Artificial Intelligence and Smart Farming: Leveraging Predictive Analytics and IoT Technologies for Precision Agriculture and Food Security

http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 11 (2025) Online ISSN Print ISSN . . 3007-3197 3007-3189 http://amresearchreview.com/index.php/Journal/about Page 480 Artificial Intelligence and Smart Farming: Leveraging Predictive Analytics and IoT Technologies for Precision Agriculture and Food Security Ch. Muhammad Rizwan Khurshid Senior Scientist (Agronomy), Agronomic Research Institute, AARI, Faisalabad, Pakistan Email: [email protected] Ali Raza Assistant Professor (Physics) Government Degree College, ThariMirwah, Khairpur, Sindh Pakistan Email: [email protected] Shahbaz Ali Shahani College Education Department Government of Sindh Email: [email protected] Dr. Sajid-ur-Rehman DG/Chief Scientist, Agriculture (Research Wing),Ayub Agricultural Research Institute, Faisalabad, Pakistan Email: dga[email protected] Prof. Dr. Naeem Iqbal Chairman, Department of Botany, GC University,Principal, Government College, Faisalabad Email: [email protected] Dr. Javed Ahmad Additional DG and Director/Chief Scientist, Wheat Research Institute, AARI, Faisalabad, Pakistan Email: [email protected] Dr. Naveed Akhtar Director/Chief Scientist, Agronomic Research Institute, AARI, Faisalabad, Pakistan Email: [email protected] Dr. Qurban Ali Director Research/Chief Scientist Research (Head Quarters), Ayub Agricultural Research Institute, Faisalabad, Pakistan Email: [email protected] Dr. Rizwana Qammar Scientific Officer (PBG), Oilseeds Research Institute, AARI, Faisalabad, Pakistan Email: [email protected] Maham Sajid Scientific Officer (PBG), Agronomic Research Institute, AARI, Faisalabad, Pakistan Email: [email protected] The integration of Artificial Intelligence (AI), Internet of Things (IoT), and predictive analytics has emerged as a transformative approach in modern agriculture, aiming to enhance productivity and ensure food security. This study investigated the impact of AI, IoT, and predictive analytics on agricultural performance and their contribution to sustainable food systems. A quantitative research design was employed, collecting data from 286 farmers, agronomists, and practitioners using smart farming technologies. Structured questionnaires were utilized to capture the extent of A B S T R A C T http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 11 (2025) Online ISSN Print ISSN . . 3007-3197 3007-3189 http://amresearchreview.com/index.php/Journal/about Page 481 technology adoption, IoT integration, predictive analytics utilization, agricultural productivity, and perceived food security outcomes. Descriptive statistics, correlation analysis, and regression models were conducted to examine relationships among the variables. Findings revealed that IoT integration, AI adoption, and predictive analytics utilization had significant positive effects on agricultural productivity, with IoT being the strongest predictor. Agricultural productivity, in turn, significantly influenced food security, highlighting the pathway through which technological adoption translated into broader societal benefits. The study also identified challenges related to sensor calibration, data quality, digital literacy, and resource constraints, indicating that technology adoption required complementary socio-institutional support for maximum impact. The results underscored the importance of integrating digital tools, training, and policy interventions to strengthen resilience in farming systems. These insights contribute to the understanding of smart agriculture and provide evidencebased guidance for policymakers, technology developers, and farmers aiming to enhance sustainable food production. Keywords: Agricultural productivity, Artificial Intelligence, Food security, Internet of Things, Predictive analytics, Smart farming Introduction Artiificial Intelligence (AI) was already a revolutionary force in farming, allowing determining accuracy and effectiveness in agricultural activities as never before. Smart predictive analytics combined with the Internet of Things (IoT) and AI enabled farmers to constantly check the condition of their land, weather, and crops and make data-driven decisions (Majeed et al., 2024; MDPI, 2024). The combination of technologies was an assurance of overcoming the major issues of resource inefficiency, variability of yield, and environmental degradation (Batool, Sumra, Awan&Raza, 2025). The issue of food security had become more urgent at the global level because of population increase, climate change, and decreased aridable lands (Hasan, Islam and Sadeq, 2022). The time-honored method of agriculture was not usually able to sustain the increasing demand without having a significant impact on the natural resources. Conversely, a solution based on the AI and IoT was also suggested as sustainable, which could allow reaching high productivity with a reduced environmental cost: precision agriculture (Babar and Akan, 2024; Thakur and Chhabra, 2025). Even with this promise, there was no uniform adoption of these technologies. A high initial cost, a shortage in technical expertise, connectivity problems, and doubts about AIwarzed suggestions have been among the major obstacles that many farmers, especially those in the developing world, had to overcome (Frontiers, 2024; Ayim, Kassahun and Addison, 2022; Batool et al., 2025). The constraints had slowed the process of scaling smart farming solutions to large scale other than pilot studies (Majeed et al., 2024; Babar and Akan, 2024). As such, the purpose of this study was to examine how predictive analytics and internet of things could be utilized in achieving scalability of precision agriculture especially under conditions where food security was most threatened. The research focused on creating actionable implications to the researchers, technology developers, http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 11 (2025) Online ISSN Print ISSN . . 3007-3197 3007-3189 http://amresearchreview.com/index.php/Journal/about Page 482 policymakers, and farmers by examining existing implementations, adoption issues, and strategies of integrations. Research Background The application of AI in agriculture had developed over the last time period (10 years). Although the previous systems were based on using simple algorithmic rules, the most recent outputs had concentrated on the machine learning and deep learning methods of forecasting crops, identifying diseases, and maximizing the use of inputs (Espinel, 2024). Such AI models had been able to utilize vast datasets of remotesensing, field sensors, and historic weather conditions to provide accurate information. Indicatively, ensemble tree-based models proved to be very accurate in crop suitability and productivity (Nti, 2023). Along with the maturation of AI, the technologies of IoT became more strong, less expensive, and less consuming of energy. Fields covered with networks of sensors had gathered real-time data on soil moisture, temperature, humidity, and nutrient values and transferred it to edge or cloud platforms use of low-power wide-area networks (Xu, 2022). This had provided close real-time visibility of the conditions of the field, which was necessitated in data-driven managing of the farm. Not only did smart farming systems ensure that productivity had improved, but it had also brought sustainability. The combination of IoT and AI-based predictive analytics had aided in targeted irrigation, accurate oversight of fertilizers, and immediate identification of pests or diseases to decrease wastage and environmental effects (Kumar, 2024). These systems had generated practical observations that served to avert the destruction of crops and maximizing the use of resources, which strengthens the possibility of regenerative practice. However, there had been major obstacles to adoption. Research in the developing world had proposed technological intricacy, deficiency in stakeholder cooperation, insufficient policy environments, and poor infrastructure (e.g., connectivity) as the significant obstacles (Islam, 2024). These barriers had hampered the fair proliferation of the smart farming technologies, particularly, to smallholder farmers in resourcestarved environments. Research Problem Despite the reported significant advantages of the AI and IoT-based precision agriculture in pilot-projects and the experimental conditions, the distance between the technological opportunities and practical, large-scale implementation was still significant. There were a lot of resource-poor and smallholder farmers who did not access the required infrastructure or financial resources and technical knowledge. As a result, they had not been able to make the most out of smart farming solutions that could increase their productivity and resiliency. In addition, integration of these technologies in the daily farm decision making was not comprehensive even after technologies were set up in areas where the infrastructure was installed. Data interoperability, scalability of systems and the confidence of the AI-driven recommendations had become such a challenge that farmers struggled to depend on predictive analytics with 100 percent of assurance. Unless such mechanisms have been put in place to overcome these systemic and institutional obstacles, the potential http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 11 (2025) Online ISSN Print ISSN . . 3007-3197 3007-3189 http://amresearchreview.com/index.php/Journal/about Page 483 transformative momentum of AI-enabled precision agriculture was doomed to existence as a proof-of-concept study. Research Objectives To examine the state-of-the-art AI and IoT technologies deployed in precision agriculture for predictive analytics. To identify and analyze the primary adoption barriers (technical, socio-economic, infrastructural) affecting the uptake of smart farming among diverse agricultural contexts To propose an integrated framework for applying predictive analytics and IoT systems into farm management workflows, emphasizing cost-efficiency, usability, and scalability. Research Questions Q1. What were the current AI and IoT technologies used in precision agriculture, and how had these evolved in recent years? Q2. What were the major barriers faced by farmers (especially smallholders) in adopting AIand IoT-based smart farming systems? Q3. How could predictive analytics models be integrated practically into farm management decisions in a way that is accessible and trustworthy to farmers? Significance of the Study The study was significant in the sense that it covered an urgent disconnection between the technological potential of smart farming and its practical adoption. Through an evaluation of technical and socio-economic barriers, the research allowed gaining a full picture of the reasons of why farmers had not invested in AI and IoT innovations to the extent they could have. The integration framework proposed was to help the stakeholders such as the researchers, technology developers, policymakers and farmers to design scalable easy to use systems that could not last in the pilot stages. Moreover, the research was very timely when it comes to the issue world food security. As populations grew, climate change, and resource degradation were putting the stability of agriculture under threat, the concept of data-driven precision agriculture was an attractive approach to sustainable production. The study added to the current discussion of how digital transformation can be utilized to create stability in the long-term food system considering that predictive analytics and IoT may be successfully utilized. Literature Review Advances in AI and Predictive Analytics in Modern Agriculture According to recent studies, there was a vast improvement in AI-based yield prediction, in which machine-learning algorithms like the Random Forest, XGBoost, and deep-learning hybrid systems were all successful in surpassing traditional statistical methods. Such models worked previously since they combined multisource inputs, such as soil parameters, satellite images, and weather indices to come up with a more accurate yield prediction (Li et al., 2023; Ramcharan et al., 2022). This development showed that predictive analytics with AI capabilities helped farmers http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 11 (2025) Online ISSN Print ISSN . . 3007-3197 3007-3189 http://amresearchreview.com/index.php/Journal/about Page 484 have advance warning of how to plan and prevent risks. Research also found out that deep learning enhanced accuracy of plant disease detection particularly with convolutional neural networks that were trained on massive image datasets. The EfficientNet and attention-based CNNs had proven to be highly resistant to noise throughout the environment, and thus, they were more appropriate in the actual field (Sethy and Behera, 2020; Islam et al., 2024). Through these systems, disease outbreaks could be detected early to minimize losses of crops as well as to make more timely interventions. It was also noted by researchers that advances have been made in integrating remote sensing with AI models to monitor crops. The application of hyperspectral, multispectral, and thermal imagery and machine-learning pipelines had made it more accurate to identify stress, pest infestation, and growth anomalies in nutrients (Zhang et al., 2021; Khanna et al., 2022). This unification enhanced prediction accuracy because the models could predict the micro-level and field-level variability. Smart Farming IoT Sensing Systems, IoT Sensing Systems and Edge Computing IoT architecture had developed at a high rate such that farmers can install networks of low-power sensors and can monitor soil moisture, pH, humidity, and leaf wetness in real time. It was found that in regions with poor coverage, affordable sensors with the use of LPWANs such as LoRaWAN and NB-IoT were able to enhance the reliability of data transmission (Sharma et al., 2022; Mekala and Viswanathan, 2021). This infrastructure supported the principle underpinning precision agriculture through constant and quality data gathering. An additional point that was highlighted by researchers is that edge computing was the key to a shorter and less intensive consumption of latency and bandwidth. With some of the initial analytics and data filtering to edge nodes, even farms with lowspeed internet connectivity could have used real-time decision-support systems (Tzounis et al., 2021; Singh et al., 2023). This made it possible to have timely control over irrigation, detect anomalies and adjustment without relying on cloud servers largely. In addition, interoperability and data standardization were also declared as crucial issues in the adoption research of IoT. A number of research papers reported that irregular sensor calibration, irregular communication protocols, and scattered data formats typically lead to data-quality problems adversely influencing AI model to perform (Aqeel-ur-Rehman et al., 2020; Kayadibi et al., 2023). In response to this, scholars suggested a single data framework and standard sensor requirements on agricultural IoT systems. Adoption Barriers, Socioeconomic Constraints and Scaling Challenges Some of the studies revealed that the use of smart-farming technologies by farmers was low because of financial limitations, the absence of digital skills and distrust of the AI-based recommendations. High-tech solutions had been pilfered by smallholder farmers especially, unless they could be shown to provide instant economic payoff (Bronson, 2022; Rehman et al., 2021). These socioeconomic obstacles emphasized the fact that technology preparation was not a guarantee of mass adoption. Research focused on policy underlined the need to have government-directed http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 11 (2025) Online ISSN Print ISSN . . 3007-3197 3007-3189 http://amresearchreview.com/index.php/Journal/about Page 485 infrastructure creation and funding so as to scale AI-based agricultural systems. Investigation in developing countries registered that financial aids on IoT equipment, investment in computerized infrastructure, and enhanced agricultural expansion administrations positively impacted the readiness of farmers to use clever technologies (Liu et al., 2024; Adeyemi et al., 2023). The lack of institutional support did not translate into large-scale applications of smart farming solutions since they only were experienced in small demonstrations. Lastly, researchers observed that the participatory and context-based design of technologies was necessary to make technologies sustainable adoption. Co-design methods in which farmers were involved in working with the technology developers were already proven to enhance trust and long-term opportunity work with AI systems (Rose et al., 2021; van der Burg et al., 2022). These types of collaborations made sure that technological solutions were in tandem with real-world needs, cultural activities and resource limitations of farming communities. Research Methodology Research Design The research design used in this study was a quantitative and explanatory one to understand the role of Artificial Intelligence (AI), Internet of Things (IoT) technologies and predictive analytics in the development of precision agriculture and food security. This design was chosen due to the fact that it enabled the establishment of relationships between technological adoption, agricultural efficiencies, and sustainability indicators to be systematically investigated. The research also employed cross-sectional methodology since the survey was conducted at one instance on the farmers, experts in agricultural technology, and users of IoT systems. The design allowed the researchers to measure the trends, generalized the association hypothesis and measured the contribution of AI-based tools to yield optimization, disease detection, and environmental monitoring within the agricultural environment. Population and Sampling The participants were registered farmers, agronomists, and smart-farming practitioners who implemented technologies based on AI or IoTs in irrigation, soil monitoring, pest detection, or crop forecasting. The purposive sampling focused was adopted due to the fact that the study involved the sample who had the first-hand experience of using the digital farming tools. All 320 individuals were contacted and 286 out of them filled in the survey questionnaires, thus making the response rate 89%. The sample size was found satisfactory to undertake regression, correlation, and structural modelling. The demographics that were captured were the size of the farm, the type of crop, years of farming experience, and the degree of exposure to AI/IoTrelated farming solutions. Data Collection Procedure The structured questionnaire was utilized to gather data and the questionnaire had been modified based on validated tools applied in previous research on the adoption of agricultural technologies and the use of electronic agriculture. The questionnaire was divided into five sections, which included demographics, the uptake of IoT http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 11 (2025) Online ISSN Print ISSN . . 3007-3197 3007-3189 http://amresearchreview.com/index.php/Journal/about Page 486 technologies, engagement in AI-driven applications, predictive analytics, and apparent influences on the productivity and food security. The measurement of items was done through a five-point Likert scale of 1 (strongly disagree) to 5 (strongly agree). The instrument was reviewed by three agricultural technology experts before administration in the field in order to clarify the instrument and make it topical. It collected data using face to face distribution places in agricultural extension center and it held data online through farming community sites. Research was conducted ethically and was on a voluntary basis. Research Instrument The tool quantified four major constructs namely AI adoption, IoT integration, use of predictive analytics and agricultural performance outcomes. Items that were used in the adoption of AI were those concerning crop disease classification, automated decision-support systems, and yield prediction algorithms. The measure of IoT integration was the use of sensors to determine the soil moisture and temperature, humidity and real-time field conditions. Predictive analytics involved use of forecasting models to determine the output of crops, water requirement and trends of pest infestation. Agricultural performance had measures like growth in yield, reduction in resource cost, reduction in input costs and general contribution to food security. The alpha test of reliability was carried out and the constructs all achieved above the desirable level of 0.70, which means high internal consistency. Data Analysis Techniques The analysis of data was done on SPSS and SmartPLS software. On the one hand, descriptive statistics were calculated to summarize demographic data and to determine central tendencies of the most important variables. The Pearson correlation analysis was then conducted to determine the strength and direction of relationships between adoption of AI, adoption of IoT, predictive analytics, and agricultural outcomes. The predictive impact of AI, IoT, and analytics on the performance of agriculture was analyzed using multiple regression analysis. Another type of analysis done was structural equation modelling (SEM), which was done to test the theoretical model to see the direct and indirect relationships between constructs. To check the soundness of the results, normality tests, multiple collinearity tests and reliability tests were carried out before SEM. Ethical Considerations The institutional review board provided the ethical approval of the study before the collection of data. The participants were made aware of why the research had to be done and how their answers would be hold confidential and be utilized only with academic interests. All the respondents gave their consent. Personal identifiers have not been gathered and the information has been safely stored in password-protected files. It was declared to the participants that drop-outs had no consequences. These ethical processes were used to make the right procedures in line with the set research standards. http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 11 (2025) Online ISSN Print ISSN . . 3007-3197 3007-3189 http://amresearchreview.com/index.php/Journal/about Page 487 Results and Analysis Overview of Findings This section gave the results of the empirical work based on descriptive statistics, correlation analysis, regression modelling, and structural modelling. The results provided the roles of artificial intelligence (AI) as well as Internet of Things (IoT) technologies and predictive analytics to enhance farming performance, resource effectiveness, and food safety. There were several tables to summarize the quantitative data which was further analyzed in a narrative format explaining the findings. Table 1. Descriptive Statistics of Key Study Variables (N = 286) Variable Mean SD Minimum Maximum AI Adoption 3.87 0.72 1.90 4.98 IoT Integration 3.94 0.69 2.10 4.89 Predictive Analytics Utilization 3.76 0.75 1.80 4.95 Agricultural Productivity 4.01 0.66 2.30 4.97 Food Security Contribution 3.92 0.71 2.00 4.90 Descriptive statistics revealed that the levels of AI adoption were moderate among the respondents as shown by a mean of 3.87. This indicated that farmers had progressively incorporated AI-driven decision support systems like automated disease-detecting systems, automated yield management programs, and automated irrigation programs in their farming activities. On the same note, the integration of IoT was revealed to be the variable with the highest mean (M = 3.94), which proved that technology involving sensors, water monitors, and automated weather stations were already commonly employed in the precision agriculture models. The standard deviations were also relatively low, meaning that the responses were uniform to participants. The findings also revealed that Predictive Analytics Utilization registered a mean of 3.76 meaning that most farmers had already embarked on the use of forecasting model to estimate crop yields, determine climatic risks, and predict epidemics of pests. Predictive analytics although in their emergent state proved to be more and more relevant in agriculture decision-making. The Agricultural Productivity has the mean of 4.01; it showed that the majority of respondents noted that crop yield, operational efficiency, and overall resource use had a significant change towards a better outcome due to digital farming tools. In terms of Food Security Contribution, the average of 3.92 indicated that the respondents felt that hi-tech technologies were a factor which created better food http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 11 (2025) Online ISSN Print ISSN . . 3007-3197 3007-3189 http://amresearchreview.com/index.php/Journal/about Page 488 availability, stability, and sustainability at either the communal, or the national level. All variables high means were combined which obstructed that already digital technologies in agriculture were starting to show significant positive results. The latter descriptive findings formed a basis of conducting other analyses like correlation and regression analysis to establish the relationship between variables. Figure 1 Descriptive Statistics of Key Study Variables (N = 286) Table 2 Correlation Matrix of Study Variables Variables AI Adoption IoT Integration Predictive Analytics Agricultural Productivity Food Security AI Adoption 1 0.62 0.58 0.55 0.49 IoT Integration 0.62 1 0.64 0.59 0.54 Predictive Analytics Utilization 0.58 0.64 1 0.61 0.57 Agricultural Productivity 0.55 0.59 0.61 1 0.66 Food Security Contribution 0.49 0.54 0.57 0.66 1 http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 11 (2025) Online ISSN Print ISSN . . 3007-3197 3007-3189 http://amresearchreview.com/index.php/Journal/about Page 495 under real-field conditions. Plant Methods, 20, 1–15. Kayadibi, M., et al. (2023). Standardization challenges in agricultural IoT systems. Sensors, 23(5), 2765. Khanna, M., et al. (2022). Remote sensing and machine learning for precision agriculture. Agricultural Systems, 198, 103389. Khalid, M., Sarfraz, M. S., Iqbal, U., Aftab, M. U., Niedbała, G., &Rauf, H. T. (2023). Real-time plant health detection using deep convolutional neural networks. Agriculture, 13(2), 510. https://doi.org/10.3390/agriculture13020510 Kumar, V., & Singh, M. (2024). A comprehensive review on smart and sustainable agriculture using Internet of Things (IoT) and AI. Computers and Electronics in Agriculture. https://doi.org/10.1016/j.compag.2024.108993 Liu, F., Su, L., Luo, P., Tao, W., Wang, Q., & Deng, M. (2024). Prediction models of growth characteristics and yield for Chinese winter wheat based on machine learning. Agronomy, 14(4), 839. https://doi.org/10.3390/agronomy14040839 Li, Z., et al. (2023). Machine learning models for yield prediction across diverse environments. Field Crops Research, 294, 108882. Liu, Y., et al. (2024). Policy support and digital transformation of agriculture. Land Use Policy, 135, 106241. Majeed, H., Azeem, W., Malik, A. A., Ahmad, U., & Sultan, H. (2024). Cultivating Innovation: An In-Depth Exploration of IoT and AI in Smart Agriculture. Pakistan Journal of Scientific Research. Mekala, M. S., &Viswanathan, P. (2021). A review on IoT-based smart farming. Sustainable Computing, 30, 100512. Nawaz, M., & Babar, M. I. K. (2025). IoT and AI for smart agriculture in resourceconstrained environments: challenges, opportunities and solutions. Discover Internet of Things, 5, Article 24. https://doi.org/10.1007/s43926-025-00119-3 Nti, I. K., Abubakar, A., &Nartey, E. (2023). A predictive analytics model for crop suitability and productivity: Toward Agenda 2030. Sustainable Computing: Informatics and Systems, 44, 100 893. https://doi.org/10.1016/j.suscom.2023.100893 Pacal, I., Kunduracıoğlu, İ., Alma, M. H., et al. (2024). A systematic review of deep learning techniques for plant diseases. Artificial Intelligence Review, 57, 304. https://doi.org/10.1007/s10462-024-10944-7 Precision Agriculture (2024). Integrative approaches in modern agriculture: IoT, ML and AI for disease forecasting amidst climate change. Precision Agriculture, 25, 2589–2613. https://doi.org/10.1007/s11119-024-10164-7 Rehman, A., Saba, T., Kashif, M., Fati, S. M., Bahaj, S. A., & Chaudhry, H. (2022). A revisit of Internet of Things technologies for monitoring and control strategies in smart agriculture. Agronomy, 12(1), 127. https://doi.org/10.3390/agronomy12010127 Ramcharan, A., et al. (2022). Multi-source data fusion for crop yield prediction. Computers and Electronics in Agriculture, 197, 106971. http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 11 (2025) Online ISSN Print ISSN . . 3007-3197 3007-3189 http://amresearchreview.com/index.php/Journal/about Page 496 Rehman, A., et al. (2021). Barriers to precision agriculture adoption among farmers. Technology in Society, 66, 101728. Rose, D. C., et al. (2021). Co-design principles for digital agriculture. Agricultural Systems, 190, 103109. Shafik, W., Tufail, A., De Silva, L. C., et al. (2025). Deep learning technique for plant disease classification and pest detection and model explainability elevating agricultural sustainability. BMC Plant Biology, 25, Article 1491. https://doi.org/10.1186/s12870-025-07377-x Sethy, P. K., &Behera, S. (2020). Deep learning approaches for plant disease detection. Journal of Plant Pathology, 102, 1–11. Sharma, S., et al. (2022). IoT sensing technologies for agricultural monitoring. Sensors, 22(11), 4153. Singh, A., et al. (2023). Edge computing for real-time agricultural analytics. IEEE Internet of Things Journal, 10(12), 10455–10467. Tzounis, A., et al. (2021). The role of IoT and edge computing in smart agriculture. Biosystems Engineering, 201, 1–17. Thakur, P., &Chhabra, V. (2025). A comprehensive review on smart farming. International Journal of Research in Agronomy. van der Burg, S., et al. (2022). Responsible innovation and AI adoption in agriculture. Agriculture and Human Values, 39, 739–755. Xu, J., Xu, Q., & Ying, X. (2022). Review of agricultural IoT technology. Artificial Intelligence in Industry Applications, 1, 100001. Zhang, Y., et al. (2021). Hyperspectral imaging and machine learning for crop monitoring. Sensors, 21(3), 912.