OPEN DEI aims, among other objectives, at harmonizing and coordinating different Digital Transformation (DT) approaches under a common Reference Architecture Framework (RAF) so that experiences and lessons learned could be shared and assessed crossing the boundaries of specific applicative sectors.

The deliverable D2.1 Reference Architecture for cross-domain Digital Transformation aims to investigate and identify the current practices (including definition, concepts, processes, models, and templates) for the development of a Reference Architecture Framework relevant to the four sectors targeted by OPEN DEI, as well as providing recommendations (i.e. based on the identified principles) to successfully design and operate Digital Platforms supporting DT journeys.

The Reference Architecture Framework (RAF) proposed promotes reusability as a driver for interoperability, recognizing that the data-driven services for DT should reuse information and services that already exist and may be available from various sources inside or beyond the organizational boundaries of the adopting organizations. Information and services should be retrievable and be made available in interoperable formats (e.g. adhering to the FAIR principles). To this end, the core reusable Model Building Blocks (MBBs), mainly representing information sources and services, should make their data or functionality accessible using resource-oriented approaches. The reusable building block approach finds a suitable application by mapping solutions against the conceptual building blocks of a Reference Architecture that allows reusable components to be detected, which also promotes rationalization.

Starting from a Reference Architecture case analysis and relying on the requirements coming from the Innovation Actions under the CSA umbrella, the OPEN DEI project has defined the approach for designing a common Reference Architecture Framework able to describe the Cross Domain Digital Transformation.

The extensive use of sensors and connected devices is a common scenario in the implementation of many Digital Transformation solutions and in many industrial sectors. The huge amount of available data is able to cover many business scenarios. Data-driven pipelines and workflows management is nowadays crucial for data gathering, processing, and decision support. To cope with this complexity OPEN DEI has adopted the following 6C architecture, adapted from the one suggested by the German Industrie 4.0 initiative, and based on the following pillars (using a bottom-up reading):

  • Connection, making data available from/to different networks, connecting systems and digital platforms, among several IT culture and cross organizations’ boundaries, start from the capability to make data available from/to different physical and digital assets. Different devices or sensors are used to acquire a variety of IoT data, but also many systems are based on unstructured or multi-media files. Data and information may also come from existing IT systems, using sector-specific protocols or more common standards coming from the Internet of Things (IoT) world used to realize data transfers.
  • Cyber, modeling and in-memory based solutions to convert data into information, leveraging several information conversion mechanisms. Digital models (of assets, data and information) will be then shared with upper layers of the pyramid in order to improve the self-healing properties of the overall system.
  • Computing, storing and using data on the edge or on cloud. Many of modern digital platforms use a combination of cloud and edge computing models, based on driving factors for established a more centralized and powerful computation capabilities, or faster, connectivity-friendly and secure computing at the edge of the digital networked platform. The forces fueling the demand for distributing computing technologies are advancing rapidly. This will create a paradigm shift for organizations moving along new digital transformation pathways, with potential changes affecting all players in the target business ecosystem.
  • Content/Context, correlating collected data for extracting information, creating a digital space for data-information continuum, not something to push out to one side of the adopted information architecture. Modern businesses need a holistic approach with the end goal driving the data (processing) and information needs. However, exploiting data is not as straightforward. So, data needs to be acquired (captured, entered via a data pipeline) and processed with a goal and context in mind, making it information, which essentially is about processed data, before moving to the next levels.
  • Community, sharing data between people and connecting stakeholders for solving collaboration needs. Networked organizations will be able to collect and share knowledge and opportunities in the widest number of sectors so that its members can make the right decisions. The of community around organizations could become increasingly important to collect and share information in a push&pull fashion.
  • Customization, personalizing allows to add value to data following each own user perspective and to match their expectations. Multiple strategies can make it possible to address all aspects of the end user expectations and empower an individual to progress through platform functionalities in a natural way. Democratizing access to data is a promising approach to help unlock the value of data, but even the most advanced technology is of little value if people do not embrace it. This is a lesson that many businesses have learned the hard way; in order to avoid pitfalls, it is paramount to properly understand end user expectations and build the platform from the ground up while keeping in mind that the intended audience, even within a single organization, can be very diverse and must be properly segmented and with specific and varying

In this scenario, complex systems based on distributed intelligence will be increasingly designed and operated based on accurate data sharing and analysis techniques. But as one of the upper layers is showing, the “smart” functions of the platforms will gain more power by leveraging the network and community effects, such that organizations’ habits are changed while their dimensions of business are expanded.

6C Architectural Model

The above-mentioned 6C Architecture principles have driven the design of the OPEN DEI RAF, developed around the main concept of Data Spaces, identifying three main different layers described in the following using a bottom-up reading approach:

  • Field Level Data Spaces, it includes the Smart World Services able to collect data and support the interaction with the IoT Systems (configuration, calibration, data acquisition, actuation, etc.), Automation and Smart Assets (robots, machinery, and related operations) and Human Systems (manual operations, supervision, and control, etc.).
  • Edge Level Data Spaces, it defines the typical edge operations from the data acquisition (from the logical perspective) to the data processing through the data brokering. The edge services will play a key role also for data analytics (i.e. validating and improving models for data analysis).
  • Cloud Level Data Spaces, it includes data storage, data integration and data intelligence operations on the cloud. The cloud services will process big data, deploy algorithms, integrate different source platforms and services, provide advanced services such as AI prediction and reasoning.

OPEN DEI Reference Architecture Framework

Furthermore, all these horizontal Data Spaces spines will feed the OPEN DEI Reference Architecture Framework a main orthogonal dimension, named X-Industry Data Spaces, characterized by following components:

  • Trusted and Security, incorporating technical frameworks and infrastructures that complements the previous to support trusted and secure exchange, which embraces:
    • Applications Hub, an infrastructure which collects the recipes required for the provision of applications (e.g. deployment, configuration and activation) in a manner that related data access/usage control policies can be enforced.
    • Security Services, a technical framework to support Identity Access Management, Usage Control and other security services.
    • Connectors and Secure Gateways, a technical framework for trusted connection among involved parties.
  • Data Sharing, incorporating technical frameworks and infrastructures for an effective and auditable data sharing, which more specifically embraces:
    • Transaction Manager, a distributed ledger/blockchain infrastructure for logging selected data sharing transactions.
    • Data Models and Ontologies, to leverage common standard and information representations.
    • Data Sharing API, a technical framework for effective data sharing: a data sharing API.
  • Data Trading, incorporating technical frameworks and infrastructures for the trading (offering, monetization) of data, which embraces:
    • App Marketplace, enabling the offering of applications and application building blocks which can be integrated plug&play to enrich existing data spaces.
    • Data Marketplace, enabling the offerings around data resources with associated terms and conditions including data usage/access control policies as well as pricing schemas.
    • Business Support Functions, enabling data/applications usage accounting as well as implementing Clearing House, Payment and Billing functions.

Finally, all the mentioned layers serve the realization of Digital Transformation X-Industry Pilots, for enabling applications (sometimes sector specific) for supporting business scenarios from experiments.

OPEN DEI consortium has made the exercise of mapping existing sector-specific Digital Platforms to the RAF.

The aims to provide an easy mapping framework among the OPEN DEI RAF and the selected sector specific Digital Platforms, by adopting easy tools intended to highlight the important functional building blocks common to data-driven Digital Platforms across industries.

Below is a high-level mapping table to cross-reference the OPEN DEI RAF service layers to the ones included in the compared DEMETER Digital Platform. Secondly, a graphical representation of the main mappings is provided in the form of a picture comparing the approaches.

OPEN DEI RAF vs. DEMETER Digital Platform Mapping