Content uploaded by Krish Jayachandran
Author content
All content in this area was uploaded by Krish Jayachandran on Feb 21, 2024
Content may be subject to copyright.
Content uploaded by Aali Mana
Author content
All content in this area was uploaded by Aali Mana on Feb 20, 2024
Content may be subject to copyright.
Smart Agricultural Technology 7 (2024) 100416
Available online 17 February 2024
2772-3755/© 2024 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-
nc-nd/4.0/).
Sustainable AI-based production agriculture: Exploring AI applications and
implications in agricultural practices
A.A. Mana
a
, A. Allouhi
a
, A. Hamrani
b
,
*
, S. Rahman
c
, I. el Jamaoui
d
, K. Jayachandran
e
a
Ecole Sup´
erieure de Technologie de F`
es, U.S.M.B.A, Route d’Imouzzer, BP 242 Fez, Morocco
b
Department of Mechanical and Materials Engineering, Florida International University, Miami, FL 33174, USA
c
Interdisciplinary Research Center for Renewable Energy and Power Systems (IRC-REPS), King Fahd University of Petroleum and Minerals (KFUPM), Dhahran 31261,
Saudi Arabia
d
Department of Agricultural Chemistry, Geology and Pedology, Faculty of Chemistry, University of Murcia, 30100 Murcia, Spain
e
Agroecology Program, Department of Earth and Environment, Florida International University, Miami, FL 33199, USA
ARTICLE INFO
Editor: Stephen Symons
Keywords:
Articial intelligence
Precision agriculture
Energy
Sustainability
ABSTRACT
In general, agriculture plays a crucial role in human survival as a primary source of food, alongside other sources
such as shing. Unfortunately, global warming and other environmental issues, particularly in less privileged
nations, hamper the Agricultural sector. It is estimated that a range of 720 to 811 million individuals experienced
food insecurity. Today’s agriculture faced signicant difculties and obstacles, as do the surveillance and
monitoring systems (climate, energy, water, elds, works, cost, fertilizers, diseases, etc.). The COVID-19
pandemic has exacerbated the susceptibilities and insufciencies inherent in worldwide food systems. Current
agricultural practices tend to prioritize productivity and protability over environmental conservation and long-
term sustainability. To establish sustainable agriculture capable of meeting the needs of a projected ten billion
people in the next 30 years, substantial structural and automation changes are required. However, these ob-
stacles can be overcome by employing smart technologies and advancing Articial Intelligence (AI) in agricul-
tural operations. AI is believed to contribute to global sustainability goals in multiple sectors, particularly in the
incorporation of renewable energy. It is anticipated that AI will revitalize both existing and new agricultural
elds by retrotting, installing and integrating automatic devices and instruments. This paper presents a
comprehensive review of the most promising and novel applications of AI in the agriculture industry. Further-
more, the role of AI in the transition to sustainability and precision agriculture is investigated.
Introduction
Background
Currently, agriculture faces critical challenges, including climate
change, water scarcity, environmental degradation, and dependence on
conventional energy sources. Intensive transformation and landscaping
can reduce biodiversity, pollute air and water sources, and put human
and animal health at risk [1,2]. As such, a worldwide agricultural mu-
tation is needed to switch from conventional to modern automated ap-
proaches [3]. Such approaches consider the agriculture farms as
factories and plants and animals as production units [4,5]. The devel-
opment of sustainable agriculture has attracted the attention of several
countries focusses on economic, social, and environmental sustainability
[6,7]. Smart innovations will be necessary to transition to a smarter and
more sustainable agricultural sector [8]. Articial intelligence (AI) seeks
to develop technologies and systems capable of performing like human
intelligence [9,10]. Recently, AI has been shown to be essential for in-
formation and services in the elds of health [11], education [12], trade
[13], and others.
Moreover, AI holds signicant potential to enhance the sustainability
of the agricultural industry through various applications. For instance, it
can contribute to identifying the optimal time for harvesting fruits and
vegetables, reducing waste, and monitoring the health of soil and crops.
Using AI, real-time monitoring of crop production can be improved,
enabling more effective and supervised processing. Additionally, the
integration can lead to efcient water use, conserving this valuable
resource while maximising crop yields [14]. Additionally, the deploy-
ment of robots and drones in agricultural farms can detect weeds, pests
and diseases, nutrient-decient spots, monitor crop yield and quality,
and other applications.
* Corresponding author.
E-mail address: ahamrani@u.edu (A. Hamrani).
Contents lists available at ScienceDirect
Smart Agricultural Technology
journal homepage: www.journals.elsevier.com/smart-agricultural-technology
https://doi.org/10.1016/j.atech.2024.100416
Received 9 May 2023; Received in revised form 13 January 2024; Accepted 16 February 2024
Smart Agricultural Technology 7 (2024) 100416
2
Benos et al. [15] delved into the use of AI and machine learning in
agriculture, covering aspects like crop, water, soil, and livestock opti-
mization. The study highlighted crop management as the most promi-
nent area, with a focus on maize, wheat, and sheep. To foster smart and
sustainable farming, key elements include ecosystem conservation,
adopting modern technologies, effective resource management, and
providing robust services in AI-based agriculture [16]. To ensure agri-
cultural sustainability and maximise crop productivity while preserving
the environment [17], it is imperative to improve, optimize, and
modernize farming practises. Machine learning techniques have found
applications in various aspects of sustainable agriculture, including crop
recognition [18,19], crop disease identication [20,21], weed detection
[22,23], water management [24,25], animal health [26,27], and live-
stock production [28,29]. AI has contributed a lot to agriculture and has
protected crops from adverse weather changes and provided food se-
curity. Talaviya et al. [30] presented the analysis of some applications of
AI in agriculture that included irrigation, weeding, and spraying using
robots, sensors, and drones. Such AI-based techniques and equipment
conserve water, pesticides, and, herbicide consumption and at the same
time, maintain soil fertility, manage efcient use of manpower, and
enhance the production quality [30]. The promising AI technologies
being used or under various stages of technological development for
sustainable agriculture, practices can be summarized as follows:
•Crop and soil health monitoring
•Automated weeding
•Intelligent spraying
•Insect and Plant Disease Detection
•Livestock Health Monitoring
•Harvesting, plowing, and pruning
•Produce grading and sorting
•Energy security
However, smart agriculture is a practical and promising alternative
to satisfy global food demand while maintaining a balance between the
agricultural industry and the environmental ecosystem. The concept of
smart agriculture refers to all management practices that use AI, ma-
chine learning, data-driven, and recent technologies to ensure the
quality and quantity of agricultural products. This concept is one of the
major elements in the 4.0 revolution. AI can help farmers at every step,
from soil preparation for seeding to harvesting with robots and
computer-assisted engines. In 2017, the total estimated value of in-
vestments in the AI technologies in agricultural sector was 518.7 million
dollars and is expected to reach 2.6 billion by 2025, with an annual
increase of 16.2 % [31].
Scope and motivation of the study
The aim of this review article is to analyse the existing practices and
technologies that are used and are sustainable for the ecological and
digital transition in the agriculture sector. Preliminary understanding is
that AI and machine-learning techniques have the potential to provide
solution to improve agriculture and sustain the agro-production. The
present scope of work denes sustainable agriculture’s dimensions and
provides a comprehensive review of AI utilization in the agriculture
industry. The review also aims at providing a useful discussion on the
most promising applications of AI in agriculture sector. Specically, this
review paper is intended to propose answers to the following research
questions.
- How can AI promote agriculture practices and accelerate
sustainability.
- To what extent the use of can AI promote the sustainability of agri-
culture in developing countries with agriculture-based economies?
- What are the limitations and challenges transitioning to smart agri-
cultural practices?
- The future of AI in Agriculture: Farmers as AI engineers?
AI applications in agriculture
Sustainable agriculture is a fundamental approach to meeting soci-
ety’s food and rural needs while safeguarding the ability of future gen-
erations to meet their own needs. To achieve this, it is based on
understanding of ecosystem services. Articial intelligence emerges as a
powerful tool for promoting sustainable agriculture, optimizing various
farming aspects such as water and energy management, precision agri-
culture and smart farming techniques [32].
AI-driven agriculture plays a vital role in improving precision and
contributing to overall sustainable farming practices. Through AI algo-
rithms, farmers can gain insight to efciently manage irrigation and
conserve water resources. In addition, AI facilitates automating labor-
intensive tasks, such as crop harvesting, pruning, and plowing in agri-
culture, with the aid of autonomous tractors and harvesters guided by AI
technologies, thus reducing the need for extensive human intervention.
However, the world faces a twofold challenge: an ever-growing
population and widespread hunger. To address these critical issues, an
integrated approach is necessary to address the challenges related to soil
fertility, water scarcity, energy insecurity, pests, and diseases that affect
crops and animals. Sustainable agriculture is dened by a set of methods
that ensure food production in harmony with ecological, economic, and
social limits [33].
In the realm of agriculture, the concept of "secured smart sustainable
agriculture (SSSA)" embodies an integrated framework that amalgam-
ates several essential branches to catalyze a revolution, ensuring the
sustainability, efciency, and security of the agricultural sector. Picture
this framework as a dynamic ecosystem, vividly illustrated in Fig. 1,
where each branch assumes a pivotal role in sculpting the future of
agriculture. Sustainable Agriculture encompasses practices such as crop
rotation and organic farming, augmenting yields while concurrently
minimizing the environmental footprint. On the other hand Data
Nomenclature
ADM Agricultural Decision-Making
AI Articial intelligence
ANNs Articial Neural Networks
BMBF Federal Ministry of Education and Research
CNNs Convolutional Neural Networks
DT Digital twin
DL Deep Learning
EC Edge Computing
FAO Food and Agriculture Organization
IPCC The Intergovernmental Panel on Climate Change
IOT Internet of things
ML Machine learning
MLP Multi-layer perceptron
M2M Machine to Machine
NDVI Normalized Difference Vegetation Index
PA Precision Agriculture
PLF Precision Livestock Farming
PWM Precision Water Management
SSCM Site-specic crop management
SSSA Secured Smart Sustainable Agriculture
RPAS Remotely Piloted Aircraft System
UAV Unmanned Aerial Vehicle
A.A. Mana et al.
Smart Agricultural Technology 7 (2024) 100416
3
Security stands as a pivotal facet, diligently safeguarding sensitive
agricultural data garnered through the deployment of sensors, drones,
and IoT devices.
Furthermore, blockchain Technology, with its decentralized and
immutable ledgers, bestows transparency and traceability advantages,
thus guaranteeing the unassailable integrity of supply chains and
nancial transactions. Precision Agriculture harnesses cutting-edge
technology, including GPS, sensors, and analytics, to facilitate real-
time decision-making, thereby mitigating resource wastage and
elevating crop yields. In this perspective, renewable Energies are inte-
grally woven into this framework, enabling the assimilation of clean
energy sources such as solar panels and wind turbines into farming
operations, thereby curtailing carbon emissions and energy expenditure.
IoT-Based Solutions (Internet of Things) intricately connect an array
of devices and sensors, delivering real-time monitoring and fostering
data-driven decision-making for a more resource-efcient and secure
agricultural practice.
On another front, AI was dened in 1956, as ‘the science and engi-
neering of making intelligent machines.’ The main notion was to create
a technology able to perform like a human intelligence [9]. This was
achieved by studying human brain processes to develop intelligent
software and systems capable of offering the optimal result for all valid
input [10]. The eld of AI is rapidly expanding, including Machine
Learning and Deep learning. The main objective of machine learning is
to obtain computational models of complex non-linear relationships or
complex models in the data, whereas AI can be described as a tool of
decision making and advanced analytics [34]. MLs are often used to
trace models in data,as well as to achieve high performance [35].
ML algorithms are trained using three prominent methods commonly
recognised as ‘supervised’ where the system learns from labelled data;
‘unsupervised’ where the unlabelled system nds patterns in the data;
and ‘reinforcement’ learning from new situations using a trial-and-error
method [36]. Fig. 2 shows the different ML used in literatures. However,
it’s important to note that the representation of ML in Fig. 2 lacks the
inclusion of reinforcement learning, a signicant omission that deserves
acknowledgment. Reinforcement learning holds particular signicance
as it involves learning optimal decision-making strategies through in-
teractions with an environment, receiving feedback in the form of re-
wards or penalties [37]. This method overlooks a crucial aspect of ML
algorithms, as reinforcement learning plays a vital role in various AI
applications, such as robotics, gaming, and autonomous systems [38].
Fig. 1. Main pillars of Secured Smart Sustainable Agriculture.
Fig. 2. Articial intelligence dimensions.
A.A. Mana et al.
Smart Agricultural Technology 7 (2024) 100416
4
Its ability to learn from experiences and optimize decisions based on
feedback distinguishes it as a key paradigm within ML.
AI has signicantly contributed towards improving the efciencies of
many engineering and social led to problems of different industries.
These technologies have even been introduced in the agricultural value
chain from production to transport, to distribution, and to marketing.
Agricultural robots have added a high value of AI and help in several
stages of agricultural production such as increasing crop yield, opti-
mising irrigation, detecting soil content, monitoring crops, and weeding.
From another point of view, intelligent systems can process information,
provide complex reports, and serve farmers in decision making and
complying efciently with quality requirements. Consequently, AI has
the potential to provide essential solutions to address different chal-
lenges in this industry and will make it possible to produce better results
more effectively. The diverse form, in which AI can intervene in the
agriculture sector, operates through the usage of new information,
communication technologies, and internet of things.
Precision water management
Precision water management (PWM) is a data-driven approach in
agriculture that optimizes water usage by applying water precisely at the
right time, place, and crop growth stage. By leveraging technology and
data analysis, PWM aims to conserve water resources, promote sus-
tainability, and enhance overall farming efciency [39]. This approach
involves sensible use of water to achieve sustainable water consumption.
In agriculture, PWM refers to accurate and appropriate application of
high-quality water at the proper time, place, and crop growth stage. Due
to numerous technologies and instruments, several ways have been
proposed to achieve such objective.
Therefore, AI is pivotal role in revolutionizing water precision
management in agriculture lies in its ability to conduct data analysis and
real-time monitoring, thus optimizing irrigation practices for sustain-
able farming. The transformative power of AI in this domain can be
attributed to the following key aspects:
•Insights from Geospatial Data: AI and big data technologies analyses
geospatial data to offer valuable information on soil moisture levels,
weather patterns, and crop water requirements [40].
•Real-time Sensor Data and Weather Forecasts: By integrating geo-
spatial data with real-time sensor data and weather forecasts, AI
enables more accurate and efcient irrigation scheduling [41].
Farmers can rely on AI’s analysis to determine optimal irrigation
schedules and amounts, avoiding over- or under-watering their
crops.
•Identifying Areas of Inefciency: AI-powered sensors monitor water
usage on farms, detecting inefciencies and patterns. This helps
farmers identify areas where water is wasted or used inefciently,
allowing corrective measures to be taken [42].
•Reducing Water Wastage: AI ne-tunes irrigation practices to mini-
mize water wastage. By precisely controlling the application of
water, AI enables farmers to maximize crop yields while minimizing
water usage, which is particularly crucial in regions with limited
water resources or unpredictable weather patterns [43,44].
Many technologies have been developed to control the communi-
cation between machines and different nodes settled in agricultural
farms. These Machine to Machine technologies are efcient for moni-
toring soil moisture content and temperature at periodic intervals to
automate the irrigation with precise requirement [45]. Remotely
controlled sensors can be used to observe both biological and climatic
conditions [46,47].
Furthermore, ML becomes a valuable tool, with the aid of measured
real time data from the agriculture farm, for making right decisions for
Table 1
Literature survey on water management models.
Ref. Inputs Algorithms Method model Technology Performance
Choudhary et al. [61] Climatic conditions, soil
moisture content
Partial Least Square
Regression (PLSR)
Evapotranspiration
model
Economic hardware,
sensors, (IoT).
Increased efciency and
economic feasibility
Anand et al. [62] Temperature,
soil humidity
Fuzzy Logic Controller Penman–Monteith
model
Wireless Sensor Networks
(Sensor nodes, hub, and
control unit)
Automated drip irrigation
water conservation
Subathra et al. [63] Climatic conditions, soil
moisture content
topography
ANN method Soil moisture model _______ Precision and Robustness of soil
moisture prediction, water
saving
Chen et al. [64] Soil water content and
meteorological data
Convolutional neural
network-)
Pearson correlation,
soil water content
autocorrelation
Deep learning
Near-infrared (NIR)
spectroscopy
Prediction accuracy ninety-three
%
Arvind et al. [65] Moisture, weather
forecast and water level
Machine Learning
algorithm –
___________ IoT, ZigBee technology,
Arduino microcontroller
Drought prediction
Poblete et al. [66] Meteorological data, soil
composition
Articial neural
network (ANN)
Machine learning
techniques
Evapotranspiration
model
Unmanned aerial vehicle
(UAV)
remote sensing platforms
Performance to predict water
stress
Melit and manghanem
[67]. Melit and
Benghanem[68]
Different conditions ANN networks Optimal model sizing Hybrid intelligent systems
(HIS)
Sizing of optimal stand-alone
photovoltaic systems
Richards and Cnibeer. [69] Different conditions Regression
comparison
Optimal model sizing Standalone power supply
(SAPS)
Seasonal Variability of Solar
Insolation
(PV) panels with H2 storage
Hernandez and Medina
[70]
Different conditions and
inputs
Genetic algorithms Optimal model sizing Sizing grid-connected PV-
system
Stability voltage distribution
Ammmar and Oualha [71] Climatic data Feed Forward Neural
Network
Adaptive Neuro Fuzzy
Inference System
Optimal model sizing Solar pumping systems Photovoltaic power forecast
Achite et al. [72] meteorological and
hydrological
ANN, ANFIS, SVM, and
DT
Hydrological Drought
Modeling
machine learning
techniques
ML accurately predicted
drought, with SVM
outperforming
Chandel et al. [73] Crop data and images AlexNet, GoogLeNet
and Inception V3
Water stress modeling Deep learning GoogLeNet achieved remarkable
accuracy
A.A. Mana et al.
Smart Agricultural Technology 7 (2024) 100416
5
improving water usage efciency and managing evapotranspiration
process. As summarised in Table 1, ML allows for a correct and effective
use of resources. One of the essential aspects of digitalization is the
large-scale use of wireless sensor networks centrally controlled by ML.
Furthermore, the commercialization of thermal cameras, facilitated by
advancements in graphics and high-speed real-time computer process-
ing, has opened new opportunities for estimating soil hydraulic condi-
tions through the acquisition of thermal indices [48]. With recent
advances, AI-based reasoning about soil water balance and forecasting
can optimise hydraulic variables and protect land against erratic cli-
matic conditions and disasters [49,50].
Furthermore, solar photovoltaic water pumping has become tech-
nologically mature and commercially acceptable for water pumping in
the agricultural sector in rural areas. A great advantage of using solar PV
based water pumping is that the water requirement and sunshine
availability timings coincide. However, due to the intermittent nature of
solar radiation, more than one energy sources can be integrated to
assure continuous energy availability for irrigating the crops. Moyo [51]
presented a comprehensive AI-based modelling of a solar/diesel hybrid
water pumping system with the objective of optimizing its performance.
The results revealed that the ANFIS-based MPPT system was able to
generate maximum energy from PV modules under prevailing weather
conditions [51]. Furthermore, hybrid power systems reduce life cycle
costs relative to standalone power systems and also provide continuous
and reliable energy [52]. Karar et al. [53] used the internet of things to
minimise water wastage in irrigation process based on meteorological
data measured through sensors (ambient temperature, relative humid-
ity, soil moisture, etc.) and the multilayer perceptron neural network
approach. The proposed model could manage sensor data to automati-
cally control the operation of the water pump. For sustainable irrigation,
the management information system along with the online adaptation of
climatic conditions help improve crop productivity and reducing the
overall cost [54]. Abidin et al. [55] used moisture content data in the soil
with an intelligent irrigation control system to reduce the use of water
usage for cultivation. The technology of wireless sensor networks was
used in the agriculture sector to promote Precision Agriculture (PA)
[56]. Wireless sensors with fuzzy controllers [57] have been recom-
mended for the automation of the irrigation system [58]. Xiao and Liu
[59] proposed the use of microcontroller units to facilitate a smart
irrigation system. Karar et al. [53] used a smart controller based on MLP
neural networks for water irrigation system [60].
Integrated food safety
Conventionally, agricultural productivity and crop diversication is
strongly linked to protection against weeds and diseases or infestation
by pests and insects. In the face of these risks, food security becomes
crucial [74]. Thereby, the faster the detection, better will be the
implementation of security parameters and the measures necessary for
the preservation of crops [75]. In recent years, ML and DL have been
applied to protect through insect pest monitoring [76], weed detection
[77], and identication of plant diseases [78].
Machine learning, using remote sensing to recognise species and
diseases [79], is applied using decision trees, random forests, and neural
networks to extract features and object classication. Deep learning has
emerged as useful method with big data and visual technologies. Con-
volutional neural networks are the easiest type of DL to process 2D
images with fewer errors but depend on high volumes of measured and
expert data sets (Fig. 3). The AI-based disease detection process involves
image collection, image labelling, data splitting and storage, and
dividing the data set into training, validation, and testing data subsets.
The model is trained and validated using the data subsets, as dened,
and then the model results are tested against the third data set to provide
the decision on whether the disease exists in the crop or not.Deep CNNs
have gained the interest of researchers in intelligent integrated man-
agement. CNN based on deep learning, coupled with remote sensing and
big data, is faster and more dependable [80,81]. Details of input data,
algorithm, performance, and limitations, of using ML approaches for
weed and disease detection are summarized in Table 2. It is observed
from the Table that imagery data along with appropriate ML method
provides the most accurate detection of the presence of weeds in the
crops.
Precision livestock farming
Monitoring the health of livestock plays an important role in modern
agriculture by ensuring the welfare of animals and producing high-
quality products. Advanced articial intelligence techniques can use
sensors and cameras to monitor animal health in real time, dropping the
need for traditional training. Compared to conventional methods, AI-
based systems have several advantages, such as identifying patterns
and potential health problems before they become serious. [94,95].
These systems can be based on various models and technologies such as
IoT, edge computing, and Distributed Ledger Technologies [96]. How-
ever, the implementation of such a system comes with several potential
concerns, including initial cost, technical expertise and requirements for
specialised equipment, ethical concerns, and doubts about job relocation
of jobs [97]. Despite these challenges, leveraging AI for Precision live-
stock farming systems can signicantly enhance animal welfare and
decision-making process. By providing information on animal
Fig. 3. Banana disease detection using intelligent algorithms [82].
A.A. Mana et al.
Smart Agricultural Technology 7 (2024) 100416
6
behaviour, feeding, and environment, AI helps farmers optimize their
livestock management systems.
In recent years, the need of establishing comparable standards in
large-scale livestock farming systems has been discussed to improve
decision-making and data interchange. Both farmers and farm-
integrated approaches were advocated for adoption, along with a
consensus on the requirements for data exchange [98]. This would allow
farmers and regional stakeholders to realize the benets of sharing data
effectively.
Studies on energy usage in dairy farms have been explored, high-
lighting the signicance of prediction models to analyse energy con-
sumption and evaluate the effects of modications in infrastructure
equipment and management practices. The literature commonly reports
a reduction of 35 % in energy usage with the adoption of grazing-based
dairy systems [99]. Various methods have been used to forecast energy
consumption in dairy farming, including the CART decision tree, the
random forest ensemble, articial neural networks, and support vector
machine [99,100].
Furthermore, animal health is also a key aspect of livestock pro-
duction. The monitoring can focus on sound analysis that have the po-
tential to be automated for large-scale farming, thus providing an
efcient and cost-effective way to track animal well-being [101]. The
use of machine learning algorithms, including face-face recognition
through convolutional neural networks became widely used. For
example, the review [102] identied relevant sensors to measure animal
health, such as cameras (2D and 3D), microphones, thermistors and
accelerometers, and highlighted how these technologies can be used to
improve pig health, leading to better outcomes for both animals and the
industry. In summary, these studies prove the potential of advanced
technology in promoting animal welfare and improving the efciency of
livestock farming.
Crop productivity and fertility
Precision agriculture or site-specic crop management is dened as
an agricultural system that uses technology, satellite and aerial images,
climate forecasts, and prediction applications to improve the produc-
tivity and the protability indicators of the elds. Based on collected
data, AI could foster agro-technologies and increase crops quality, pro-
ductivity, and hence the protability. ML makes it more achievable by
learning from the analysis of measured data and performing agricultural
production with a high degree of accuracy (Fig. 4)
For example, AI may response to crop health issues or soil nutrient
deciencies based on the measured data [103]. AI techniques can
examine photo-sanitary models, soil health, and the amount of fertilisers
required [104]. Therefore, the risks of plant and the soil degradation
may reduce, and yields can cope with the market trends, maximize the
income from different farms [105], and ensure better crop mapping for
decision-making (Fig. 5) [106]. Under the recent Soil monitoring sce-
narios, once farmers submit a sample of their agricultural soil to the
monitoring agency, they will receive a detailed summary of their eld
soil contents. Based on the results obtained, an appropriate deci-
sion/action is taken and communicated to the farmer about the presence
and type of bacteria, fungi and wide-ranging microbial progression
[107].
AI for harvesting, pruning and ploughing
AI-based robots have revolutionized agricultural tasks, offering
advanced capabilities in areas such as harvesting, pruning, and plowing.
These robots have brought signicant improvements to farming opera-
tions, resulting in increased productivity and efciency. One remarkable
example is the development of a sowing, pruning, and harvesting robot
designed to work efciently in dense vegetation. This small and exible
robot minimizes its impact on the environment and exhibits impressive
obstacle avoidance capabilities, reducing operating time by 49 %
Table 2
AI performance and limitations for weed and disease detection.
Application Inputs Method/algorithms Performance Limitations Ref
Weed
detection
UAV images Fully Convolutional Network
(FCN) method
Weed mapping: 94 % weed
recognition: 88 %
Requires vast human
expertise
H. Huang et al.
[83]
Weed
prevention
multispectral, hyperspectral
cameras, and GPS data
ROBOTs. Sensor machine
learning
Saves time and removes resistant
weeds.
Expensive and affect soils Brazo. [84]
Weed
prevention
Yield sensing and imagery data Colour-based and Texture
Based algorithms;
High accuracy 92.9 % Expensive Sujaritha et al.
[85].
Weed
detection
(RGB)/ hyper spectral images Deep Convolutional neural
networks
High accuracy 98.23 % Requires big data. Assad and Bais.
[86]
Disease
detection
UAV images (RGB) CNN Overall accuracy 89 %,94 % Requires big data and
human expertise
Bah and Canals.
[87]
Disease
detection
Multispectral Imaging and
sensing data
Phenotyping technology,
remote sensing methods
Early season detection and
performance
Require big data Ampatzidis and
Partel [88]
Disease
detection
Expert systems Web-Based Expert System High performance Internet dependence Beiranvand . [89]
Disease
detection
UAS images
Data Base
CNN an average accuracy of 93.75 % Image segmentation affect
CNN acuraccy
Junde Chen et al.
[90]
Weed
detection
Digital Image Remote sensing methods real-time, submeter- or even
decimeter-level accuracy
– Perez-Ruiz el al
[91].
Weed
detection
Hyper spectral images SVM, ANN, and CNN Quick detection. Accepted accuracy Che’Ya [92]
Weed
detection
Thermal images,Big data ANNs Performance. Reduces trial and
error.
Requires big data, expensive Zamani and el. [93]
Fig. 4. Site-specic crop management based on a three-dimensional approach
that assesses inputs and outputs from elds to watersheds and regional
scales [8].
A.A. Mana et al.
Smart Agricultural Technology 7 (2024) 100416
7
compared to traditional controllers [108]. Precision farming, which uses
appropriate technology and practices to produce high-quality agricul-
tural products, is one area where AI can be applied. Here are some ex-
amples of how AI can be used for harvesting, pruning, and plowing:
- Smart Spectrometer: This is a spectrometer with integrated articial
intelligence that can estimate properties such as substance concen-
trations and compositions. It can be integrated into a harvesting
vehicle, where quality is determined by predicting sugar and acid in
grapes in the eld [109].
- Complexity-Driven CNN Compression: This is a type of model
compression that can accelerate Convolutional Neural Networks
(CNNs) on low-power devices. It can be used for pruning CNNs for
resource-constrained edge AI [110].
- Selective Harvesting Robotics: This is an area of research that aims to
develop robots for selective harvesting. The task of selective har-
vesting is not easy for robots, but it allows for improved farm man-
agement and can optimize the food-production chain [111].
- Pruning and Harvesting Manipulators: These are manipulators used
in the agricultural robotics eld. They can be used for pruning
grapevines and apple trees, as well as harvesting strawberries, to-
matoes, apples, sweet-peppers, and iceberg lettuce [112].
- Plant Counting with UAV RGB Images and Deep Learning Networks:
Utilizing Unmanned Aerial Vehicles (UAVs) equipped with RGB
cameras and deep learning networks, AI demonstrates its capability
in accurately counting plants within agricultural elds, such as rice
paddies. The methodologies employed—image processing and deep
learning algorithms—form the foundational technology that extends
beyond plant counting. These methodologies serve as the basis for
various AI-driven agricultural tasks like harvesting, pruning, and
plowing, leveraging similar data processing, decision-making algo-
rithms, and precision achieved through advanced image analysis
[113,114].
These are just a few examples of how AI can be used for harvesting,
pruning, and plowing in agriculture. As AI technology continues to
advance, we can expect to see even more innovative applications in
this eld.
Sustainable AI based agriculture
Predictive analytics for smart energy planning
Predictive analytics is a powerful tool that empowers farmers to
optimize their energy usage and make informed decisions in agriculture.
By analyzing data from diverse sources, predictive analytics offers
valuable insights on crop selection, optimal planting times, appropriate
fertilizers, irrigation schedules, and pest or disease management. One
application of predictive analytics in agriculture is smart energy plan-
ning, promoting the use of renewable energy sources and energy con-
servation techniques to enhance sustainability and decrease fossil fuel
dependency. Smart agriculture further leverages predictive analysis,
integrating Agro IoT systems, renewable energy sources, and vertical
farming techniques to increase yields while minimizing water and en-
ergy consumption, resulting in more sustainable farming practices.
The integration of AI-driven energy management and IoT-based
weather forecasting holds immense potential in transforming agricul-
ture practices for greater efciency and sustainability. Utilizing
machine-learning algorithms, AI accurately predicts energy demands,
detecting potential inefciencies in energy usage. With historical data
analysis, farmers can proactively allocate energy resources, minimizing
environmental impact. The coupling of IoT with meteorological sensors
enhances weather forecasts, surpassing conventional methods’ limita-
tions and providing predictions that are more precise. Proposed solu-
tions incorporate deep neural networks and CNNs, supported by
meteorological satellite data, to bolster weather forecasting. Recurrent
neural networks improve long-term climate modeling with short-term
memory capabilities [115]. Managed through CPU or cloud platforms,
the collected data ensures precise results, accessible via mobile appli-
cations for real-time monitoring and decision-making in agricultural
activities [116]. This fusion of AI-driven energy management and
IoT-based weather forecasting presents a transformative approach to
sustainable and efcient agriculture practices. Hybrid models
combining CNN with RNN architectures may enhance accuracy in
managing both spatial and temporal data [117]. Table 3 summarizes
diverse energy predictive solutions.
AI-based energy management for agriculture products
To modernise the agriculture sector, electrication may be an
effective approach and can result in environmental and economic ben-
ets [126]. In the present times, the agricultural energy internet (AEI)
concept is getting popular in developing relevant agriculture technolo-
gies. AEI is relatively a recent development and was derived from Energy
Internet to accomplish clean energy generation for isolated places. AEI is
a multi-energy system realized by combining power networks, heating,
gas and requires agricultural information sharing by the people [126].
AEI supports new energy industries, smart agriculture solutions, and
rural revival, Fu et al. [127]. The intelligent management and control
systems adopt standard methods for collecting the current, voltage,
electrical fault, and power classication data to carry out the real-time
safety analysis and issue warning, if any for the safety of agriculture
production, Song et al. [128]. Fu and Yang [129] provided guidelines for
protecting crops and trees from low temperatures and agricultural
deciency. The multi-source data fusion concept is useful for moni-
toring, positioning, and navigating the greenhouse environment to
Fig. 5. Crop yield map using machine intelligence algorithms [106].
A.A. Mana et al.
Smart Agricultural Technology 7 (2024) 100416
8
optimize and improve agriculture produce. For risk assessment, it is
essential to predict the meteorological variables to safeguard the growth
level of agricultural products. Mancipe and Guti´
errez [130] proposed a
data fusion strategy to predict meteorological parameters to precisely
monitor the growth of agriculture products.
Sparrow et al. [131] provided proper strategies may be needed when
AI begins to affect the agriculture sector. The study pointed out some of
the consequences which may arise due to the implementation of AI
technology in agriculture sector. According to the authors, AI may be
benecial for farmers, end users and the environment but may present
unknown risks and suggested some alternative designs and regulatory
procedures have been suggested to the risks. The civil and principled
impacts of using AI in the farming were studied to nd out how are they
correlated with AI ethics [132]based on sustainability, trust, privacy,
benets, transparency, freedom, responsibility, justice, morality, and
unanimity.
Vincent et al. [133] mentioned that currently some of the farmers are
using automated equipment fed through huge data collected by mete-
orological sensors and satellite imagery. These farmers act in accordance
with the advice provided by sophisticated computer applications.
Existing investment on precision agriculture assures an important role of
AI in agriculture [134,135]. The ML methods are expected to address
key research topics in agriculture sector, which include meteorological
parameter prediction ahead of time, economic modelling, and plant and
animal breeding. Furthermore, AI and ML methods have the potential to
improve distribution, balance energy consumption loads, and manage
uctuations in renewable energy production. The energy transition can
incorporate AI into their system for more opportunities to improve the
efciency of production and consumption.
Moreover, it is agreed that improving the efciency of biomass
processing could help agricultural regions to produce bioenergy and
addressing the challenges that hinder biomass-based energy develop-
ment [136]. Practically, agricultural biomass can feed large scale of
bioplants (100 MW and more) [137,33]. Thus, pyrolysis and gasication
technologies been widely developed [138], explaining the reason for
huge AI studies coupled with these systems [139,140]. Combined heat
and power (CHP) technology can be widely adopted for waste man-
agement and energy production [141]. CHPs are identied for their
promising efciency in modern greenhouses, compared to biogas plants
[142]. Considering the progress of cogeneration systems, if integrated
with AI methods, can be an effective and potential solution to
compensate for the different energy demands of agricultural green-
houses [143,144]. Furthermore, bioenergy systems have proven per-
formance, as well as combined and hybrid with other green technologies
[145,146]. Generally, the potential benet of AI is to simultaneously
supply electricity by controlling and monitoring main equipment, input,
and output parameters [147].
Nanotechnologies utilization for agricultural management
The exponentially increasing population, adverse effects of climate
change, growing biofuel demands, and deteriorating soil condition are
some of the alarming international food security issues. Exploring new
and sustainable options requires modern techniques to emulate infor-
mation from materials science and automation [148]. Emergence of
precision agriculture with nanotechnology and AI, offers excellent av-
enues for sustainable food production. The third Green Revolution of the
1950s and 1960s improved agriculture production and minimized the
scarcity of food and spread of malnutrition. Since then, the global
population has crossed 6 billion mark and has compelled an increase in
agriculture production. Today, this sector is facing challenges such as
decreased yields, soil quality, freshwater availability for irrigation,
fertility, and excessive use of pesticides and fertilizers [149].
For the development of the precise and sustainable agriculture sector
development, nanotechnology can offer excellent opportunities, dis-
cussed in review articles covering strategies to improve crop nutrition
and develop smart plant sensors [150,151] . Nanotechnology can
facilitate the delivery of fertilisers to tissues and organisms in a
controlled way [152], which would be benecial for plant growth and
optimal use of fertilizers and pesticides and minimise adverse effect on
soil condition [153,154]. Furthermore, nanotechnology applications in
agriculture include the plant sensor development through which the
plants can itself sense abiotic stress depending on the directed delivery
of nanomaterials [155]. Four main areas in which nanotechnology is
progressing include improving production yield, soil conditions, and
efciency of materials usage, Fig. 6 [148].
Additionally, the integration of nanotechnologies with AI-driven
methodologies amplies these advancements in agriculture. AI com-
plements nanotechnologies by providing intelligent data analysis, pre-
dictive modeling, and autonomous decision-making capabilities. The
fusion of nanotechnology and AI promises novel opportunities in pre-
cision farming, allowing for real-time monitoring, precise resource
management, and informed decision support systems. This collaboration
aims to revolutionize agricultural practices, enhancing productivity,
sustainability, and efciency in the face of global agricultural challenges
Discussions: issues and challenges related to the use of AI
As perceived before, AI can change our traditional view of the
agricultural sector and enable farmers to be agents of change, especially
in rural areas. However, AI remains a vast eld that operates in a way
Table 3
Performances and limitations of AI on predictability.
applications Technologies Inputs Performance Limitations Ref
Energy modelling AI and ANFIS Energy, climatic and
agricultural data
High accuracy using Hybrid learning method _ Ashkan et al.
[118]
Smart energy inputs ,Smart energy
pacages,ANN
Nitrogen,fuel,manure,
and electricity
Reducing 17.3 % of energy and 23 % of GHG
emissions
_ Elahi et al.
[119]
Agrivoltaics energy
prediction
Regression, ANOVA Weather and crop data Predicted LER value of 2.17 relies on the proper
arrangement of PV panels and crops in the research
region.
Low accuracy Abidi et al.
[120]
energy consumption
in greenhouses
Baysian model Climatic data and
greenhouse dimensions
The CBMA model outperformed BMA, MLP-SEOA,
MLP-SCA, MLP-BA, MLP-PSO, and MLP models.
_ Ehteram et al.
[121]
Early warning systems IoT and ML Climatic data Reduce the number of tasks and data completeness
of routing in a larger coverage agricultural
greenhouses.
Data collection Liu et al.
[122]
Weather forecasting ANN Metrological
parameters
Acceptable error percentage, fast prediction.
prediction results till 2050
Increasing of the percentage
predicted errors with time. Yahya and
Seker. [123]
Load forecasting ML, IoT Different parameters Acceptable errors,
Fast prediction.
Short term load prediction Raju and
Laxmi. [124]
Load forecasting Multi Linear
Regression (MLR)
Different parameters Higher accuracy Short term Kim et al.
[125]
A.A. Mana et al.
Smart Agricultural Technology 7 (2024) 100416
9
that is dependent on several factors relating to the nature and quantity of
data collected, modelling, design, type of algorithm used, and is highly
dependent on the way in which is all applied. All these elements can be a
hindrance to the development of AI in the agricultural sector. However,
the advantages of AI are unavoidably accompanied by several risks.
Currently, the issues and difculties raised by using AI in agriculture are
still not inevitable. In this section, the relevant challenges of AI appli-
cations in agriculture are highlighted.
Lack of trust
The question of whether we are capable of placing the total trust in
agricultural and food processing machines is the subject of a long debate
[156–158]. Today, agriculture industry is undergoing a digital revolu-
tion, but lacks the intelligence to delegate all agricultural tasks to
technology and robots.
Today, intelligent systems may help to optimize performance,
manage risks at all levels and if possible, innovate. From a security
perspective, agriculture may be at risk of conict and war due to these
systems’ susceptibility to hacking and cyber assaults. This raises the
question of entrusting decision making to algorithms based on meteo-
rological, biological, or energetic data.
Thus, experience has shown that AI has disrupted the processes of
many elds such as banking, industry, and medicine. Relying on ma-
chines and decision support systems can pose major ethical and prob-
lems and programs may have a "cold" logic. The limitations of trusting
these algorithms in agriculture can be seen as the emerging self-driving
car technology. It will be difcult to dene ethics for insurance to cover
liability in case of fatal decisions and accidents. The need for normative
rules for these innovations is the next step soon.
Impact of articial intelligence on the workforce
The ability of AI to replace humans in performing cognitive tasks is
one of the critical issues. Digitalization of agriculture will make it
possible to our production-related tasks at a lower cost and in a shorter
time, which will change the function of farmers and reduce their
numbers to the minimum possible. The digitalization of activities and
relationships creates new dangers for occupational health and safety
(OHS), such as an increase in mental load and a blurring of personal and
professional boundaries. These signicant technical advancements
cause people, especially older workers, to doubt their knowledge, which
can put them in situations where they are incompetent and cause sig-
nicant personal destabilization. As a result, learning new technologies
Fig. 6. Nanotechnology Applications in agriculture [148].
A.A. Mana et al.
Smart Agricultural Technology 7 (2024) 100416
10
or procedures through training can be challenging, and this can lead to
feelings of frustration and failure that may demotivate people, creating a
vicious cycle of professional disinsertion. Good change management will
consist of anticipating these psychosocial risks, dening, and imple-
menting an approach that allows the project to be implemented without
signicant disruption to new working contexts, tasks/responsibilities,
and skills.
AI application in the workplace creates new concerns about occu-
pational dangers and safeguarding farmers from the effects of the
changes in labor [159–161]. Psychological dangers, such as emotions of
reliance, loss of autonomy and identity, and extra mental load, are
linked to human-machine interaction, in addition to the physical risks
brought on by malfunctioning robot/s. With legal, ethical, social, and
regulatory issues at stake, such as the emergence of technostress, the
exploitation of personal data in the workplace, the security and
decision-making latitude dimensions of human-machine interactions,
the transparency of monitoring algorithms, etc., there is growing
concern about the risks of articial intelligence and machine learning in
the workplace. This concern comes from its integration to automated
machines and in farmer monitoring.
The question of data-driven algorithms
The future of smart agriculture lies in the efcient collection and
analysis of data. Data are not readily available, particularly at a local
farm scale [162], and if available could contain uncertainty. The
measured data contains information and trends about weather, soil,
crops, water resources and more could be extracted and used as decision
support for farmers, researchers, agricultural advisors, and market ser-
vices. Development of an open-source database, at a global and local
scale, will serve as the baseline for scientists, economist, and farmers.
This database will contain satellite imagery, Internet of Things (IoT)
sensors data, soil, crops, water, tillage and surface temperature data [39,
163]. At the data collection stage, IoT network can help collect data
measured from sensors located in the eld, in the soil probes,
tower-mounted devices on tractors, providing real-time accessibility.
The next stage concerns the integration of collected information with
data from cloud-based systems (soil types, present and future weather
conditions, cost models, etc.) to extract insights and patterns by machine
learning models. These predictive models help farmers and scientists to
detect existing and future issues. The challenge now lies in promoting
global efforts for the availability, accessibility and usability of data in
agriculture. Integrating these stochastic algorithms in deterministic
approaches such as biophysical models is a hot topic that research is
working on. Scientists are actively working on the integration of the
physical aspect in the ML algorithms for making the whole system more
realistic application. There is also signicant legal ambiguity around
machine learning in general, as with any breakthrough. Although there
are still standards that must be followed, the sector is evolving, con-
cerning the concepts of ethics and accountability.
The question of interpretability
Good design may be able to reduce some of the hazards mentioned
above, making them important considerations for those who develop AI
for agriculture. The characteristics of AI, such as its proneness to bias
and algorithms nature, are linked to several the problems involved with
its application in agriculture. In fact, concerns about interpretability
have recently arisen in response to the emergence of machine learning.
Some machine learning algorithms operate in a very opaque manner and
their conclusions are still not fully explicable or justied [164,165].
Therefore, interpretability may also be of an instrumental relevance for
several reasons, including the right to explanation of stakeholders who
may be impacted by an ML decision in agriculture.
AI and ML models, however powerful it may be, but still considered
mysterious and black boxes. Now, it is difcult to measure and justify
their results or outcomes. Interpretability is the inherent issue with the
use of AI. There are two distinct levels of interpretability in machine
learning models.
1) Low interpretability: This includes ML models such as support vector
machine (SVM), neural networks, and deep learning. The lack of
interpretability is justied by using structure of multiple inter-
connected layers containing different types of neurones; in the case
of deep learning or complex geometrical foundations; in the case of
SVM.
2) High interpretability: This level includes the classical regression al-
gorithms such as linear, multiple linear, decision trees, Ridge and
least absolute shrinkage and selection operator Lasso regressions.
Although these models are inherently interpretable but there is a
limited relationship between explain ability and the accuracy of the
predictability in the future.
To this end, it is essential to avoid technological determinism while
considering the future of agriculture. This is not meant to ignore the
possibility of agricultural uses being facilitated by technology, but
rather to improve intelligent systems and adapt them in the agro-food
sector.
Explainable AI (XAI) is a branch of AI that has been specically
developed to ensure transparency and comprehensibility. This is
particularly important in elds such as agriculture, where decisions
made by AI systems can have signicant impacts on crop yields, water
usage, and overall sustainability [166]. The use of XAI in smart agri-
culture can help build trust in AI-based technologies, identify potential
biases or errors, and ensure that these technologies are transparent,
accountable, and in compliance with relevant regulations and standards.
Specic issues of use of AI in agriculture sector in developing countries
The agriculture sector has gone through several transformational
revolutions throughout history that have signicantly affected produc-
tion and efciency. From the rst agricultural revolution, which enabled
people to settle down around 10,000 years ago, to the most recent
agricultural revolution called "digital agricultural revolution". This
revolution is characterized by the integration of IoT technologies and
the emergence of Big Data to connect agricultural systems, resulting in
unprecedented levels of efciency and productivity. All These changes
have played a critical role in the development of agriculture. However,
the development and adoption of smart agricultural technologies in
developing countries is signicantly lagging and varies depending on
the country and region. The main obstacles to the adoption and use of
intelligent agricultural technologies in developing countries include a
lack of nancing (initial investment), low awareness of modern agri-
cultural technologies and processes, limited technical capabilities, and
restricted access to information sources on agricultural technologies.
Cultural factors, biases, and traditional agricultural practices also play a
role in resistance to the adoption of intelligent agricultural technologies.
Furthermore, the lack of coherent policies and legal frameworks to
support the use and adoption of intelligent agricultural technologies is a
signicant obstacle. Developing countries do not have the same re-
sources or expertise as developed countries to support research and
innovation in intelligent agricultural technologies. In addition, farmers
may be limited by low-income levels, which hinder their ability to invest
in agricultural technologies.
In this section, a roadmap is presented which outlines the crucial
issues surrounding AI and agriculture in developing countries. The aim
is to provide a thorough understanding of the questions being posed and
their relevance for researchers, businesses, and policymakers who are
not well versed in agriculture and rural realities. It should be noted that
while the section highlights important issues and provides an overview,
it does not explore specic case studies for each country.
A.A. Mana et al.
Smart Agricultural Technology 7 (2024) 100416
11
Logistic issues
From a logistical point of view, the provision of low-cost agricultural
machinery by AI-based platforms enables farmers to access on-demand
tractor services and optimise their resources [167,168]. Several tech-
nologies can help farmers who cannot afford to buy equipment to
respond to the uncertainty caused by climate change in a faster and more
cost-effective way.
In addition, AI and algorithmic decision-making systems used in
agriculture can be applied to improve efciency in various sectors and
have been used in the implementation of complex public procurement.
These systems can also help reduce fraud and corruption and remove
excessive and burdensome bureaucracy. Developing countries can
therefore build on decision-making systems in agricultural administra-
tion, logistics and supply chains, including input programmes, to
improve efciency and accountability [169].
To conduct a comprehensive study on improving agricultural logis-
tics, it is essential to gather data on various aspects such as the number
and locations of logistical services, their capabilities, challenges, and
benets. Additionally, the research must explore opportunities for
scaling these services and examine the government’s role in enabling
them.
It is also crucial to review existing regulatory frameworks that apply
to public infrastructures like satellites and advanced technologies. These
frameworks need to address not only the capacity benets but also safety
and security challenges posed by these systems. Furthermore, the study
should assess how Agricultural Decision-Making (ADMs) can enhance
the efciency of logistics operations, minimize corruption, and
contribute to more inclusive and equitable schemes. This analysis would
help identify areas where ADMs could be implemented to streamline
logistics operations and improve their overall effectiveness in support-
ing agricultural production. By giving smallholder farmers access to
market data, pricing, and demand, ADMs can aid in the creation of more
inclusive and equitable programmes. This can assist farmers in more
protably planning their produce. ADMs may also be used to identify
farmers who may be facing food insecurity, enabling policymakers to
provide these farmers specialised assistance. This may result in more just
and environmentally friendly agriculture systems that are advantageous
to both farmers and consumers.
Gender inequalities
The challenges faced by women in agriculture are multiple and
complex and they are particularly widespread in developing countries.
Women farmers face numerous obstacles that limit their productivity
and potential, including unequal land ownership, limited access to
capital, training, and agricultural inputs, as well as cultural discrimi-
nation based on gender [170].
One of the most signicant challenges faced by women farmers is the
lack of access to land ownership. In many countries, women are not
allowed to own or inherit land. This limits their ability to invest in their
farms, make long-term plans, and access nancing. Access to capital and
nancial services is another major challenge for women farmers.
Women are often excluded from formal nancial services such as loans
and credit due to their lack of collateral and limited nancial knowl-
edge. This limits their ability to invest in their farms, purchase inputs,
and access markets.
The use of AI algorithms in agriculture can potentially exacerbate
these inequalities. AI algorithms are often trained on biased datasets,
which can perpetuate existing gender discrimination in agriculture.
However, responsible use of AI can also help to mitigate some of these
inequalities and make agriculture more accessible to women.
To promote gender parity in agriculture participation, intentional
policies are needed to address discriminatory anomalies and use AI
technologies to include women in new forms of work, entrepreneurship,
and innovation. This could include policies aimed at increasing women’s
land ownership, improving access to nancial services, providing
training and inputs, and promoting the use of AI to connect women
farmers with markets and buyers.
Overall, it is essential to address the challenges faced by women in
agriculture to promote sustainable development and reduce poverty. By
encouraging gender parity in agriculture participation and using AI
technologies responsibly, we can help to ensure that women farmers
have the resources, income, and control over their own projects that
they need to succeed.
Innovation and data access
The agricultural sector has been transformed by the emergence of
technology and the improvement of analytical tools. These advance-
ments have opened new opportunities for innovation, which are already
being observed and are likely to multiply. The use of cutting-edge
technology is not always necessary, as innovation can be achieved by
increasing yields to generate data, digitizing physical assets, integrating
data across industries, exchanging data, and encoding unique capabil-
ities. However, the use of large and valuable data faces constraints,
including limited access and capacity in emerging economies and
contractual rules that restrict public access to data [171].
In agriculture, increasing yields is the most relevant approach to
generating data, with sensors installed on the equipment to improve
accuracy and inputs. These data can be used to improve the design,
operation, maintenance, and repair of assets, creating new and better
services and business models. However, the use of big data and AI
technology raises concerns about user privacy, which must be resolved
before these innovations can be implemented on a large scale, particu-
larly in Africa.
In the context of AI in agriculture, the exclusive property rights of
data and their implications are illustrated using digital information on
sequences. Although these advancements have transformed genomics
and gene editing, the implications for agriculture and livestock have not
received enough attention. At the macroeconomic level, the circulation
of information and the governance of information on digital sequences
are a current issue.
Taking the example of Africa, the use of AI in agricultural infra-
structure raises several crucial questions [172,173]. These include how
to promote the use of data for African agricultural innovations, how
farmers and agricultural businesses can create value for their customers
using data and analytical tools they own or could have access to, and
who should own the data collected on farms. It is essential to address
these questions to successfully implement AI in ourishing countries and
develop new and innovative services and agricultural models.
Climate change
Global energy consumption and greenhouse gas emissions have
skyrocketed due to the widespread and extensive use of digital tech-
nology, which is a major factor in climate change. Digital technology
alone is predicted to use 20 % of global power by 2025 and produce 14
% of greenhouse gas emissions by 2040, therefore this trend is projected
to continue [174] . Huge volumes of data storage demand a lot of power,
which greatly contributes to climate change, the effects of which are
already disproportionately felt in developing nations.
It is critical to investigate how AI and data initiatives could help
developing countries cope with the consequences of climate change to
solve this issue. However, it is signicant to emphasise that, compared to
other nations, ourishing nations are more consumed with pressing
problems such as famine, drought, and political instability, and there-
fore measures to ameliorate the consequences of climate change
generally take longer to develop and are less important. Investigating
alternate data compression techniques might help eliminate the need for
expensive, power-guzzling machines and data farms.
Furthermore, it is important to focus on raising awareness and pro-
moting sustainable practices. This can be achieved by developing AI and
data projects that are tailored to the specic needs and circumstances of
these countries. For example, AI algorithms can be developed to help
familial farmers optimise their crops and reduce water usage, or to help
A.A. Mana et al.
Smart Agricultural Technology 7 (2024) 100416
12
local communities had better manage natural resources such as forests
and rivers. In addition, shifting to renewable energy sources such as
solar, wind, and hydropower is an alternative approach to mitigate the
environmental impact of digital technology. The extensive use of fossil
fuels to power data centres and computer facilities has led to an enor-
mous increase in greenhouse gas emissions. The utilization of renewable
energy can signicantly reduce the carbon footprint of digital technol-
ogy. Additionally, the implementation of AI and data initiatives can
substantially help monitor and predicting the effects of climate change.
AI algorithms can collect and analyse data on environmental factors
such as weather patterns and soil moisture, helping us to understand the
impact of climate change on agriculture, animals, and habitats [175].
Data obtained from these initiatives can be used to help communities
plan and adapt to changing climate and inform policy decisions.
Conclusion
Considering the challenge posed by global warming, taking tangible
measures toward more inclusive and ecologically sustainable models has
become a global priority.
The evolution of agriculture towards the new agro-food 4.0 will
encourage businesses and farmers to invest in automation and articial
intelligence. Integration of green energy, slowly introduced into agri-
culture, will nd technological support based on DL and other ap-
proaches to improve production and improve agricultural security. The
current article proposes the use of computer vision technologies and
articial intelligence in the agricultural sector based on the globally
used AI technologies reported in the literature. Importantly, this review
provides a detailed understanding of promising applications in agri-
culture. The positive impacts of the use of AI in agriculture sector
include:
•The strength of AI methods in detecting, analysing, and estimating
data surpasses that of traditional techniques, especially when uti-
lizing deep learning algorithms such as CNNs, RNNs, or other
computational networks.
•Wireless technology and IoT may use the latest communication
protocols and sensors to better manage water resources and avoid
excessive irrigation losses or lack of water.
•Different integrated methods can be expended to create a sustainable
environment and increased production. Applications include
planting, fertilising, crop weeding, spraying, and harvesting.
•AI can be implemented for distant meteorological monitoring and
control of agricultural practises.
•The use of AI algorithms in agriculture can potentially exacerbate
gender inequalities, but responsible use of AI can also help mitigate
some of these inequalities and make agriculture more accessible to
women.
•Innovation can be accomplished with the use of cutting-edge tech-
nology by encoding special talents, digitising physical assets, inte-
grating data throughout agriculture, improving yields to create data,
and exchanging data.
•Renewable energy sources and AI offer a promising solution for
reducing the carbon footprint of agriculture.
CRediT authorship contribution statement
A.A. Mana: Investigation, Writing – original draft, Writing – review
& editing. A. Allouhi: Conceptualization, Investigation, Methodology,
Supervision, Validation, Writing – review & editing. A. Hamrani:
Investigation, Methodology, Writing – review & editing. S. Rahman:
Writing – review & editing. I. el Jamaoui: Writing – review & editing.
K. Jayachandran: Writing – review & editing.
Declaration of competing interest
The authors declare that they have no known competing nancial
interests or personal relationships that could have appeared to inuence
the work reported in this paper.
Data availability
No data was used for the research described in the article.
Supplementary materials
Supplementary material associated with this article can be found, in
the online version, at doi:10.1016/j.atech.2024.100416.
References
[1] M.G. Betts, et al., Global forest loss disproportionately erodes biodiversity in
intact landscapes, Nature 547 (7664) (2017) 441–444.
[2] X. Qi, Y. Fu, R.Y. Wang, C.N. Ng, H. Dang, Y. He, Improving the sustainability of
agricultural land use: an integrated framework for the conict between food
security and environmental deterioration, Appl. Geogr. 90 (2018) 214–223.
[3] H.C.J. Godfray, T. Garnett, Food security and sustainable intensication, Philos.
Trans. R. Soc. B Biol. Sci. 369 (1639) (2014) 20120273.
[4] S. Juhola, N. Klein, J. K¨
ayhk¨
o, T.S.S. Neset, Climate change transformations in
Nordic agriculture? J. Rural Stud. 51 (2017) 28–36.
[5] P.G. Pardey, J.M. Beddow, T.M. Hurley, T.K.M. Beatty, V.R. Eidman, A bounds
analysis of world food futures: global agriculture through to 2050, Aust. J. Agric.
Resour. Econ. 58 (4) (2014) 571–589.
[6] Y. Lu, N. Nakicenovic, M. Visbeck, A.S. Stevance, Policy: Five priorities for the
UN sustainable development goals, Nature 520 (7548) (2015) 432–433.
[7] C.A. Helander, K. Delin, Evaluation of farming systems according to valuation
indices developed within a European network on integrated and ecological arable
farming systems, Eur. J. Agron. 21 (1) (2004) 53–67.
[8] A.A. Mana, A. Allouhi, A. Hamrani, A. Jamil, K.O.A. Barrahmoune, D. Daffa,
Survey Review on Articial Intelligence and Embedded Systems for Agriculture
Safety: A proposed IoT Agro-meteorology System for Local Farmers in Morocco,
Smart Embed. Syst. Appl. (2023) 211.
[9] H. Lu, Y. Li, M. Chen, H. Kim, S. Serikawa, Brain intelligence: go beyond articial
intelligence, Mob. Networks Appl. 23 (2) (2018) 368–375.
[10] V. Parekh, D. Shah, M. Shah, Fatigue detection using articial intelligence
framework, Augment. Hum. Res. 5 (1) (2020) 5.
[11] I. Keshta, AI-driven IoT for smart health care: Security and privacy issues,
Informatics Med. Unlocked 30 (2022) 100903.
[12] L. Zheng, C. Wang, X. Chen, Y. Song, Z. Meng, R. Zhang, Evolutionary machine
learning builds smart education big data platform: Data-driven higher education,
Appl. Soft Comput. 136 (2023) 110114.
[13] Y. Zhou, P.D. Lund, Peer-to-peer energy sharing and trading of renewable energy
in smart communities - trading pricing models, decision-making and agent-based
collaboration, Renew. Energy (2023).
[14] S.A. Nikolidakis, D. Kandris, D.D. Vergados, C. Douligeris, Energy efcient
automated control of irrigation in agriculture by using wireless sensor networks,
Comput. Electron. Agric. 113 (2015) 154–163.
[15] L. Benos, A.C. Tagarakis, G. Dolias, R. Berruto, D. Kateris, D. Bochtis, Machine
learning in agriculture: A comprehensive updated review, Sensors 21 (11) (2021)
3758.
[16] M.G. Lampridi, C.G. Sørensen, D. Bochtis, Agricultural sustainability: A review of
concepts and methods, Sustainability 11 (18) (2019) 5120.
[17] F. Zecca, The Use of Internet of Things for the Sustainability of the Agricultural
Sector: The Case of Climate Smart Agriculture, Int. J. Civ. Eng. Technol. 10 (3)
(2019).
[18] J. W¨
aldchen, M. Rzanny, M. Seeland, P. M¨
ader, Automated plant species
identication—Trends and future directions, PLoS Comput. Biol. 14 (4) (2018)
e1005993.
[19] K. Pushpanathan, M. Hana, S. Mashohor, W.F. Fazlil Ilahi, Machine learning in
medicinal plants recognition: a review, Artif. Intell. Rev. 54 (1) (2021) 305–327.
[20] A. Abade, P.A. Ferreira, F. de Barros Vidal, Plant diseases recognition on images
using convolutional neural networks: A systematic review, Comput. Electron.
Agric. 185 (2021) 106125.
[21] Y. Yuan, L. Chen, H. Wu, L. Li, Advanced agricultural disease image recognition
technologies: A review, Inf. Process. Agric. (2021).
[22] A. Wang, W. Zhang, X. Wei, A review on weed detection using ground-based
machine vision and image processing techniques, Comput. Electron. Agric. 158
(2019) 226–240.
[23] W.H. Su, Advanced Machine Learning in Point Spectroscopy, RGB-and
hyperspectral-imaging for automatic discriminations of crops and weeds: A
review, Smart Cities 3 (3) (2020) 767–792.
[24] S.S. Virnodkar, V.K. Pachghare, V.C. Patil, S.K. Jha, Remote sensing and machine
learning for crop water stress determination in various crops: a critical review,
Precis. Agric. 21 (5) (2020) 1121–1155.
A.A. Mana et al.
Smart Agricultural Technology 7 (2024) 100416
13
[25] A.Y. Sun, B.R. Scanlon, How can Big Data and machine learning benet
environment and water management: a survey of methods, applications, and
future directions, Environ. Res. Lett. 14 (7) (2019) 73001.
[26] R. Garcia, J. Aguilar, M. Toro, A. Pinto, P. Rodriguez, A systematic literature
review on the use of machine learning in precision livestock farming, Comput.
Electron. Agric. 179 (2020) 105826.
[27] N. Li, Z. Ren, D. Li, L. Zeng, Automated techniques for monitoring the behaviour
and welfare of broilers and laying hens: towards the goal of precision livestock
farming, animal 14 (3) (2020) 617–625.
[28] J.L. Ellis, et al., Synergy between mechanistic modelling and data-driven models
for modern animal production systems in the era of big data, animal 14 (S2)
(2020) s223–s237.
[29] D. Lovarelli, J. Bacenetti, M. Guarino, A review on dairy cattle farming: Is
precision livestock farming the compromise for an environmental, economic and
social sustainable production? J. Clean. Prod. 262 (2020) 121409.
[30] T. Talaviya, D. Shah, N. Patel, H. Yagnik, M. Shah, Implementation of articial
intelligence in agriculture for optimisation of irrigation and application of
pesticides and herbicides, Artif. Intell. Agric. (2020).
[31] “AI in Agriculture Market by Technology (Machine Learning, Computer Vision,
Predictive Analytics), Offering, Application (Precision Farming, Drone Analytics,
Agriculture Robots, Livestock Monitoring), Offering, and Geography - Global
Forecast to 2025,” 2019.
[32] A.K. Kar, S.K. Choudhary, V.K. Singh, How can articial intelligence impact
sustainability: A systematic literature review, J. Clean. Prod. (2022) 134120.
[33] A.A. Mana, A. Allouhi, K. Ouazzani, A. Jamil, Feasibility of agriculture biomass
power generation in Morocco: Techno-economic analysis, J. Clean. Prod. (2021)
126293.
[34] D.M. Dimiduk, E.A. Holm, S.R. Niezgoda, Perspectives on the impact of machine
learning, deep learning, and articial intelligence on materials, processes, and
structures engineering, Integr. Mater. Manuf. Innov. 7 (3) (2018) 157–172.
[35] M. Schmidt, H. Lipson, Distilling free-form natural laws from experimental data,
Science (80-.) 324 (5923) (2009) 81–85.
[36] S.S. Mousavi, M. Schukat, E. Howley, Deep reinforcement learning: an overview,
in: Proceedings of SAI Intelligent Systems Conference, 2016, pp. 426–440.
[37] Z. Zhang, J. Boubin, C. Stewart, S. Khanal, Whole-eld reinforcement learning: A
fully autonomous aerial scouting method for precision agriculture, Sensors 20
(22) (2020) 6585.
[38] R. Gautron, O.A. Maillard, P. Preux, M. Corbeels, R. Sabbadin, Reinforcement
learning for crop management support: Review, prospects and challenges,
Comput. Electron. Agric. 200 (2022) 107182.
[39] M.T. Linaza, et al., Data-Driven Articial Intelligence Applications for Sustainable
Precision Agriculture, Agron. 2021 11 (6) (Jun. 2021) 1227, https://doi.org/
10.3390/AGRONOMY11061227. Vol. 11, Page 1227.
[40] G.P. Obi Reddy, B.S. Dwivedi, G. Ravindra Chary, Applications of geospatial and
big data technologies in smart farming. Smart Agriculture for Developing Nations:
Status, Perspectives and Challenges, Springer, 2023, pp. 15–31.
[41] L. Gong, et al., An IoT-based intelligent irrigation system with data fusion and a
self-powered wide-area network, J. Ind. Inf. Integr. 29 (2022) 100367.
[42] K. Kudashkina, M.G. Corradini, P. Thirunathan, R.Y. Yada, E.D.G. Fraser,
Articial Intelligence technology in food safety: A behavioral approach, Trends
Food Sci. Technol. 123 (2022) 376–381.
[43] Z.H. Kok, A.R. Mohamed Shariff, M.S.M. Alfatni, S. Khairunniza-Bejo, Support
Vector Machine in Precision Agriculture: A review, Comput. Electron. Agric. 191
(Dec. 2021) 106546, https://doi.org/10.1016/J.COMPAG.2021.106546.
[44] S. Bitla, S. Santhan, S. Bhagat, A. Pandey, V. Nath, Smart irrigation system: A
review, in: International Conference on Nanoelectronics, Circuits and
Communication Systems, 2018, pp. 569–578.
[45] Y. Shekhar, E. Dagur, S. Mishra, S. Sankaranarayanan, Intelligent IoT based
automated irrigation system, Int. J. Appl. Eng. Res. 12 (18) (2017) 7306–7320.
[46] J. Muangprathub, N. Boonnam, S. Kajornkasirat, N. Lekbangpong,
A. Wanichsombat, P. Nillaor, IoT and agriculture data analysis for smart farm,
Comput. Electron. Agric. 156 (2019) 467–474.
[47] K. Jha, A. Doshi, P. Patel, M. Shah, A comprehensive review on automation in
agriculture using articial intelligence, Artif. Intell. Agric. 2 (2019) 1–12.
[48] S. Fuentes, R. De Bei, J. Pech, S. Tyerman, Computational water stress indices
obtained from thermal image analysis of grapevine canopies, Irrig. Sci. 30 (6)
(2012) 523–536.
[49] K.E. Adikari, S. Shrestha, D.T. Ratnayake, A. Budhathoki, S. Mohanasundaram, M.
N. Dailey, Evaluation of articial intelligence models for ood and drought
forecasting in arid and tropical regions, Environ. Model. Softw. 144 (2021)
105136.
[50] F.A. Prodhan, et al., Deep learning for monitoring agricultural drought in South
Asia using remote sensing data, Remote Sens 13 (9) (2021) 1715.
[51] R.T. Moyo, “Articial intelligence based solar/diesel hybrid water pumping
system.” 2021.
[52] M. Fadaee, M.A.M. Radzi, Multi-objective optimization of a stand-alone hybrid
renewable energy system by using evolutionary algorithms: A review, Renew.
Sustain. energy Rev. 16 (5) (2012) 3364–3369.
[53] M.E. Karar, M.F. Al-Rasheed, A.F. Al-Rasheed, O. Reyad, IoT and neural network-
based water pumping control system for smart irrigation, arXiv Prepr.
arXiv2005.04158 (2020).
[54] F. Khan, O. Reyad, Application of intelligent multi agent based systems for E-
healthcare security, arXiv Prepr. arXiv2004.01256 (2020).
[55] M.S.B.Z. Abidin, S. Shibusawa, S. Buyamin, Z. Mohamed, Intelligent control of
capillary irrigation system for water-saving cultivation, in: 2015 10th Asian
Control Conference (ASCC), 2015, pp. 1–5.
[56] P.J. Kia, A.T. Far, M. Omid, R. Alimardani, L. Naderloo, Intelligent control based
fuzzy logic for automation of greenhouse irrigation system and evaluation in
relation to conventional systems, World Appl. Sci. J. 6 (1) (2009) 16–23.
[57] M.E. Karar, A simulation study of adaptive force controller for medical robotic
liver ultrasound guidance,” Arab, J. Sci. Eng. 43 (8) (2018) 4229–4238.
[58] P. Patil, B.L. Desai, Intelligent irrigation control system by employing wireless
sensor networks, Int. J. Comput. Appl. 79 (11) (2013).
[59] J. Xiao, D. Liu, Water-saving irrigation intelligent control system based on
STC89C52 MCU. Future Control and Automation, Springer, 2012, pp. 223–230.
[60] B.A. El-Rahiem, M.A.O. Ahmed, O. Reyad, H.A. El-Rahaman, M. Amin, F.A. El-
Samie, An efcient deep convolutional neural network for visual image
classication, in: International conference on advanced machine learning
technologies and applications, 2019, pp. 23–31.
[61] S. Choudhary, V. Gaurav, A. Singh, S. Agarwal, Autonomous Crop Irrigation
System using Articial Intelligence, Int. J. Eng. Adv. Technol. 8 (5S) (2019)
46–51.
[62] K. Anand, C. Jayakumar, M. Muthu, S. Amirneni, Automatic drip irrigation
system using fuzzy logic and mobile technology, in: 2015 IEEE Technological
Innovation in ICT for Agriculture and Rural Development (TIAR), 2015,
pp. 54–58.
[63] M.S.P. Subathra, C.J. Blessing, S.T. George, A. Thomas, A.D. Raj, V. Ewards,
Automated Intelligent Wireless Drip Irrigation Using ANN Techniques. Advances
in Big Data and Cloud Computing, Springer, 2019, pp. 555–568.
[64] H. Chen, et al., A deep learning CNN architecture applied in smart near-infrared
analysis of water pollution for agricultural irrigation resources, Agric. Water
Manag. 240 (2020) 106303.
[65] G. Arvind, V.G. Athira, H. Haripriya, R.A. Rani, S. Aravind, Automated irrigation
with advanced seed germination and pest control, in: 2017 IEEE Technological
Innovations in ICT for Agriculture and Rural Development (TIAR), 2017,
pp. 64–67.
[66] T. Poblete, S. Ortega-Farías, M.A. Moreno, M. Bardeen, Articial neural network
to predict vine water status spatial variability using multispectral information
obtained from an unmanned aerial vehicle (UAV), Sensors 17 (11) (2017) 2488.
[67] A. Mellit, M. Menghanem, M. Bendekhis, Articial neural network model for
prediction solar radiation data: application for sizing stand-alone photovoltaic
power system, IEEE Power Engineering Society General Meeting 2005 (2005)
40–44.
[68] A. Mellit, M. Benghanem, Sizing of stand-alone photovoltaic systems using neural
network adaptive model, Desalination 209 (1–3) (2007) 64–72.
[69] B.S. Richards, G.J. Conibeer, A comparison of hydrogen storage technologies for
solar-powered stand-alone power supplies: A photovoltaic system sizing
approach, Int. J. Hydrogen Energy 32 (14) (2007) 2712–2718.
[70] J.C. Hern´
andez, A. Medina, F. Jurado, Optimal allocation and sizing for
protability and voltage enhancement of PV systems on feeders, Renew. Energy
32 (10) (2007) 1768–1789.
[71] R. Ben Ammar, M. Ben Ammar, A. Oualha, Photovoltaic power forecast using
empirical models and articial intelligence approaches for water pumping
systems, Renew. Energy 153 (2020) 1016–1028.
[72] M. Achite, M. Jehanzaib, N. Elshaboury, T.W. Kim, Evaluation of machine
learning techniques for hydrological drought modeling: A case study of the Wadi
Ouahrane basin in Algeria, Water 14 (3) (2022) 431.
[73] N.S. Chandel, S.K. Chakraborty, Y.A. Rajwade, K. Dubey, M.K. Tiwari, D. Jat,
Identifying crop water stress using deep learning models, Neural Comput. Appl.
33 (2021) 5353–5367.
[74] E. Skotadis, et al., A sensing approach for automated and real-time pesticide
detection in the scope of smart-farming, Comput. Electron. Agric. 178 (2020)
105759.
[75] C.L. Chung, K.J. Huang, S.Y. Chen, M.H. Lai, Y.C. Chen, Y.F. Kuo, Detecting
Bakanae disease in rice seedlings by machine vision, Comput. Electron. Agric. 121
(2016) 404–411.
[76] A.N. Alves, W.S.R. Souza, D.L. Borges, Cotton pests classication in eld-based
images using deep residual networks, Comput. Electron. Agric. 174 (2020)
105488.
[77] L. Nanni, G. Maguolo, F. Pancino, Insect pest image detection and recognition
based on bio-inspired methods, Ecol. Inform. 57 (2020) 101089.
[78] S. Shrivastava, S.K. Singh, D.S. Hooda, Soybean plant foliar disease detection
using image retrieval approaches, Multimed. Tools Appl. 76 (24) (2017)
26647–26674.
[79] M. Zomeni, J. Tzanopoulos, J.D. Pantis, Historical analysis of landscape change
using remote sensing techniques: An explanatory tool for agricultural
transformation in Greek rural areas, Landsc. Urban Plan. 86 (1) (2008) 38–46.
[80] M.E. Karar, F. Alsunaydi, S. Albusaymi, S. Alotaibi, A new mobile application of
agricultural pests recognition using deep learning in cloud computing system,
Alexandria Eng. J. 60 (5) (2021) 4423–4432.
[81] G. Pattnaik, V.K. Shrivastava, K. Parvathi, Transfer learning-based framework for
classication of pest in tomato plants, Appl. Artif. Intell. 34 (13) (2020) 981–993.
[82] M.G. Selvaraj, et al., AI-powered banana diseases and pest detection, Plant
Methods 15 (1) (2019) 92.
[83] H. Huang, J. Deng, Y. Lan, A. Yang, X. Deng, L. Zhang, A fully convolutional
network for weed mapping of unmanned aerial vehicle (UAV) imagery, PLoS One
13 (4) (2018) e0196302.
[84] “Fighting weeds: Can we reduce, or even eliminate, herbicides by utilizing
robotics and AI? | Genetic”.
[85] M. Sujaritha, S. Annadurai, J. Satheeshkumar, S.K. Sharan, L. Mahesh, Weed
detecting robot in sugarcane elds using fuzzy real time classier, Comput.
Electron. Agric. 134 (2017) 160–171.
A.A. Mana et al.
Smart Agricultural Technology 7 (2024) 100416
14
[86] M.H. Asad, A. Bais, Weed detection in canola elds using maximum likelihood
classication and deep convolutional neural network, Inf. Process. Agric. (2019).
[87] M.D. Bah, A. Haane, R. Canals, Deep learning with unsupervised data labeling
for weed detection in line crops in UAV images, Remote Sens 10 (11) (2018)
1690.
[88] Y. Ampatzidis, V. Partel, UAV-based high throughput phenotyping in citrus
utilizing multispectral imaging and articial intelligence, Remote Sens 11 (4)
(2019) 410.
[89] Z. Beiranvand, Integration of expert system and fuzzy theory for diagnosis wheat
plant diseases, QUID Investig. Cienc. y Tecnol. (1) (2017) 1924–1930.
[90] J. Chen, J. Chen, D. Zhang, Y.A. Nanehkaran, Y. Sun, A cognitive vision method
for the detection of plant disease images, Mach. Vis. Appl. 32 (2021) 1–18.
[91] M. Perez-Ruiz, J. Martínez-Guanter, S.K. Upadhyaya, High-precision GNSS for
agricultural operations. GPS and GNSS Technology in Geosciences, Elsevier,
2021, pp. 299–335.
[92] N.N. Che’Ya, E. Dunwoody, M. Gupta, Assessment of weed classication using
hyperspectral reectance and optimal multispectral UAV imagery, Agronomy 11
(7) (2021) 1435.
[93] S.A. Zamani, Y. Baleghi, Early/late fusion structures with optimized feature
selection for weed detection using visible and thermal images of paddy elds,
Precis. Agric. 24 (2) (2023) 482–510.
[94] C. Aquilani, A. Confessore, R. Bozzi, F. Sirtori, C. Pugliese, Precision Livestock
Farming technologies in pasture-based livestock systems, Animal 16 (1) (2022)
100429.
[95] V.R. Pathmudi, N. Khatri, S. Kumar, A.S.H. Abdul-Qawy, A.K. Vyas, A systematic
review of IoT technologies and their constituents for smart and sustainable
agriculture applications, Sci. African 19 (Mar. 2023) e01577, https://doi.org/
10.1016/J.SCIAF.2023.E01577.
[96] R.S. Alonso, I. Sitt´
on-Candanedo, ´
O. García, J. Prieto, S. Rodríguez-Gonz´
alez, An
intelligent Edge-IoT platform for monitoring livestock and crops in a dairy
farming scenario, Ad Hoc Networks 98 (Mar. 2020) 102047, https://doi.org/
10.1016/J.ADHOC.2019.102047.
[97] B.C. Stahl, Articial Intelligence For a Better future: an Ecosystem Perspective On
the Ethics of AI and Emerging Digital Technologies, Springer Nature, 2021.
[98] C. Bahlo, P. Dahlhaus, H. Thompson, M. Trotter, The role of interoperable data
standards in precision livestock farming in extensive livestock systems: A review,
Comput. Electron. Agric. 156 (Jan. 2019) 459–466, https://doi.org/10.1016/J.
COMPAG.2018.12.007.
[99] P. Shine, J. Upton, P. Sefeedpari, M.D. Murphy, Energy Consumption on Dairy
Farms: A Review of Monitoring, Prediction Modelling, and Analyses, Energies
2020 13 (5) (Mar. 2020) 1288, https://doi.org/10.3390/EN13051288. Vol. 13,
Page 1288.
[100] P. Shine, T. Scully, J. Upton, M.D. Murphy, Annual electricity consumption
prediction and future expansion analysis on dairy farms using a support vector
machine, Appl. Energy 250 (2019) 1110–1119.
[101] M.P. Mcloughlin, R. Stewart, A.G. McElligott, Automated bioacoustics: methods
in ecology and conservation and their potential for animal welfare monitoring,
J. R. Soc. Interface 16 (155) (2019) 20190225.
[102] M. Benjamin, S. Yik, Precision livestock farming in swine welfare: a review for
swine practitioners, Animals 9 (4) (2019) 133.
[103] A. Hamrani, A. Akbarzadeh, C.A. Madramootoo, Machine learning for predicting
greenhouse gas emissions from agricultural soils, Sci. Total Environ. 741 (2020)
140338.
[104] A.K. Mahlein, Plant disease detection by imaging sensors–parallels and specic
demands for precision agriculture and plant phenotyping, Plant Dis 100 (2)
(2016) 241–251.
[105] D.I. Patrício, R. Rieder, Computer vision and articial intelligence in precision
agriculture for grain crops: A systematic review, Comput. Electron. Agric. 153
(2018) 69–81.
[106] “Crop yield analysis — Sustainability and articial intelligence lab.” http://susta
in.stanford.edu/crop-yield-analysis (accessed Jan. 02, 2021).
[107] I. Kumar, J. Rawat, N. Mohd, S. Husain, Opportunities of articial intelligence
and machine learning in the food industry, J. Food Qual. 2021 (2021).
[108] T. Otani, et al., Agricultural Robot under Solar Panels for Sowing, Pruning, and
Harvesting in a Synecoculture Environment, Agriculture 13 (1) (2022) 18.
[109] J. Krause, et al., SmartSpectrometer—Embedded optical spectroscopy for
applications in agriculture and industry, Sensors 21 (13) (2021) 4476.
[110] M. Zawish, S. Davy, L. Abraham, Complexity-driven cnn compression for
resource-constrained edge ai, arXiv Prepr. arXiv2208.12816 (2022).
[111] G. Kootstra, X. Wang, P.M. Blok, J. Hemming, E. Van Henten, Selective harvesting
robotics: current research, trends, and future directions, Curr. Robot. Reports 2
(2021) 95–104.
[112] V. Tinoco, M.F. Silva, F.N. Santos, L.F. Rocha, S. Magalh˜
aes, L.C. Santos, A review
of pruning and harvesting manipulators, in: 2021 IEEE International Conference
on Autonomous Robot Systems and Competitions (ICARSC), 2021, pp. 155–160.
[113] X. Bai, et al., RPNet: Rice plant counting after tillering stage based on plant
attention and multiple supervision network, Crop J (2023).
[114] X. Bai, et al., Rice Plant Counting, Locating, and Sizing Method Based on High-
Throughput UAV RGB Images, Plant Phenomics 5 (2023) 20.
[115] X. Ren, et al., Deep learning-based weather prediction: a survey, Big Data Res 23
(2021) 100178.
[116] D.K. Krishnappa, D. Irwin, E. Lyons, M. Zink, CloudCast: Cloud computing for
short-term mobile weather forecasts, in: 2012 IEEE 31st International
Performance Computing and Communications Conference (IPCCC), 2012,
pp. 61–70.
[117] R. Castro, Y.M. Souto, E. Ogasawara, F. Porto, E. Bezerra, STConvS2S:
Spatiotemporal Convolutional Sequence to Sequence Network for weather
forecasting, Neurocomputing 426 (Feb. 2021) 285–298, https://doi.org/
10.1016/j.neucom.2020.09.060.
[118] A. Nabavi-Pelesaraei, S. Raee, F. Hosseini-Fashami, K. Chau, Articial neural
networks and adaptive neuro-fuzzy inference system in energy modeling of
agricultural products. Predictive Modelling For Energy Management and Power
Systems Engineering, Elsevier, 2021, pp. 299–334.
[119] E. Elahi, Z. Khalid, Estimating smart energy inputs packages using hybrid
optimisation technique to mitigate environmental emissions of commercial sh
farms, Appl. Energy 326 (2022) 119602.
[120] M.A.Z. Abidin, M.N. Mahyuddin, M.A.A.M. Zainuri, Optimal efcient energy
production by PV module tilt-orientation prediction without compromising crop-
light demands in Agrivoltaic systems, IEEE Access (2023).
[121] M. Ehteram, A.N. Ahmed, P. Kumar, M. Sherif, A. El-Shae, Predicting freshwater
production and energy consumption in a seawater greenhouse based on ensemble
frameworks using optimized multi-layer perceptron, Energy Reports 7 (2021)
6308–6326.
[122] Y. Liu, Intelligent analysis platform of agricultural sustainable development based
on the Internet of Things and machine learning, Acta Agric. Scand. Sect. B—Soil
Plant Sci 71 (8) (2021) 718–731.
[123] B.M. Yahya, D.Z. Seker, Designing weather forecasting model using
computational intelligence tools, Appl. Artif. Intell. 33 (2) (2019) 137–151.
[124] M.P. Raju, A.J. Laxmi, IoT based online load forecasting using machine learning
algorithms, Procedia Comput. Sci. 171 (2020) 551–560.
[125] J. Kim, S. Cho, K. Ko, R.R. Rao, Short-term electric load prediction using multiple
linear regression method, in: 2018 IEEE International Conference on
Communications, Control, and Computing Technologies for Smart Grids
(SmartGridComm), 2018, pp. 1–6.
[126] X. Fu, F. Yang, Viewpoints on the Theory of Agricultural Energy Internet. Front,
Energy Res 10 (2022) 871772.
[127] X. Fu, et al., A review of key technologies and trends in the development of
integrated heating and power systems in agriculture, Entropy 23 (2) (2021) 260.
[128] I.K. Song, S.Y. Yun, S.C. Kwon, N.H. Kwak, Design of smart distribution
management system for obtaining real-time security analysis and predictive
operation in Korea, IEEE Trans. Smart Grid 4 (1) (2012) 375–382.
[129] X. Fu, F. Yang, Articial Intelligence Early Warnings of Agricultural Energy
Internet, Front. Energy Res. (2022) 594.
[130] L. Mancipe-Castro, R.E. Guti´
errez-Carvajal, Prediction of environment variables
in precision agriculture using a sparse model as data fusion strategy, Inf. Process.
Agric. 9 (2) (2022) 171–183.
[131] R. Sparrow, M. Howard, C. Degeling, Managing the risks of articial intelligence
in agriculture, NJAS Impact Agric. Life Sci. 93 (1) (2021) 172–196.
[132] M. Ryan, The social and ethical impacts of articial intelligence in agriculture:
mapping the agricultural AI literature, AI Soc. (2022) 1–13.
[133] D.R. Vincent, N. Deepa, D. Elavarasan, K. Srinivasan, S.H. Chauhdary, C. Iwendi,
Sensors driven AI-based agriculture recommendation model for assessing land
suitability, Sensors 19 (17) (2019) 3667.
[134] M.J. Smith, Getting value from articial intelligence in agriculture, Anim. Prod.
Sci. 60 (1) (2018) 46–54.
[135] X. Pham, M. Stack, How data analytics is transforming agriculture, Bus. Horiz. 61
(1) (2018) 125–133.
[136] M. Liao, Y. Yao, Applications of articial intelligence-based modeling for
bioenergy systems: A review, GCB Bioenergy 13 (5) (2021) 774–802.
[137] C. Bibbiani, F. Fantozzi, C. Gargari, C.A. Campiotti, E. Schettini, G. Vox, Wood
biomass as sustainable energy for greenhouses heating in Italy, Agric. Agric. Sci.
procedia 8 (2016) 637–645.
[138] A. Aslani, T. Mazzuca-Sobczuk, S. Eivazi, K. Bekhrad, Analysis of bioenergy
technologies development based on life cycle and adaptation trends, Renew.
energy 127 (2018) 1076–1086.
[139] J. Chen, et al., Investigation of co-combustion characteristics of sewage sludge
and coffee grounds mixtures using thermogravimetric analysis coupled to
articial neural networks modeling, Bioresour. Technol. 225 (2017) 234–245.
[140] L.M. Romeo, R. Gareta, Neural network for evaluating boiler behaviour, Appl.
Therm. Eng. 26 (14–15) (2006) 1530–1536.
[141] H. Lund, Renewable energy strategies for sustainable development, Energy 32 (6)
(2007) 912–919.
[142] T. Compernolle, N. Witters, S. Van Passel, T. Thewys, Analyzing a self-managed
CHP system for greenhouse cultivation as a protable way to reduce CO2-
emissions, Energy 36 (4) (2011) 1940–1947.
[143] J. Chen, J. Zhao, F. Xu, H. Hu, Q. Ai, J. Yang, Modeling of energy demand in the
greenhouse using PSO-GA hybrid algorithms, Math. Probl. Eng. 2015 (2015).
[144] S.S. Clair, J. Hillier, P. Smith, Estimating the pre-harvest greenhouse gas costs of
energy crop production, Biomass and bioenergy 32 (5) (2008) 442–452.
[145] A.A. Mana, “opportunities to Integration of renewable energies in Moroccan
agriculture,” Oct. 2021, Accessed: Apr. 06, 2023. [Online]. Available: https://hal.
science/tel-04025164.
[146] J. Wang, Z. Han, Z. Guan, Hybrid solar-assisted combined cooling, heating, and
power systems: A review, Renew. Sustain. Energy Rev. 133 (2020) 110256.
[147] D. Wu, R. Wang, Combined cooling, heating and power: A review, Prog. energy
Combust. Sci. 32 (5–6) (2006) 459–495.
[148] P. Zhang, Z. Guo, S. Ullah, G. Melagraki, A. Afantitis, I. Lynch, Nanotechnology
and articial intelligence to enable sustainable and precision agriculture, Nat.
Plants 7 (7) (2021) 864–876.
A.A. Mana et al.
Smart Agricultural Technology 7 (2024) 100416
15
[149] A.N. Shahzad, M.K. Qureshi, A. Wakeel, T. Misselbrook, Crop production in
Pakistan and low nitrogen use efciencies, Nat. Sustain. 2 (12) (2019)
1106–1114.
[150] G.V. Lowry, A. Avellan, L.M. Gilbertson, Opportunities and challenges for
nanotechnology in the agri-tech revolution, Nat. Nanotechnol. 14 (6) (2019)
517–522.
[151] J.P. Giraldo, H. Wu, G.M. Newkirk, S. Kruss, Nanobiotechnology approaches for
engineering smart plant sensors, Nat. Nanotechnol. 14 (6) (2019) 541–553.
[152] N. Kottegoda, et al., Urea-hydroxyapatite nanohybrids for slow release of
nitrogen, ACS Nano 11 (2) (2017) 1214–1221.
[153] S. Kabiri, F. Degryse, D.N.H. Tran, R.C. da Silva, M.J. McLaughlin, D. Losic,
Graphene oxide: A new carrier for slow release of plant micronutrients, ACS Appl.
Mater. Interfaces 9 (49) (2017) 43325–43335.
[154] B. Huang, et al., Advances in targeted pesticides with environmentally responsive
controlled release by nanotechnology, Nanomaterials 8 (2) (2018) 102.
[155] J.P. Giraldo, et al., Plant nanobionics approach to augment photosynthesis and
biochemical sensing, Nat. Mater. 13 (4) (2014) 400–408.
[156] M. Cubric, Drivers, barriers and social considerations for AI adoption in business
and management: A tertiary study, Technol. Soc. 62 (Aug. 2020) 101257, https://
doi.org/10.1016/J.TECHSOC.2020.101257.
[157] C.D. Alupo, D. Omeiza, D. Vernon, Realizing the Potential of AI in Africa: It All
Turns on Trust, Intell. Syst. Control Autom. Sci. Eng. 102 (2022) 179–192,
https://doi.org/10.1007/978-3-031-09823-9_13/COVER.
[158] M. Gardezi, R. Stock, Growing algorithmic governmentality: Interrogating the
social construction of trust in precision agriculture, J. Rural Stud. 84 (May 2021)
1–11, https://doi.org/10.1016/J.JRURSTUD.2021.03.004.
[159] S. Munnisunker, L. Nel, D. Diederichs, The Impact of Articial Intelligence on
Agricultural Labour in Europe, J. Agric. Informatics 13 (1) (Mar. 2022) 46–54,
https://doi.org/10.17700/JAI.2022.13.1.638.
[160] S. Makridakis, The forthcoming Articial Intelligence (AI) revolution: Its impact
on society and rms, Futures 90 (Jun. 2017) 46–60, https://doi.org/10.1016/J.
FUTURES.2017.03.006.
[161] M. Javaid, A. Haleem, I.H. Khan, R. Suman, Understanding the potential
applications of Articial Intelligence in Agriculture Sector, Adv. Agrochem 2 (1)
(Mar. 2023) 15–30, https://doi.org/10.1016/J.AAC.2022.10.001.
[162] D. Jim´
enez, et al., A scalable scheme to implement data-driven agriculture for
small-scale farmers, Glob. Food Sec. 23 (Dec. 2019) 256–266, https://doi.org/
10.1016/J.GFS.2019.08.004.
[163] N. Tantalaki, S. Souravlas, and M. Roumeliotis, “Data-Driven Decision Making in
Precision Agriculture: The Rise of Big Data in Agricultural Systems,” https://doi.
org/10.1080/10496505.2019.1638264, vol. 20, no. 4, pp. 344–380, Oct. 2019,
10.1080/10496505.2019.1638264.
[164] N. Ranasinghe, et al., Interpretability and accessibility of machine learning in
selected food processing, agriculture and health applications, J. Natl. Sci. Found.
Sri Lanka 50 (Nov. 2022) 263–276, https://doi.org/10.4038/jnsfsr.v50i0.11249,
no. Special Issue.
[165] M. Ryo, Explainable articial intelligence and interpretable machine learning for
agricultural data analysis, Artif. Intell. Agric. 6 (Jan. 2022) 257–265, https://doi.
org/10.1016/J.AIIA.2022.11.003.
[166] N.L. Tsakiridis, et al., Versatile Internet of Things for Agriculture: An eXplainable
AI Approach, IFIP Adv. Inf. Commun. Technol. 584 (2020) 180–191, https://doi.
org/10.1007/978-3-030-49186-4_16/FIGURES/5. IFIP.
[167] J.G.A.J. van der Vorst, J. Snels, Developments and Needs for Sustainable Agro-
Logistics in Developing Countries, World Bank, Washington, DC, 2014.
[168] R. Mazhar, A. Ghafoor, B. Xuehao, Z. Wei, Fostering sustainable agriculture: Do
institutional factors impact the adoption of multiple climate-smart agricultural
practices among new entry organic farmers in Pakistan? J. Clean. Prod. 283 (Feb.
2021) 124620 https://doi.org/10.1016/J.JCLEPRO.2020.124620.
[169] S.W. Anuga, C. Gordon, E. Boon, J.M.I. Surugu, Determinants of Climate Smart
Agriculture (CSA) Adoption among Smallholder Food Crop Farmers in the
Techiman Municipality, Ghana, Ghana J. Geogr. 11 (1) (May 2019) 124–139,
https://doi.org/10.4314/gjg.v11i1.
[170] M. Tsige, G. Synnevåg, J.B. Aune, Gendered constraints for adopting climate-
smart agriculture amongst smallholder Ethiopian women farmers, Sci. African 7
(Mar. 2020) e00250, https://doi.org/10.1016/J.SCIAF.2019.E00250.
[171] D.M. Nyariki, Household Data Collection for Socio-Economic Research in
Agriculture: Approaches and Challenges in Developing Countries, Kamla Raj
Enterp 19 (2) (May 2017) 91–99, https://doi.org/10.1080/
09718923.2009.11892696.
[172] H. Baumüller, Towards Smart Farming? Mobile technology trends and their
potential for developing country agriculture, Handb. ICT Dev. Ctries. 5 G
Perspect. English (Aug. 2017) 191–210, https://doi.org/10.1201/
9781003338376-9/Towards-smart-farming-mobile-technology-trends-potential-
developing-country-agriculture-Heike-Baum.
[173] U. Deichmann, A. Goyal, D. Mishra, Will digital technologies transform
agriculture in developing countries? Agric. Econ. 47 (S1) (Nov. 2016) 21–33,
https://doi.org/10.1111/AGEC.12300.
[174] F.N. Tubiello, et al., Estimating greenhouse gas emissions in agriculture: a manual
to address data requirements for developing countries, Estim. Greenh. gas Emiss.
Agric. a Man. to address data Requir. Dev. countries. (2015).
[175] G. Grenon, A. Hamrani, C.A. Madramootoo, B. Singh, C. von Sperber, Neural
network model predictions for phosphorus management strategies on tile-drained
organic soils, Hydrol. Res. 53 (6) (Jun. 2022) 825–839, https://doi.org/10.2166/
NH.2022.127/1060151/NH2022127.PDF.
A.A. Mana et al.
