
not address the full spectrum of nutrient requirements economic feasibility and technical limitations of
for various crops, leaving room for refinement. these systems before implementing them on a large
scale farming system.
On the other hand, Joseph et al. [12] picked a
streamlined approach. It is done by employing a Aside from that, IoT-based solutions can increase
single soil moisture sensor in their automation agricultural yields and automate a variety of farming
fertigation system to detect soil moisture content. chores by using smart devices and digital technology
This approach may simplify user interaction. [16], using sensors, communication technology,
Unfortunately, it can also limit the system's Arduino devices, and other devices to gather data in
adaptability. It depends on user input for fertilizer- real-time and regulate variables such as temperature,
related decisions, which may not align with the goal humidity, light, and soil moisture. They work well for
of reducing human intervention. Moreover, the accurate irrigation and for automating tasks like
success of such a system depends on the accuracy of harvesting, pest control, fertilisation, and caring for
the soil moisture sensor, and any calibration issues livestock. Consequently, it benefits farmers by
can lead to imprecise decisions, presenting a reducing expenses while simultaneously increasing
limitation. yields. Before implementing these systems broadly,
it's crucial to consider their costs and technical
Meanwhile, [13] proposed a solar fertigation system difficulties, as the research to date does not provide
that uses solar power as its energy source. The this information, and further research is required to
system aims to determine the optimal water and ensure that they are useful and feasible for farmers.
nutrient requirements of plants based on agronomic
models and sensor data. The system then calibrates In [17], the system uses IoT devices like sensors and
the fertilizers’ levels to be used during irrigation and Wi-Fi modules to watch over soil moisture in
fertigation, as well as the timing of irrigation fertigation systems, allowing real-time data analysis.
scheduling. However, the proposed system may not This helps farmers improve their fertigation methods,
be practical. This is due to the complex data, thus it cut costs, and increase crop yields. Research papers
needs for irrigation and fertigation processes. The explain how IoT technology can revolutionize
decision-making process relies on intelligent analysis farming by remotely managing soil moisture levels in
of various factors such as soil sensors, weather fertigation. They suggest that IoT-based systems can
conditions, crop type, and IoT analysis. Moreover, transform food production and resource management
the system also collects data on the temperature and in agriculture. However, this system does not
humidity of both the soil and atmosphere, therefore, consider the real-time monitoring that can assist
increasing the complexity of the data. farmers [17]. However, the monitoring is restricted to
the Blynk application, which lacks extensive
Long range wide area network (LoRaWAN ) based customization. Blynk provides a set of predefined
internet of thing s (IoT) system for precision widgets, but the limited customization options can
irrigation in plasticulture fresh-market tomato. pose challenges when trying to design highly
Sensors are utilized for real-time data monitoring and specialized or unique IoT interfaces.
automating irrigation systems [14]. This is mainly to
save water in crop production and can be used as a To summarize, the integration of IoT technology into
precision management tool for fresh-market tomato fertigation systems is recognized for its potential to
production. However, this system focuses only on revolutionize farm management, optimize crop
fresh-market tomato production and may not be production, and enhance resource utilization.
1582
 International Journal of Advanced Technology and Engineering Exploration, Vol 10(109)

However, the adoption of IoT faces challenges, flow, acid solution, and alkali solution. This shows
including accessibility issues due to factors such as how versatile is the smart technologies in agriculture,
affordability, connectivity, and technical skills, which significantly increasing crop yields.
may hinder widespread implementation. Existing
fertigation systems often rely on multiple sensors, Similarly, [19] applied fuzzy logic in their fertigation
each with its limitations, ranging from applicability system. By doing so, the intelligent irrigation
across different crops to the need for regular decisions are made based on data from soil moisture,
maintenance and calibration. Streamlining temperature, and humidity sensors. This approach has
approaches, such as utilizing a single soil moisture been proven to enhance irrigation. Furthermore, [20]
sensor, may simplify user interaction but could limit integrated both ML and deep learning in their
system adaptability and depend on user input for fertigation system. The accuracy of different ML
fertilizer decisions. Innovative proposals, like the algorithms are compared, such as naive bayes,
solar fertigation system, present challenges due to the logistic regression, support vector machine, decision
complexity of data requirements for irrigation tree classifier, bagging classifier, random forest
processes. Specific IoT systems targeting fresh- classifier, adaboost classifier, gradient boosting
market tomato or chili production show promise but classifier, xgboost classifier, and k-nearest neighbor.
underscore the importance of considering economic From this study, it can be concluded that efficiency
feasibility and technical limitations before large-scale can be increased and the process is simplified when
implementation. Overall, while IoT-based solutions ML are used in agricultural application. In
hold potential to enhance agriculture by automating conclusion, the implementation of AI in fertigation
tasks and remotely managing soil moisture, careful systems can greatly benefit agriculture, leading to
consideration of costs, technical challenges, and more efficient and cost-effective crop production.
customization options is crucial for ensuring The choice of algorithm depends on the specific
practicality and feasibility for farmers. requirements of the system, and it is essential to
compare and evaluate the different approaches before
2.2Implementation of AI into automation implementing them.
fertigation system
In the realm of smart farming, numerous studies have Proposed an automated irrigation and fertilization
explored a diverse range of applications for AI and system that uses fuzzy logic to control sensors [18].
machine learning (ML). These investigations span Their system has two input membership functions for
from precision agriculture and crop monitoring to soil moisture and pH value and three output
automated livestock management and predictive membership functions for water flow, acid solution,
analytics for disease detection. and alkali solution. With the help of smart
technologies, agriculture is now capable of producing
In [13], it emphasizes the pivotal role of ML significantly more crops than before. Similarly, used
algorithms in tasks such as crop selection and fuzzy logic in their fertigation system to make
management. These algorithms leverage data on soil intelligent irrigation decisions based on data from soil
quality, compatibility classification, and other moisture, temperature, and humidity sensors. The
factors, enabling informed decisions about crop fuzzy logic-based fertigation method increased
choices. irrigation application efficiency by 50 percent and
produced a larger growth gradient and chili
In a related review, [14] examines the automation and production compared to traditional methods. It
digitization of agriculture and underscores the propose an automated system that uses an artificial
benefits of employing technology for sustainable neural network (ANN) algorithm for intelligent
crop production. The references suggest that AI and decision-making. ANN is a popular algorithm for
IoT-based systems have the potential to optimize forecasting and making predictions based on parallel
water and nutrient management, reduce labor costs, reasoning, which simulates the human brain's
and enhance crop production. functioning.

Moreover, [18] proposed an automated irrigation and Furthermore, [20] integrated both ML and deep
fertilization system that utilizes fuzzy logic to control learning in their fertigation system. They compared
sensors. Their system incorporates two input the accuracy of different ML algorithms, such as
membership functions for soil moisture and pH value naive bayes, logistic regression, support vector
and three output membership functions for water machine, decision tree classifier, bagging classifier,

1583
Nurul Anis Zulaikha Izahar et al.

random forest classifier, AdaBoost classifier, gradient mobile application. In our opinion, to consistently
boosting vlassifier, XGBoost classifier, and K-nearest supply plants with enough nutrients and moisture,
neighbor. Their study concluded that using ML in proper monitoring and control of irrigation and
agricultural applications can significantly increase fertilization systems are essential.
efficiency and simplify the process. In conclusion,
the implementation of AI in fertigation systems can Nevertheless, the automated fertigation system
greatly benefit agriculture, leading to more efficient proposed by [9] is using the web as the user interface.
and cost-effective crop production. The choice of Farmers can utilize the web as an interface to interact
algorithm depends on the specific requirements of the with the system. The farmers can access the web
system, and it is essential to compare and evaluate using mobile devices such as phones and tablets to
the different approaches before implementing them. monitor the system, change the fertigation routine,
and set the combination formula of EC. However,
In the realm of smart farming, diverse studies explore mobile applications are the preferred choice for
the applications of AI and ML. These investigations farmers to access and monitor their fertigation
range from crop management to disease detection, systems while working in the field. They are user-
highlighting the pivotal role of ML algorithms in friendly, intuitive and provide convenience on site. If
informed decision-making, as seen in [13]. the farmers wanted to make some adjustments to
Additionally, [14] emphasizes the benefits of their systems, they can just easily do it through
employing AI and IoT-based systems for optimizing mobile apps. They can also stay connected to receive
water and nutrient management. Noteworthy real-time alerts and notifications as they go around
contributions from [19, 20] showcase the versatility their farms.
of smart technologies, utilizing fuzzy logic and
integrating ML and deep learning to significantly On the other hand, [23] simplifies communication by
enhance irrigation decisions and increase overall crop incorporating Telegram bot applications for
production efficiency. The proposed automated automated mobile messaging. Additionally, web-
system in [18] further exemplifies the potential of based applications are hosted in the cloud, enhancing
fuzzy logic and artificial neural networks (ANN) for accessibility. Nevertheless, it's crucial to
intelligent decision-making, underscoring the acknowledge a limitation in this context—the
transformative impact of these technologies on absence of a dedicated mobile interface for real-time
agriculture. Overall, the implementation of AI and access to the fertigation system in the dynamic
ML in fertigation systems holds significant promise farming environment.
for advancing efficiency and cost-effective crop
production, contingent on careful consideration of In contrast, [9] utilizes the web as its user interface
specific system requirements and algorithm choices. for the automated fertigation system. By using
mobile devices such as phones and tablets, farmers
2.3Application usage in automation fertigation are able to interact with the system through web
system access. This allows them to monitor the system,
The study conducted by [12] can be conveniently adjust the fertigation routine, and customize the EC
monitored via a mobile application. They use Blynk combination formula. However, it's noteworthy that
mobile application as the user interface. Blynk is an mobile applications remain the preferred choice for
Arduino-controlling mobile application that runs on farmers working in the field. Mobile apps offer user-
the internet. By using the Blynk application, they can friendliness, intuitive operation, and on-site
input nitrogen, phosphorus and potassium (NPK) convenience. They enable farmers to make necessary
requirements, fertilizer concentrations, soil moisture adjustments to their systems effortlessly while
content status, start or stop commands, and irrigation staying connected to receive real-time alerts and
mode. Also, they can use the application to monitor notifications as they navigate their farms. This level
the plant's condition without having to physically of mobility and real-time monitoring aligns well with
visit the farm. But their application is not good the dynamic nature of farming, where quick
enough since they still need to control the usage of responses to changing conditions are essential.
the fertilizers in the application manually and it does
not meet the purpose of reducing human labour. In short, the examination of how farmers interact
Like [12, 18, 21, 22] also used mobile applications to with automated fertigation systems highlights two
monitor their proposed system. The data is stored by primary methods: through mobile applications and
the system and display on the farmers’ device using web interfaces. In the case of the blynk mobile
1584
 International Journal of Advanced Technology and Engineering Exploration, Vol 10(109)

application used in one study, it provides convenient humidity, temperature, and ultrasonic sensors. Then,
monitoring capabilities, allowing users to input the data is transmitted to a website for user access.
crucial parameters. However, a notable drawback Sensors are devices that detect changes in physical
emerges as manual control of fertilizer usage is still conditions like pressure, light, temperature, and
required. Similarly, other studies favor mobile more.
applications, emphasizing the importance of vigilant
monitoring for maintaining consistent nutrient They are small but productive chili plants [27]. Bird’s
supply. In contrast, another study opts for a web- eye chili is a popular spice all around the world and
based interface, recognizing its accessibility believed to be originated in Asia. They are frequently
advantages while acknowledging the prevailing used in cooking and salads throughout India and
preference among farmers for mobile applications Asia, particularly. The bird’s eye chili is a tiny plant
due to their on-site convenience and real-time with small fruits. The pods are approximately 5cm
monitoring. Introducing a Telegram bot application long and 1cm in diameter. According to [27] during
simplifies communication in a different study but the germ phase, seed germination might be harmed
introduces a limitation by lacking a dedicated mobile by pre-fertilized growth substrate. For the initial days
interface for immediate access. This analysis after germination, the chili seed has adequate
underscores the practical considerations in choosing compounds. We start to fertilize the plants after they
user interfaces, emphasizing the need for seamless have developed more leaves. The seedling dosage
automation and on-site convenience in dynamic must be kept as low as possible. High nutrient salt
farming environments. It also suggests avenues for concentrations are vulnerable to young plants.
future research to enhance user experiences and
simplify smart farming technologies for farmers. During the first four months of growth, we must use a
fertilizer with a higher nitrogen (N) concentration
2.4Bird Eye’s chili plant care [27]. The nitrogen content can be shown by the NPK
The bird's eye chili is scientifically known as value. When the NPK value is 3-1-4, it indicates that
Capsicum frutescens [24]. Caring and measuring chili the fertilizer contains 3% nitrogen, 1% phosphorus,
plant in terms of research can be seen in [25]. It and 4% potassium. Nitrogen is necessary for plant
proposed a more efficient way to measure the height growth. Root growth is aided by phosphorus, while
and width of these plants in the field, which is flowering and fruit creation are aided by potassium. It
traditionally done manually. This innovative is essential to keep the soil damp. Dry soil and the
approach uses special devices like sensors, relays, usage of liquid fertilizer may cause damage to the
and Wi-Fi to take the measurements automatically. It roots.
also allows for the real-time collection and analysis
of soil moisture data. The research papers discuss the The aim for this study is to measure soil moisture, air
development and application of systems that utilize humidity, and temperature to determine the best
image processing to monitor plant growth in the field times to water chili plants using these sensors, as it
from a distance. This technology is seen as promising eases the farmers’ work to monitor and care for them.
for farmers, but it's important to note that the study's Through Blackbox testing, the system's effectiveness
scale was relatively small, and it’s suitability for is confirmed, showing that it meets expectations and
large-scale agriculture may require further functions well. Despite that, it's important to note that
exploration. this study is done in a controlled environment and
may not fit for large-scale plant production.
To care for chili plants, it does not only involve
fertilization; proper watering is also an important To summarize, bird eye's chili plant care delves into
factor to be taken care of. Often, watering can be the cultivation practices of capsicum frutescens, a
irregular, and caregivers may overlook factors like globally popular spice native to Asia. Most of the
temperature and soil moisture [26]. In addition, due studies provides practical insights, highlighting the
to other commitments, caregivers may find it hard importance of nitrogen-focused fertilization and
and challenging to maintain a consistent watering specific NPK values during the initial growth phase.
schedule, thus, potentially leading to suboptimal chili Beyond fertilization, the literature addresses
plant growth or even causes death. To solve this, challenges in watering practices, emphasizing the
ideal watering times are utilized using Arduino and need for consistency and the potential consequences
other various sensors. From this research, data are of irregular watering. The integration of advanced
collected from sensors, including soil moisture, air technologies, such as sensors and Wi-Fi, for

1585
Nurul Anis Zulaikha Izahar et al.

automated measurement and real-time data collection of the IoT-based fertigation system was the primary
in chili plant care is explored. While recognizing the focus. The second iteration was dedicated to the
effectiveness of these technologies through testing, development of the C-Farm mobile application,
the literature acknowledges the need for further which was created using Flutter [2931]. The third
research, especially in scaling up image processing iteration involved the implementation of fuzzy logic
systems for plant growth monitoring and assessing into the system. Each iteration consisted of six
the adaptability of Arduino-based watering systems phases, and Figure 1 provides an overview of the
in diverse environments. The study's implications mobile application development process for an
include practical recommendations for farmers and automated fertigation system using fuzzy logic [32].
stress the continual importance of innovation in smart However, for the purpose of this study, we will
agricultural practices. primarily focus on the fuzzy logic aspect of the
development since it is the main idea behind the
3.Methods development of this product. Figure 2 shows an
The smart fertigation system with mobile application architectural overview of the proposed mobile
and fuzzy logic optimization was developed using an application automated fertigation system employing
agile model [4, 27, 28] that makes iterative fuzzy logic as a decision-making engine. It consists
development, fast feedback, and adjusting at every of five elements which include the hardware setup of
level of the product cycle their priority. The the fertigation system, Arduino for data collection,
development cycle consists of three iterations. Each fuzzy logic algorithm for data processing and
iteration focuses on a specific aspect of the product. decision making, cloud firestore database, and C-
During the first iteration, the hardware development farm application.

Connect the Arduino

Construct IoT-based Develope mobile Link database with Implement fuzzy
to cloud firestore
fertigation system application using the mobile logic into the
database from googe
using Arduino Flutter application system
firebase

Figure 1 Development of the fuzzy logic based automation

Figure 2 Fuzzy logic based automation Fertigation system with mobile application

1586
 International Journal of Advanced Technology and Engineering Exploration, Vol 10(109)

Sensors has been used to read data twice a day at depending on the plant's growth phase and nutrient
predetermined times: 8:00 AM and 5:00 PM. The requirements, the nitrogen (N), phosphorous (P), and
collected data will be processed by fuzzy logic potassium (K) pump will be turned on for a duration
algorithm to produce outputs for irrigation and determined by the output of the fuzzy logic. and
fertilization purposes. The system will then carry out lastly, if the plant does not require irrigation or
three distinct procedures simultaneously. The initial fertilization, the system continues collecting data
step involves checking the state of irrigation; from the sensors as usual. The data collected by the
followed by the watering status for the plant, when system will be stored in the cloud firestore, and this
needed, the water pump is activated by the system for data will be used in the third procedure to display
a predetermined duration determined by fuzzy logic information in the C-farm mobile application. The
output. The next step involves the fertilization status; complete flow of the system can be seen in Figure 3.

Figure 3 Flowchart for the proposed automated Fertigation system

3.1Hardware and software overview Hardware Software

A well design of the hardware setup ensures the Pump
reliability of the system and choosing suitable Water Pipe NA
hardware and software is essential to build a good 9V Battery NA
and successful system. Table 1 enumerates the 9V Battery Clip NA
hardware and software components employed in the 8-Ways Relay Module NA
development of the Smart fertigation system with MAX485 TTL to RS485 NA
mobile application and fuzzy logic optimization. Module
Soil Moisture Sensor NA
Table 1 Hardware and software DHT22 Sensor NA
Hardware Software NPK Sensor NA
Arduino UNO Arduino IDE
NodeMCU Fritzing 3.2Fuzzy irrigation system
Breadboard MATLAB The foundation of our automated fertigation system
Jumper Wire Google Firebase relies on fuzzy set theory. Fuzzy inference system is
Micro Submersible Water Visual Studio Code a process of mapping from a given input to an output,
1587
Nurul Anis Zulaikha Izahar et al.

using the theory of fuzzy sets. This system used to the respective fuzzy set theory prior producing
Mamdani method [33, 34] for the fuzzy inference fuzzy output for the irrigation system.
system. Figure 4 shows the mapping of fuzzy inputs

Figure 4 Fuzzy irrigation system

3.3Crisp input and fuzzification functions for the soil moisture, humidity, and
The system, which comes with a Mobile Application, temperature sensors were generated using MATLAB
takes in three inputs: soil moisture, temperature, and [3638] and are illustrated in Figure 5, 6, and 7
humidity. To define these inputs, the system uses respectively. Meanwhile, Table 2 provides a
three linguistic regions: 'DRY', 'MEDIUM', and summary of the input variable for the sensors using
'WET' for soil moisture, 'LOW', 'MEDIUM', and the Mamdani method.
'HIGH' for humidity, and 'COLD', 'WARM', and
'HOT' for temperature [18, 21, 35]. The membership

Figure 5 Membership function for soil moisture sensor

1588
 International Journal of Advanced Technology and Engineering Exploration, Vol 10(109)

Figure 6 Membership function for humidity sensor

Figure 7 Membership function for temperature sensor

Table 2 Output description of each pump

Sensor Name of MF MF Type Range Parameters
Soil Moisture Dry Triangular (0, 100) (0, 20.5, 40)
Medium Triangular (0, 100) (41, 55, 70)
Wet Triangular (0, 100) (71, 84.5, 100)
Humidity Low Triangular (0, 100) (0, 17.5, 30)
Medium Triangular (0, 100) (31, 50.2, 70)
High Triangular (0, 100) (71, 85, 100)
Temperature Cold Triangular (0, 100) (0, 15, 30)
Warm Triangular (0, 100) (31, 35, 40)
Hot Triangular (0, 100) (41, 45, 100)

3.4Knowledge based rule 3.5Defuzzification and crisp output

In order to support approximative reasoning, When a crisp input is fuzzified to create a fuzzy
knowledge base systems use fuzzy knowledge bases value, this process is called defuzzification. The
that rely on fuzzy set theory to represent facts, rules, fuzzy output as the result of fuzzy inference engine is
and linguistic variables. In this system, each of the converted into a crisp value that can be forwarded to
three input variables - soil moisture, temperature, and the controller. The resulting fuzzy results cannot be
humidity - has three linguistic variables. This means applied to any application where decisions must only
that there are 27 rules for the input parameters, which be based on crisp values and the controller only
can be seen in Figure 8. comprehends crisp output. As a result, the fuzzy
output needs to be changed into a crisp value. Figure
9 shows the membership function for water pump
1589
Nurul Anis Zulaikha Izahar et al.

rate from MATLAB. The output for this membership pump. The water pump rate is measured in unit
function is in unit second. Using Mamdani, method, second.
Table 3 exhibits the state of output variable for water

Figure 8 Irrigation IF-THEN rules

Figure 9 Membership function for water pump rate

Table 3 Description of water pump

Sensor Name of MF MF Type Range Parameters
Water Pump Short Triangular (0, 15) (-5, 0, 5)
Medium Triangular (0, 15) (5, 10, 15)

3.6Fuzzy fertilization system 3.7Crisp input and fuzzification

Figure 10 shows how the fuzzy inputs are mapped to The system receives nitrogen, phosphorus, and
fuzzy output via the use of fuzzy set theory for the potassium as three crisp inputs. This system defines
fertilization system. Depending on the stage of all the input variables according to three linguistic
plants’ growth, the fuzzy logic for fertilization is regions: "LOW", "MEDIUM", and "HIGH". The
different. This is due to the different NPK fertilizer parameter values in the input description below vary
ratios during each stage. according to the stage of plant growth. This resulted
from the various NPK fertilizer ratios used at each
phase.
1590
 International Journal of Advanced Technology and Engineering Exploration, Vol 10(109)

3.8Root phase phosphorous, and potassium sensor for the root phase
The root growth is aided by phosphorus (P). Thus, is displayed in Figure 11, 12 and 13 respectively.
the NPK fertilizer ratios for the root phase are 1-2-1. Based on Mamdani method, Table 4 exhibits the state
The membership function of the nitrogen, of the input variable of phosphorous sensor.

Figure 10 Fertilization fuzzy inference system

Figure 11Membership function for nitrogen sensor

1591
Nurul Anis Zulaikha Izahar et al.

Figure 12 Membership function for phosphorous sensor

Figure 13 Membership function for potassium sensor

Table 4 Input variable sensor using mamdani method for fertilization

Sensor Name of MF MF Type Range Parameters
Nitrogen Low Triangular (0, 125) (0, 20.83, 41.66)
Medium Triangular (0, 125) (41.66, 62.5, 83.34)
High Triangular (0, 125) (83.34, 104.2, 125)
Phosphorous Low Triangular (0, 250) (0, 42.16, 83.33)
Medium Triangular (0, 250) (83.33, 125.5, 166.7)
High Triangular (0, 250) (166.7, 208.8, 250)
Potassium Low Triangular (0, 125) (0, 20.83, 41.66)
Medium Triangular (0, 125) (41.66, 62.5, 83.34)
High Triangular (0, 125) (83.34, 104.2, 125)

3.9Leafy phase and fruit phase fertilization system, the fuzzy logic rules are shown
Nitrogen (N) ensures good plant growth. Hence, the in Figure 14. Similar to the irrigation system, the
NPK fertilizer ratios for the leafy phase are 2-1-1. rules were created using the Mamdani technique, and
Meanwhile in the Fruit Phase, Potassium (K) aids in the simulation was performed using MATLAB. The
the growth of flowers and fruits. The NPK fertilizer fuzzy inference will collect and analyze the crisp
ratios are therefore 1-1-2 for the fruit phase. Thus, the input data, such as nitrogen, phosphorous, and
membership function and input description will be potassium, using an interference rule base.
regulated to the stated ratio accordingly. Furthermore, these IF-THEN rules were applied for
all phases of plant growth. Figure 15 illustrates the
3.10Knowledge base rule fuzzy logic controller of the fertilization system using
There are three different linguistic domains for each Mamdani method. Also, the controller was the same
of the system's three input variables. Therefore, there for all phases of plant growth.
are 33 = 27 rules for the input parameters. For the

1592
 International Journal of Advanced Technology and Engineering Exploration, Vol 10(109)

Figure 14 Fertilization IF-THEN rules

Figure 15 Fuzzy logic controller of the fertilization system using mamdani method

3.11Defuzzification and crisp output function's parameters were uniform for all phases of
Just like the defuzzification process in irrigation plant growth. Figures 16, 17, and 18 illustrate the
system, the crisp value supposedly to be fed to the membership function for nitrogen, phosphorus, and
controller is created by defuzzifying the fuzzy output potassium pump rate respectively, while Table 5 lists
of the fuzzy inference engine. The membership the input variables for each of them.

Figure 16 Membership function for nitrogen pump rate

1593
Nurul Anis Zulaikha Izahar et al.

Figure 17 Membership function for phosphorous pump rate

Figure 18 Membership function for potassium pump rate

Table 5 Input variable of using mamdani method for nitrogen, phosphorus and potassium pump
Sensor Name of MF MF Type Range Parameters
Nitrogen Pump Short Triangular (0, 10) (-3.33, 0, 3.33)
Medium Triangular (0, 10) (3.33, 6.67, 10)
Phosphorous Pump Short Triangular (0, 10) (-3.33, 0, 3.33)
Medium Triangular (0, 10) (3.33, 6.67, 10)
Potassium Pump Short Triangular (0, 10) (-3.33, 0, 3.33)
Medium Triangular (0, 10) (3.33, 6.67, 10)

3.12Use case diagram of C-farm application application programming interface (API) and details
The use case diagram for the C-Farm mobile about the NPK fertilizer ratios based on the stage of
application is shown in Figure 19. A use case plant growth. This section discusses fuzzy reasoning
diagram can be used to summarize details about the rules and is divided into two, one for irrigation and
users and the fertigation system. From the Figure, the another for fertilization. The fuzzy reasoning rule is
sensor data, water use, and NPK fertilizer usage are appropriate for evaluating the precision of the
all presented in the C-Farm application, allowing membership function that was completed within
farmers to monitor their plants. All the data will be earlier stage. Some input is necessary to be supplied
retrieved from the cloud firestore database [39]. The for assessing fuzzy logic reasoning module.
farmer has access to monthly water and NPK 3.13Irrigation
fertilizer usage as well as previous sensor data in Figure 20 shows the fuzzy reasoning rule table for
addition to the most recent data. In the C-Farm the irrigation system when including the value of
application, the farmers may also view weather input using the Mamdani method. The output value,
forecast data that is retrieved from the weather the water pump rate is produced by adding an input
1594
 International Journal of Advanced Technology and Engineering Exploration, Vol 10(109)

value of 63.4 for soil moisture, a value of 56.9 from these usability assessments and user feedback
Celsius for humidity, and a value of 29.7 Celsius for highlighted several positive aspects of the smart
temperature. A crisp output is generated at a water fertigation system with mobile application and fuzzy
pump rate of 2.44 seconds based on these provided logic optimization. Users appreciated the system's
inputs. intuitive design, which facilitated seamless
3.14Fertilization navigation and task completion. The real-time data
Figure 21 demonstrates the fuzzy reasoning rule table presentation, including soil moisture, temperature,
for the fertilization system when including the value and humidity, was deemed informative and beneficial
of input using the Mamdani method. This fuzzy for making informed decisions regarding irrigation
reasoning rule table was for the root phase and used and fertilization. Despite that, there are some areas
1-2-1 NPK fertilizer ratios. Combining input values that need to be improved according to the feedback.
of 78 for nitrogen, 65 for phosphorus, and 59 for Some users suggested enhancing the accessibility of
potassium yielded three output values: nitrogen pump certain features, optimizing the mobile application for
rate, phosphorous pump rate, and potassium pump different device types, and refining the visualization
rate. The crisp output is generated at a nitrogen pump of analytical data. Their input was invaluable in
rate of 1.44 seconds, a phosphorous pump rate of guiding future refinements of the system. In
6.67 seconds, and a potassium pump rate of 1.44 summary, the integration of user testing and feedback
seconds. collection methods played a pivotal role in the
3.15Usability assessment and user feedback evaluation of the system's usability. These insights do
The evaluation of the proposed system included a not only provide validation of the system's
comprehensive assessment of usability, user- effectiveness but also directed us toward refinements
friendliness, and overall system performance. This and enhancements that will further optimize the user
process involved the application of user testing and experience. The user-centric approach to assessment
feedback collection methods, making sure that the ensures that the Smart fertigation system with mobile
system is effective in real-world usage scenarios. application and fuzzy logic optimization is not only
These methods play their part in gathering invaluable technologically advanced but also aligned with the
insights into the system's usability and its alignment practical needs and expectations of its end-users in
with end-users' needs and expectations. The findings agricultural settings.

Figure 19 Use case diagram

1595
Nurul Anis Zulaikha Izahar et al.

Figure 20 Irrigation fuzzy reasoning rule from MATLAB

Figure 21 Fertilization fuzzy reasoning rule from MATLAB

4.Results Figure 23 shows the page for the chili plant

In this section, we present the outcomes of our study, monitoring. As for this figure, these are the chili
offering a detailed examination of the performance plants in Section 1. The information is divided into
and efficacy of the proposed smart fertigation system two parts which are irrigation system and fertigation
with mobile application and fuzzy logic optimization. system. In the irrigation part, the data includes the
The results are delineated through two main last watered status, including day, date, and time with
subsections: remote monitoring system and testing the volume of water used for the irrigation. It also
evaluation. displays the soil moisture, temperature, and humidity
value. In the fertilization part, the data includes the
4.1Remote monitoring system day, date, and time of the most recent fertilization as
Figure 22 shows the interface of the C-farm ratio of well as the amount of NPK fertilizer used. The values
fertilizer page. Information like the NPK fertilizer for nitrogen, phosphorus, and potassium are also
ratio based on the stage of plant growth can be shown. Both irrigation and fertigation part will
viewed on this page. The ratio of fertilizer for the display the latest data that are retrieved from the
root phase, leafy phase, and fruit phase are 1-2-1, 2- cloud firestore database. Additionally, the info icon
1-1, and 1-1-2 respectively. There is also a “Home” in the fertilization part displays a popup message
icon on the application bar which led the user back to regarding the NPK fertilizer ratio. On the right side
the home page. of the application bar, there are three different icons.
The analytics icon is the first icon to the right that

1596
 International Journal of Advanced Technology and Engineering Exploration, Vol 10(109)

leads user to the analytics page, while the history

icon, the second icon to the right leads user to the
history page.

Figure 24 C-farm analytics page

4.2Testing evaluation
For the purpose of instructing the water pump for
Figure 22 C-Farm ratio of fertilizer page
optimizing the water distribution across the farm, the
system makes use of input variables including soil
Figure 24 shows the interface of C-Farm analytics
moisture, humidity, and temperature in an effort to
page. The user will be directed to the analytics page
increase the irrigation system's efficiency and
after clicking the analytics icon. On display, there
reliability. The system's three inputs will be used to
will be a bar graph showing monthly water and NPK
determine the irrigation output. In terms of
fertilizers usage. The graph can be scrolled to the
fertilization, the nitrogen pump, phosphorous pump,
right to view the previous month.
and potassium pump are each controlled by input
variables like nitrogen, phosphorous, and potassium.
The system will calculate the fertilization output
using the three inputs. Both irrigation and
fertilization system output are summarized in the
respective Table 6 and Table 7, based on the data
gathered. Specific to fertilization, the data were
collected when the chili plant was at its root phase.

The smart fertigation system with mobile application

and fuzzy logic optimization represents a significant
enhancement of existing automated fertigation
systems. A meticulously designed hardware setup
ensures the system's reliability, underscoring the
importance of selecting appropriate hardware and
software components to construct an effective and
successful system. The integration of a fuzzy logic
algorithm further augments the system's efficiency.
Additionally, the inclusion of a mobile application
empowers agricultural business owners with the
ability to remotely monitor and manage their farms,
contributing to improved overall operational control.
Figure 23 C-farm plant monitoring page
1597
Nurul Anis Zulaikha Izahar et al.

4.2.1Validation of the fuzzy logic algorithm which excels in managing the inherent uncertainty in
The fuzzy logic algorithm validation process was agricultural systems.
rigorous and thorough. It commenced with the
meticulous collection of real-time data, serving as the Extensive simulation testing simulated various
foundation for creating precise fuzzy logic rules that environmental conditions and growth stages,
governed irrigation and fertilization replicating real-world variability. The algorithm's
recommendations. The dataset included vital responsiveness and adaptability were evaluated,
parameters such as soil moisture, temperature, demonstrating its readiness for practical agricultural
humidity, nitrogen, phosphorous and potassium. applications. Furthermore, a direct comparison with
Following rule-based design, a comprehensive set of traditional methods confirmed the algorithm's
fuzzy logic rules was established, expertly mapping superior performance in optimizing water and
input variables to irrigation and fertilization nutrient delivery, establishing it as a key component
recommendations. This granular approach ensured within the smart fertigation system with mobile
tailored responses to the dynamic needs of cultivated application and fuzzy logic optimization.
plants, highlighting the adaptability of fuzzy logic,

Table 6 Irrigation testing data

Soil moisture (%) Temperature(oC) Humidity (%) Water Pump(ml)
63.4 29.7 56.9 48.8
58.7 29.5 39 48.4
63.4 29.7 56.9 48.8
75.3 27.9 56.9 45.8
51 26.9 70 150
49.8 26.6 59 43.8
51.3 27.7 69 48
36.5 26.4 76 200
65.2 26.9 76 44.2
63.4 29.7 56.9 48.8

Table 7 Fertilization testing data

Nutrient in soil (mg/kg) Pump (ml)
Nitrogen Phosphorous Potassium Nitrogen Phosphorous Potassium
78 65 59 28.8 133.4 28.8
56 69 73 27.4 133.4 27.4
74 63 55 26 133.4 26
68 79 43 31.8 133.4 31.8
53 49 62 24.8 133.4 24.8
61 47 52 25.4 133.4 25.4
60 55 63 23.4 133.4 23.4
61 65 69 26.2 133.4 26.2
82 78 79 31.8 133.4 31.8
78 65 59 28.8 133.4 28.8

4.2.2Hardware component testing each component is reliable and accurate during its
Hardware component testing was a critical phase of operation across a diverse range of conditions.
the evaluation process, ensuring the reliability and Integration testing: Our evaluation also focused on
functionality of the fertigation system. It the interaction and compatibility of hardware
encompassed some key elements such as functional components within the system. This meticulous
testing, integration testing, and long-term reliability examination identified and resolved potential
testing. integration issues, ensuring the seamless operation of
Functional testing: Every hardware component, the system as a whole.
which includes sensors, pumps, and controllers, Long-term reliability testing: To assess the
underwent comprehensive functional testing. This durability and long-term reliability of the hardware
rigorous examination was done to make sure that components, they were subjected to extended periods

1598
 International Journal of Advanced Technology and Engineering Exploration, Vol 10(109)

of operation and data collection. This phase of testing receive the precise amount of water and nutrients
was important to ensure that the components are able they require. As a result, it minimizes resource
to perform consistently over an extended duration. wastage and overcomes the challenges of under or
4.2.3Accuracy and performance metrics over-irrigation and fertilization, thus, improving crop
To measure the accuracy and performance of the health and productivity.
fuzzy logic algorithm and the hardware components, Resource optimization: The system's dynamic
we established a range of specific metrics: adjustments based on environmental variables and
Accuracy in irrigation and fertilization: Metrics crop requirements mark a substantial step forward in
such as precise water distribution and optimal resource optimization. As an example, the integration
nutrient supply were employed to assess the accuracy of rain prediction data allows for intelligent water
of the system's recommendations. This ensures that management, reducing over-irrigation and conserving
delivery of the right amount of water and nutrients to water resources. Similarly, the tailored fertilization
the plants are done consistently by the system. process ensures that nutrients are applied precisely,
Energy efficiency: Energy consumption of the mitigating over-fertilization and its associated
hardware components was systematically recorded environmental consequences. This, in essence,
and analyzed. This approach ensured that the system promotes to sustainable farming practices and
operates efficiently, while minimizing the use of economic savings.
energy as it maintains optimal functionality. Improved crop yield and quality: The system helps
Reliability and stability: The system's overall to improve crop yield and quality through precise
reliability and stability were assessed through control over irrigation and fertilization. By providing
continuous operation and performance monitoring. crops with the right amount of nutrients and water at
This aspect of testing was crucial in determining the the right time, it helps maximize their growth
system's ability to maintain consistent performance potential and resilience to adverse environmental
over time. conditions. Not only does it enhance the overall
yield, but it also ensures the high quality of the crops,
In short, through rigorous validation and testing, we meeting market standards and consumer preferences.
have confirmed the Smart fertigation system with Real-time monitoring and decision support: The
mobile application and fuzzy logic optimization's mobile application's role as a real-time monitoring
reliability and robustness to be used in the real-world tool plays a crucial role. As it provides farmers with
agricultural field. The system consistently delivers continuous access to critical data, the application
precise recommendations for irrigation and empowers timely decision-making and intervention.
fertilization, and the hardware components With this level of control, it can lessen the risks
seamlessly integrate, making it durable in long-term. associated with suboptimal environmental conditions
The fuzzy logic algorithm shows adaptability and and enable proactive responses to plant health
accuracy, while a set of key metrics ensures high concerns.
performance standards, positioning our system as a Potential environmental and economic impact:
reliable, efficient, and effective solution for modern The Smart fertigation system with mobile application
agriculture. and fuzzy logic optimization offers substantial
potential, not only in terms of environmental, but also
5.Discussion economic impact.
The integration of the Smart fertigation system with Environmental impact: By reducing over-irrigation
mobile application and fuzzy logic optimization and over-fertilization, the system helps in
portrays a significant advancement in modern sustainability and water conservation. It helps
agriculture, exemplifying the potential of technology reducing the environmental consequences of
in enhancing crop management and utilizing excessive resource usage, such as nutrient runoff and
resource. Several key aspects require further water wastage. Moreover, the real-time monitoring
consideration and reflection. and predictive capabilities of the system allow for the
Efficiency of irrigation and fertilization processes: early detection of potential issues, enabling more
The incorporation of IoT sensors and fuzzy logic targeted interventions and minimizing environmental
optimization has significantly improved the damage.
efficiency of both irrigation and fertilization Economic impact: The system contributes to cost
processes. Real-time data from the sensors and the savings in agriculture by optimizing resource
application's intelligent algorithms collaboratively utilization and improving crop yield and quality. This
fine-tune these processes, guaranteeing that the crops situation increases revenue for farmers since they are

1599
Nurul Anis Zulaikha Izahar et al.

able to achieve higher crop production with reduced stations, holds the potential to enhance predictive
input costs. Additionally, it reduces the risk of crop accuracy and refine decision-making. Addressing
failure since the crops are able to adapt to changing these limitations presents an opportunity to maximize
environmental conditions, thus protecting the the system's utility and ensure its applicability across
investment done by the farmers. a broad spectrum of agricultural settings. A complete
Comparison with existing fertigation systems: list of abbreviations is shown in Appendix I.
In the context of existing fertigation systems, it's
important to acknowledge the limits of only utilizing 6.Conclusion and future work
traditional methods and partially automated systems. Agriculture is a vital industry that provides the
Traditional fertigation methods often rely heavily on world's necessity of food. To improve agriculture
human labour, requiring the farmers’ constant practices, it is essential to monitor crops, and the
presence to monitor and manage the crops. This Smart fertigation system with mobile application and
manual approach does not only lead to inefficiencies, fuzzy logic optimization is developed to assist
but also leads to human errors, and difficulties in farmers. The system uses the IoT and mobile
maintaining precision, especially in large-scale application to reduce labor costs and time, and fuzzy
farming operations. While some partially automated logic to improve precision in plant irrigation and
systems have been proposed, a continuous key issue fertilization, resulting in better crop yield and quality.
was the reliance on farmers’ decisions to manage the As a future work, the Smart Fertigation System can
fertilization. This introduces a degree of subjectivity be further improved by adding a function to notify
and may result in less accurate application of water farmers of pump malfunctions through the mobile
and fertilizer to crops, potentially leading to issues of application, integrating rain prediction into the
over-fertilization or under-fertilization. The irrigation system, and using raspberry pi instead of
shortcomings of these existing systems drive the NodeMCU as an intermediary to connect directly to
development of the proposed smart fertigation the cloud firestore database, thus can reduce the
system, designed to address these challenges and hardware complexity.
usher in a new era of automation and precision in
fertigation. The system integrates IoT technology, a In conclusion, the convergence of technology and
user-friendly mobile application for real-time agriculture finds a remarkable expression in the
monitoring, and innovative fuzzy logic decision- Smart fertigation system with mobile application and
making processes to optimize water and fertilizer fuzzy logic optimization. This innovative integration
application based on the growth phase of the plants. has the potential to reshape traditional farming
This approach represents a significant leap towards a practices and enhance agricultural outcomes.
fully automated and highly precise fertigation system, Through the utilization of IoT sensors and a user-
enhancing both the efficiency of crop management centric mobile application, the system simplifies
and the quality of the crops produced, particularly complex farming tasks and introduces a new level of
benefiting large-scale bird's eye chili growers. control and insight for farmers. The incorporation of
fuzzy logic algorithms adds a layer of sophistication,
The study has limitations that warrant consideration. enabling precise and adaptive management of
One limitation lies in the need for comprehensive irrigation and fertilization processes, thereby
calibration and validation of the fuzzy logic promoting resource efficiency and optimal crop yield.
optimization to ensure the accuracy of the system's Looking ahead, a roadmap of promising future
outcomes. Rigorous testing and calibration are developments emerges to amplify the system's
essential to guarantee that the system consistently capabilities and impact. The introduction of a pump
delivers the expected results. Another limitation is malfunction alert mechanism within the mobile
the adaptability of the system to various crops and application ensures timely responses to technical
their different growth stages. This necessitates glitches, safeguarding crops from potential damage.
parameter adjustments and experimentation to Apart from that, by incorporating predictive rain
guarantee the system's effectiveness when applied to forecasts into the system's decision-making process,
a wide range of crops, ensuring optimal performance irrigation can be intelligently adjusted to account for
under different conditions. Given the unique impending weather conditions, thus conserving water
requirements and growth patterns of different crops, resources while maintaining crop health.
meticulous fine-tuning is essential to meet these Furthermore, the transition from NodeMCU to
diverse needs effectively. Furthermore, integrating raspberry pi for cloud connectivity holds the promise
data from various sources, such as local weather of enhanced data processing power and seamless
1600
 International Journal of Advanced Technology and Engineering Exploration, Vol 10(109)

communication, reducing hardware complexity and methodologies. In international conference on trends

increasing reliability. This transition enables the in electronics and informatics 2019 (pp. 1147-52).
system to handle more complex algorithms and IEEE.
analyses, facilitating smarter and more informed [5] Hakim IM, Singgih ML, Gunarta IK. Critical success
factors for internet of things (IoT) Implementation in
agricultural management. Automotive Companies, Indonesia. Sustainability.
2023; 15(4):1-18.
To cater to the diverse needs of farmers and crops, an [6] Duang-ek-anong S, Pibulcharoensit S, Phongsatha T.
expansion of the system's adaptability to different Technology readiness for internet of things (IoT)
plant species and growth stages would broaden its adoption in smart farming in Thailand. International
application and utility. Customizable algorithms Journal of Simulation: Systems, Science &
could provide tailored recommendations, optimizing Technology. 2019; 20:1-6.
the system's effectiveness across various agricultural [7] En GWW, Devanthran H. The development of smart
scenarios. farming technologies and its application in Malaysia.
International Journal of Scientific and Technology
Research.2020; 9(8):561-66.
In essence, the Smart Fertigation System's journey is [8] Zaman NB, Raof WN, Saili AR, Aziz NN, Fatah FA,
a testament to the ongoing pursuit of precision and Vaiappuri SK. Adoption of smart farming technology
sustainability in agriculture. The envisioned among rice farmers. Journal of Advanced Research in
enhancements collectively chart a course toward an Applied Sciences and Engineering Technology. 2023;
agricultural landscape characterized by efficient 29(2):268-75.
resource utilization, improved yield, and [9] Abidin SA, Ibrahim SN. Web-based monitoring of an
technological sophistication, ensuring a brighter and automated fertigation system: an IoT application. In
more secure future for global food production. 12th Malaysia international conference on
communications 2015 (pp. 1-5). IEEE.
[10] Ahmed OM, Osman AA, Awadalkarim SD. A design
Acknowledgment of an automated fertigation system using IoT. In
None. international conference on computer, control,
electrical, and electronics engineering 2018 (pp. 1-5).
Conflicts of interest IEEE.
The authors have no conflicts of interest to declare. [11] Ruan J, Liao P, Dong C. The design and research on
intelligent fertigation system. In international
Author’s contribution statement conference on intelligent human-machine systems and
Nurul Anis Zulaikha Izahar: Investigation, development,
cybernetics 2015 (pp. 456-9). IEEE.
testing, analysis, writing, and editing. Mohd Noor
[12] Joseph C, Thirunavuakkarasu I, Bhaskar A, Penujuru
Derahman: Idea generation, data validation, correction,
A. Automated fertigation system for efficient
study conception, design, data collection, supervision, and
utilization of fertilizer and water. In international
overall project coordination. Mohamad Afendee
conference on information technology and electrical
Mohamed: Editing, manuscript testing, manuscript
engineering 2017 (pp. 1-6). IEEE.
preparation, data validation, feedback, project support and
[13] Ahmad U, Alvino A, Marino S. Solar fertigation: a
technical content review. Imas Sukaesih Sitanggang:
sustainable and smart IoT-based irrigation and
Editing, manuscript testing, manuscript preparation,
fertilization system for efficient water and nutrient
research coordination data validation, feedback, data
management. Agronomy. 2022; 12(5):1-24.
analysis support.
[14] Zhang H, He L, Di GF, Choi D, Elia A, Heinemann P.
LoRaWAN based internet of things (IoT) system for
References precision irrigation in plasticulture fresh-market
[1] Elferink M, Schierhorn F. Global demand for food is tomato. Smart Agricultural Technology. 2022;
rising. Can we meet it. Harvard Business Review. 2:100053.
2016; 7(4):1-5. [15] Khairodin FN, Rahman TA, Elijah O, Saharuddin HI.
[2] Boopathy S, Anand KG, Priya ED, Sharmila A, Smart IoT system for chili production using Lora
Pasupathy SA. Iot based hydroponics based natural technology. In integrated emerging methods of
fertigation system for organic veggies cultivation. In artificial intelligence & cloud computing 2021 (pp. 22-
third international conference on intelligent 33). Cham: Springer International Publishing.
communication technologies and virtual mobile [16] Rode S, Saraf R, Veigas J, Shetty N, Shardul S.
networks 2021 (pp. 404-9). IEEE. Design and fabrication of IoT based agricultural
[3] Kummar L, Al-aani FS, Kahtan H, Darr M, Al-bashirl automation system. In Biennial international
H. Data visualisation for smart farming using mobile conference on nascent technologies in engineering
application. International Journal of Computer Science 2023 (pp. 1-6). IEEE.
and Network Security. 2019; 19:1-7. [17] Zailani MZ, Jumaat SA. A monitoring system of soil
[4] Mathur B, Satapathy SM. An analytical comparison of moisture for fertigation system using IOT application.
mobile application development using agile
1601
Nurul Anis Zulaikha Izahar et al.

Evolution in Electrical and Electronic Engineering. [31] Bhagat SA, Dudhalkar SG, Kelapure PD, Kokare AS,
2021; 2(2):856-66. Bachwani PS. Review on mobile application
[18] Nandi PK, Mahmood MA. An automated irrigation development based on flutter platform. International
and fertilization management system using fuzzy Journal for Research in Applied Science and
logic. In international conference on electrical Engineering Technology. 2022; 10(1):803-9.
information and communication technology 2021 (pp. [32] Azry AS, Derahman MN, Mohamad ZA, Rahiman
1-5). IEEE. AR, Muzakkari BA, Mohamed MA. Fuzzy logic-
[19] Jha K, Doshi A, Patel P, Shah M. A comprehensive based intelligent irrigation system with mobile
review on automation in agriculture using artificial application. Journal of Theoretical and Applied
intelligence. Artificial Intelligence in Agriculture. Information Technology. 2022; 100(18):5323-34.
2019; 2:1-12. [33] Manan A, Wiley V, Lucas T. Programmer selection
[20] Durai SK, Shamili MD. Smart farming using machine using modified fuzzy mamdani method. Lontar
learning and deep learning techniques. Decision Komputer : Jurnal Ilmiah Teknologi Informasi. 2019;
Analytics Journal. 2022; 3:100041. 10(2):108-18.
[21] Pezol NS, Adnan R, Tajjudin M. Design of an internet [34] Wawan W, Zuniati M, Setiawan A. Optimization of
of things (IoT) based smart irrigation and fertilization national rice production with fuzzy logic using
system using fuzzy logic for chili plant. In Mamdani method. Journal of Multidisciplinary
international conference on automatic control and Applied Natural Science. 2021:1-8.
intelligent systems 2020 (pp. 69-73). IEEE. [35] Awalludin N, Gerhana YA, Mulyana E, Lukman N,
[22] Mohanraj I, Gokul V, Ezhilarasie R, Umamakeswari Dauni P, Dzikrayah F. Implementation of fuzzy logic
A. Intelligent drip irrigation and fertigation using method to measure soil moisture and environmental
wireless sensor networks. In technological innovations temperature in automatic watering system based on
in ICT for agriculture and rural development 2017 (pp. internet of things. In 8th international conference on
36-41). IEEE. wireless and telematics 2022 (pp. 1-5). IEEE.
[23] Saban M, Bekkour M, Amdaouch I, El GJ, Ait AB, [36] Sivanandam SN, Sumathi S, Deepa SN. Introduction
Chaari MZ, et al. A smart agricultural system based on to fuzzy logic using MATLAB. 2007; Springer.
PLC and a cloud computing web application using [37] Sharma S, Obaid AJ. Mathematical modelling,
LoRa and LoRaWan. Sensors. 2023; 23(5):1-16. analysis and design of fuzzy logic controller for the
[24] Pinatih PT, Dewi NN, Sutadarma IW. Green chili control of ventilation systems using MATLAB fuzzy
(Capsicum frutescens) ethanol extracts lowered logic toolbox. Journal of Interdisciplinary
triglyceride levels in rats fed with high-fat diet. Mathematics. 2020; 23(4):843-9.
Biomedicine. 2022; 42(6):1275-80. [38] Pierre NJ, Sefu B, Venuste S, D’amour MJ, Daniel K,
[25] Gupta C, Tewari VK, Machavaram R, Shrivastava P. Pierre NJ, et al. Smart crops irrigation system with
An image processing approach for measurement of low energy consumption. Journal of Appropriate
chili plant height and width under field conditions. Technology. 2023; 9(1):9-19.
Journal of the Saudi Society of Agricultural Sciences. [39] Sukarsa I, Antara IK, Buana PW, Bayupati I,
2022; 21(3):171-9. Wisswani NW, Puteri DW. Data storage model in
[26] Adek RT, Fikry M, Naluri H. Automatic control low-cost mobile applications. Indonesian Journal of
system using arduino UNO and web-based monitoring Electrical Engineering and Computer Science. 2022;
for watering Chili plants. Journal of Informatics and 28(2):1-11.
Telecommunication Engineering. 2022; 5(2):510-9.
[27] Almasri AK. A proposed hybrid agile framework Nurul Anis Zulaikha Izahar is a
model for mobile applications development. highly motivated and dynamic
International Journal of Software Engineering and Its Computer Science student of Universiti
Applications. 2016; 7(2):1-9. Putra Malaysia (UPM). Her academic
[28] Nazir M. Software quality assurance and android journey at UPM, underscores her
application development: a comparison among dedication to mastering complex
traditional and agile methodology. Lahore Garrison concepts in the realm of computer
University Research Journal of Computer Science and science.
Information Technology. 2020; 4(4):1-29. Email: nurulaniszulaikha@gmail.com
[29] Haidar R, Ibrahim K, Asaad R, Ahmad Z, Shamkuwar
S. Building smart mobile apps with flutter and open Mohd Noor Derahman is a computer
AI AI-powered text and images and chatbots. Journal science professional with extensive
For Research in Applied Science and Engineering experience and a strong academic
Technology. 2023; 11(6):904-8. background. His research interests
[30] Rimal K, Shah KB, Jha AK. Advanced multi-class revolve around Computer Networks,
deep learning convolution neural network approach Distributed Heterogeneous Systems,
for insect pest classification using TensorFlow. and networking. Derahman has worked
International Journal of Environmental Science and as a lecturer at the University Putra
Technology. 2023; 20(4):4003-16.

1602
 International Journal of Advanced Technology and Engineering Exploration, Vol 10(109)

Malaysia since 2003 and gained practical experience with a Appendix I

focus on computer science and network-related subjects. S. No. Abbreviation Description
Email: mnoord@upm.edu.my 1 AI Artificial Intelligence
2 ANN Artificial Neural Networks
Mohamad Afendee Mohamed 3 API Application Programming Interface
received his PhD in Mathematical 4 EC Electrical Conductivity
Cryptography from Universiti Putra 5 IoT Internet of Things
6 K Phosphorus
Malaysia in 2011. Upon completion, he
7 LoRaWAN Long Range Wide Area Network
served the university for three years as
8 ML Machine Learning
a senior lecturer. From 2014, he moved 9 N Nitrogen
to Universiti Sultan Zainal Abidin and 10 NPK Nitrogen, Potassium, Phosphorus
later assumed an associate professor 11 P Pottasium
position. His current research interests include both
theoretical and application issues in the domain of data
Security, and Mobile and Wireless Networking. Dr
Mohamed has authored and co-authored more than 100
articles that have appeared in various journals, book
chapters and conference proceedings.
Email: mafendee@unisza.edu.my

Imas Sukaesih Sitanggang received

the PhD Degree in Computer Science
from Faculty of Computer Science and
Information Technology, Universiti
Putra Malaysia in 2013. She is a
lecturer in Computer Science
Department, IPB University, Indonesia.
Her main research interests include
Spatial Data Mining and Data Warehousing.
Email: imas.sitanggang@apps.ipb.ac.id

1603
 © 2023. This work is published under
http://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding
the ProQuest Terms and Conditions, you may use this content in accordance
with the terms of the License.