Currently, there is a significant lack of methodologies for the development and validation of digital twins. MTP seeks to address this gap through its participation in the PlatGDIA project.


Digital twins (DT), known as Digital Twins in English, are software artifacts that virtually represent (or replicate) a real system (Actual System - AS), continuously updating with real-time data and having the ability to interact with and influence the AS. Although DTs have been used in the industry for years for specific cases, they also find applications in fields such as medicine (biotechnology), engineering (materials), and science. However, the global vision of an Artificial Intelligence-intensive Digital Twin (DTAI), developed from various AI modules and other functionalities, has not yet reached its full potential.

The PlatGDIA project aims to 'contribute to the development of an advanced and competitive industry by exploiting the possibilities of three deep technologies: Artificial Intelligence (AI), including Machine Learning (ML), new materials, and industrial biotechnology.'

Through this project, the goal is to generate an outcome that facilitates the incorporation of DTAIs in the sectors of Industry 4.0, New Materials, and Biotechnology, and that is also adaptable to other sectors such as automotive and aerospace.

MTP participates in the PlatGDIA project with the purpose of reducing development effort by being able to employ a common framework for the development and validation of DTAIs, which will include a clear definition of the techniques, procedures, and tools to be used, thus facilitating its institutionalization, and improving its quality.

General Objectives of PlatGDIA

  • Research and development of a methodology for the development and validation of DTs, consisting of a lifecycle adapted to digital twins and a platform for creating AI-intensive digital twins (PlatGDIA or PlatAIDT, acronym in English).
  • Reduce the effort required in the development of complex DTs (embedded AI) through the research and development of a library of DT components, called DTBlocks, which can be used to build new DTs of great complexity.
  • Validate the DTMeth methodology through experimentation in the development and validation of DTs with embedded AI in specific Industry 4.0 scenarios.

Main Project Results

  • Methodology for the development and validation of DTs with embedded AI (DTMeth), including a lifecycle adapted to digital twins (DT-SDLC) for their development and deployment.
  • Software platform for creating AI-intensive digital twins (PlatGDIA) that facilitates the development and validation of these.
  • Software library of reusable components for DTs with and without embedded AI (called DTBlocks) that facilitates the construction of new DTs.
  • Several software components that facilitate the development and validation of DTAIs:
  • Digital twins intensive in Artificial Intelligence:
  1. DT of an IoT sensor network.
  2. DT of an eVTOL UAV, with application in the industrial environment.
  3. DT of an intelligent and connected industrial laboratory.
  4. DT of a mobile autonomous vehicle oriented towards tasks of high environmental complexity.
  5. DT of a complete manufacturing process in the industrial domain.

Consortium and Work Plan

The PlatGDIA consortium consists of two groups: the CDTI group formed by MTP as the coordinator, CEAMSA, ELLIOT, PLEXUS, and SORALUCE, and the AEI group formed by GRADIANT as the coordinator, UCLM, and IDEKO.

The project has a duration of 4 years and is structured into several work packages:

  • WP0: Project Management and Coordination
  • WP1: Conceptualization of Methodology and Description of Digital Twins
  • WP2 and WP3: Research in Development and Validation of Software-Intensive Digital Twins (DTMeth), respectively.
  • WP4: Development of AI-Intensive Digital Twins

In summary, the project focuses on establishing a comprehensive methodology for the development and validation of AI-intensive digital twins, with particular emphasis on the development of software-intensive digital twins.