In this master's thesis, we have addressed the topic of digital twins and their implementation using the Microsoft Azure platform. The thesis is organized into two parts, with the first part focusing on the theory, including the technologies and concepts that were applied in the second part, which is the practical section. In the practical part of the research, we carried out the process of implementing a digital twin using a case study of a motherboard manufacturing factory. This procedure was performed using existing solutions within the Azure platform, along with additional software solutions and the implementation of concepts relevant to artificial intelligence, specifically machine learning. The result of this process is a system that serves as proof of an innovative technological concept, where we identified the characteristics and potential commercial outcomes of applying a digital twin using existing solutions within the Microsoft Azure platform. Microsoft Azure platform offers numerous services and features applicable to digital twins, providing a wide range of possibilities. However, the central service, Azure Digital Twins, has limitations related to data consumption capacity, which can pose challenges when applied to a complex real-world system with numerous IoT devices. Despite these limitations, the Azure platform can be utilized for implementing digital twins, but it requires substituting certain services with other software solutions that offer better performance. Ultimately, the feasibility of implementing digital twins using cloud infrastructure depends solely on the specific underlying system on which we intend to build the digital twin.