Application of Multi-perspective Modelling Approach for Building Digital Twin in Smart Agriculture

The article offers a multi-perspective modelling approach for building a digital twin (DT) in smart agriculture. The proposed MultiDigiII methodology involves the development of conceptual models at different levels of abstraction, so that complex systems can be more fully represented in the conceptual model. In the development of complex systems, it is necessary to use more than one approach to modelling, and one, simplified conceptual model development is not enough to fully reflect the essence of the system. The MultiDigiII methodology is used in the development of the digital twin for smart agriculture. Agriculture is a very complex system to model due to various influencing changing natural conditions. Therefore, remote sensing in a dynamic environment can provide the missing spatial information needed for crop models to improve yield prediction. A more precise and appropriate model can be built based on remote sensing and IoT data. Building the digital twin by the application of a multi- perspective view is important to provide a comprehensive model, which includes not only an object from the field (for example, a potato) but also influencing obj ects. Such an approach allows getting a richer picture of processes and possible outcomes. The result of this study is the proposed MultiDigiII methodology for creating a digital twin and the results of a case study in the real-life situation of smart agriculture.