Abstract
Agriculture, as a foundational sector of primary production and a critical factor in global food security, is called upon to increase its productivity by approximately 60% by the end of the 21st century, according to estimates by the United Nations Food and Agriculture Organization (FAO). This demand, compounded by the rapidly growing global population and increasing pressure on natural resources, necessitates not only a quantitative but also a qualitative restructuring of agricultural practices, integrating sustainable intensification approaches that align with the principles of ecological management and biodiversity conservation. The concept of Agriculture 4.0, which has emerged in recent years as an adaptation of Industry 4.0 principles to the agricultural sector, focuses on advanced digitization and automation technologies, including the Internet of Things (IoT), cloud computing, artificial intelligence (AI), and big data analytics. These collectively aim to create a "smart" agricu ...
Agriculture, as a foundational sector of primary production and a critical factor in global food security, is called upon to increase its productivity by approximately 60% by the end of the 21st century, according to estimates by the United Nations Food and Agriculture Organization (FAO). This demand, compounded by the rapidly growing global population and increasing pressure on natural resources, necessitates not only a quantitative but also a qualitative restructuring of agricultural practices, integrating sustainable intensification approaches that align with the principles of ecological management and biodiversity conservation. The concept of Agriculture 4.0, which has emerged in recent years as an adaptation of Industry 4.0 principles to the agricultural sector, focuses on advanced digitization and automation technologies, including the Internet of Things (IoT), cloud computing, artificial intelligence (AI), and big data analytics. These collectively aim to create a "smart" agricultural environment. The integration of these technologies within Precision Agriculture practices represents a significant advancement, allowing real-time data acquisition, condition monitoring, and decision-making that enhances efficiency, reduces waste, and optimizes resource use in agricultural production systems. In this data-rich and ever-evolving environment, sensor networks, as a core element of IoT, play an essential role in automating agricultural processes by enabling data collection from the physical environment, capturing a wide range of parameters (e.g., temperature, soil moisture, animal movement), as well as other critical information, depending on the application. However, the large and heterogeneous volumes of data generated by these networks pose significant challenges in data management, interoperability, and scalability, especially when real-time processing, analysis, and utilization are required. This doctoral dissertation proposes an approach that combines middleware and cloud computing technologies to develop an integrated system based on the Internet of Things, addressing the automation needs of modern agricultural production systems. The multi-layered, hierarchical structure of the proposed system provides a robust infrastructure for aggregating and analysing data from distributed sources in a central cloud environment, supporting large-scale operations adaptable to diverse agricultural settings. Additionally, the proposed middleware aims to achieve flexible connectivity, scalability, and high interoperability, facilitating the seamless integration and coordinated operation of various devices and platforms. A central feature of the middleware is its integration of context awareness, which enables continuous system analysis and real-time adaptation, improving the reliability and efficiency of data management in agricultural production systems and enabling the development of automated applications. In designing the middleware, special emphasis was placed on the framework for context modelling, which provides an expressive representation of the relationships and attributes of data, achieving semantic relevance and supporting inference, while also providing a flexible and adaptable foundation for automation in agricultural production systems. To validate the functional applicability and reliability of the proposed system, two case studies were conducted: one in a crop cultivation environment and the other in a livestock unit. The results of these evaluations highlighted the middleware's capacity to optimize data processing efficiency, offering actionable insights that support strategic decision-making in precision agriculture. The case studies also demonstrate the system's potential to support sustainable intensification goals, as well as proactive actions in response to climate change-induced phenomena. Overall, this dissertation is expected to make a substantial contribution to contemporary scientific knowledge, specifically in the field of Agriculture 4.0, by proposing an integrated, intelligent, and sustainable data management architecture. The proposed framework addresses key interoperability challenges across systems, enabling seamless collaboration among various technological platforms and applications. Simultaneously, it promotes the automation of agricultural production processes, thereby enhancing efficiency and sustainability in the agricultural sector. In this way, the dissertation is anticipated to support the implementation of practices that favour a more interconnected and efficient agriculture, harnessing cutting-edge technologies to address the challenges of modern agricultural production.
show more
