Development of a decision support system for controlled environment vertical farming: The case of Hybrid African violet

Traditional agricultural practices face many challenges such as increasing global population, changing climate and resource scarcity. Droughts, floods and unpredictable weather threaten harvests, while valuable soil erodes and pests and diseases damage crops. At this critical juncture, a revolutionary approach to agricultural production has emerged: Controlled Environment Agriculture (CEA). It is an approach that goes beyond the traditional concept of greenhouse cultivation, bringing plants to life with precisely calibrated LED lighting that not only transitions to a closed environment for plant growth, but also mimics the sun's spectrum for optimal photosynthesis. Temperature and humidity are meticulously balanced in CEA systems to mimic ideal growing conditions. Advanced irrigation systems deliver water and nutrients directly to the roots, eliminating wasted resources and maximizing efficiency. Sensors continuously monitor all growth parameters in the environment, allowing farmers or growers to fine-tune production and growing conditions for each individual crop. This level of control turns the CEA into a powerful tool to ensure food safety, efficiency and sustainability in the 21st century. CEA offers techniques that can be applied not only in the production of food crops but also in the production of high value-added ornamentals. In this study, we propose a transformative solution for growing hybrid African violets year-round in a meticulously controlled indoor environment that maximizes the vigor and health of the plants. This study also proposes a decision support system model based on artificial intelligence, machine learning, image processing and expert opinion that can help decision makers in ornamental plant production to make fast and accurate decisions, minimize errors, losses and costs, and increase profitability. When this system is considered as a low-scale pilot study, it is expected that the system will be applicable to large-scale projects or facilities. The 'know-how' obtained from the study includes basic studies that can support decision makers in making fast and correct decisions in plant cultivation and can provide guidance to decision makers in different plant culture cultivation to be carried out both indoors and outdoors. As a result of our study, the decision support system we developed detected critical growth symptoms such as early flowering, lack of growth and root rot in the plants included in the test. In this way, necessary isolation or treatment measures were taken and losses in other plants were prevented. In addition, a variety-specific database was created, providing reference information to growers for future production studies.