Application of Multi-Model Databases in Digital Twins Using the Example of a Quality Assurance Process

As digitalization in factories continues, companies increasingly want to establish their production as a cohesive digital representation. One method to achieve this over the lifecycle of an asset is the digital twin (DT). It contains model descriptions of products and processes and incorporates asset-specific information. Due to the variety of data generated, efficient handling poses a challenge. Previously used databases often follow a relational approach, which is not suitable for storing extensive, heterogeneous, unstructured data. NoSQL, which can basically handle this data, however, only supports one data model (e.g., graphs), resulting in multiple databases for each data model. We propose the use of multi-model databases (MMD) for digital twins which can store the whole range of data models needed within one single system. Since MMDs found only little application in this context, advantages over the aforementioned approaches for their use in digital twins are shown. The MMD allows to connect all generated data consistently which thereby enables an efficient cross-data model query. The development and application of a digital twin in an MMD is demonstrated by an industrial quality inspection process. In accordance with the DT, all necessary data, such as inspection equipment, plans and processes, are represented in a suitable data model inside the MMD. Finally, the results of the application of MMD in DTs are discussed and aspects for future work are pointed out.