teknowlogy´s top IT trends for 2019 – 2/10 Virtual data platforms
The success of digital transformation is closely linked to the availability of information. The focus is on collecting data from different sources, such as operational systems or the internet, and analyzing it to develop new services and business models. Status and consumption data from connected devices and machines is also increasingly joining this flow of information.
This focus on new, external data sources sometimes obscures the fact that large amounts of data are already available in companies. However, these pieces of information are scattered across various departments and databases, as they are intended for specific applications such as sales, production, product development, supply chain, or after-sales. But if processes are to be aligned to customer requirements more efficiently and with greater agility, then data from different silos must be integrated. Only then are comprehensive analyses and correlations of different data possible.
In our opinion, a possible approach to solving this problem is the use of a virtual, shared data platform consisting of (in-memory) databases, data lakes, and data repositories, which in turn merges various items of data in different formats (such as real-time data, historical data, and information from external, publicly available data sources) and establishes relationships between them.
This data platform acts as a central, virtual data hub for all business processes and is continuously supplied with data from applications. In the case of a digital factory, these can be PLM, ERP, MES, or CRM applications, for example. In addition, the platform is also fed with data from the ecosystem, for example from engineering partners, suppliers, contract manufacturers, marketing agencies, resellers, service partners, or customers. This principle can of course also be applied to other industries.
The platform enables flexible use of this data for business processes, data analysis, cognitive solutions, augmented reality, and virtual reality as well as robotics.
Virtual data platforms are thus suitable both for optimizing existing processes and supporting new business processes and models, for example in the area of IoT. The data is accessible via interfaces, with APIs and microservices playing an increasingly important role here. What is essential is that the data is used within the framework of legal requirements and taking aspects related to privacy and data security into consideration.