Towards Federated Digital Twin Platforms

Mirgita Frasheri, Prasad Talasila, Vanessa Scherma · 2025

Read Towards Federated Digital Twin Platforms by Mirgita Frasheri, Prasad Talasila, Vanessa Scherma (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2505-04324.txt.

Introduction

Mirgita Frasheri, Prasad Talasila, and Vanessa Scherma extend a digital twin platform toward federation in two thousand twenty-five. Digital twins mirror real systems to tune work and run cyber physical gear. Twins hold models and data that others could reuse. Reuse cuts cost and sparks joint work across teams.

Their platform, DT as a Service, now links many copies of itself. Users on distinct copies can find and share twin assets. They can also reshape and alter assets made by peers. The paper is a demo of that linked mode. It shows the blocks that make twin growth and care easier across sites.

Catalog record

CREATORMirgita Frasheri, Prasad Talasila, Vanessa Scherma
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE16 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2505-04324

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Towards Federated Digital Twin Platforms Mirgita Frasheri, Prasad Talasila, Vanessa Scherma Abstract Digital Twin (DT) technology has become rather popular in recent years, promising to optimize production processes, manage the operation of cyber-physical systems, with an impact spanning across multiple application domains (e.g., manufacturing, robotics, space etc.). DTs can include different kinds of assets, e.g., models, data, which could potentially be reused across DT projects by multiple users, directly affecting development costs, as well as enabling collaboration and further development of these assets. To provide user support for these purposes, dedicated DT frameworks and platforms are required, that take into account user needs, providing the infrastructure and building blocks for DT development and management. In this demo paper, we show how the DT as a Service (DTaaS) platform has been extended to enable a federated approach to DT development and management, that allows multiple users across multiple instances of DTaaS to discover, reuse, reconfigure, and modify existing DT assets. 1 Introduction Interest in Digital Twins (DTs) has soared in the past years, with the techonology promising to optimize production processes, as well as delivering run-time services such as monitoring, fault detection and diagnosis, as well as reconfiguration of processes or systems [ 6 ] . A DT is usually defined as a virtual replica of a physical or cyber-physical system, referred to as the physical twin (PT), with continuous bi-directional communication between the two [ 11 ] . As such, the operation of the DT is impacted by the real data coming from the PT, and in turn the operation and behaviour of the PT is influenced by the analyses and services provided by the DT. Note however, that a DT is not to substitute the functionality of the PT, but rather enhance it, and in this case we can refer to the PT as a DT-enabled system [ 6 ] . DTs are useful across application domains, from manufacturing, to robotics and space, in monitoring the degradation of materials, or possible mission failures, and executing viable mitigation strategies to avoid failures or extended down-times. In recent years, the focus of the DT research community has started to shift from the tools and methodologies for building individual DTs in rather ad-hoc ways, to platforms and frameworks that support the creation, execution, and more generally the management of possibly different DTs, as well as providing configurable services that enable DT-PT communication [ 8 ] . In addition, platforms such as the Digital Twin as a Service (DTaaS) [ 19 ] allow users to share DTs, as well as other finer grained assets, e.g., models, tools, and data, thus making it possible for users to resuse these across different DT applications and users themselves. Nevertheless, the collaboration of users has remained at a higher level of abstraction, e.g., in terms of exchanging plug-and-play models, or tools. The flexible reconfiguration and execution of DTs is not addressed to a satisfactory extent. To address this issue, DTaaS has been extended to integrate the Gitlab DevOps (combination of development and operation activities) framework with the ultimate goal of configurable and scalable execution of multiple DTs in heterogeneous computing environments. The rest of this paper is organised as follows. An overview of the existing DTaaS platform is provided in Section 2 , including additional user needs and requirements, with the DTaaS extension presented in Section 3 . A brief summary of related DT platforms is given in Section 4 . The final remarks are made in Section 5 . 2 DTaaS Platform Overview The DT as a service (DTaaS) platform facilitates the development and management of DTs [ 19 ] , by providing users with means for (i) developing and sharing reusable DT components named DT assets, (ii) a development workspace to work on these assets, (iii) integrated platform services for communication, data storage and visualisation, and (iv) a structure of DT life-cycle phases. In the context of DTaaS, assets can consist of models, data, functions, services, model solvers, each encapuslating parts of a DT that could be shared across other DT projects. Users have access to private workspaces in which they can develop these DT assets, and trial run DTs there. It is also possible to use private workspaces and create assets that are protected by Intellectual Property rights. DTaaS enables users to define a set of life-cycle phases for any DT. Figure 1 shown an example of DT life-cycle on the DTaaS platform. The create phase: include all operations needed for the discovering the available DT assets on the platform, creation of a DT instance, e.g., selection of suitable OS execution environment, installing all required dependencies; execute : run a DT instance, establish bidirectional communication with PT, run DT simulators, integrate DT with visualisation tools, and so on; reconfigure : change DT configurations before executing the same, e.g., reconfigure monitors, change models used inside the DT, and so on. This step is only required if a user intends to receive different insights from the DT; terminate : save the state of the DT and support graceful shutdown of the DT, including termination operations that may affect the PT. The user has control over which phases to include and their specification. Figure 1 : DT life-cycle phases. Dashed arrows indicate possible actions, whereas solid arrows indicate the taken action. In general, a user can work with the platform in one of two ways. The first consists in the user uploading to the platform a pre-existing and pre-packaged DT. In this case, the DT would consist of possibly tightly coupled services and models, ensuring the operation of the DT itself and the communication with the PT. Nevertheless, such coupling of models, and components in general would prevent reusability across projects – even for the same user. The second, and preferred way to work with the platform, which fully exploits its features, is to realize the DT by bringing together relevant available assets, such as data, models, functions, and tools. These would have been made available for sharing previously, by the same, or other users. 2.1 User Needs & Requirements DTaaS provides a structured way for dealing with the different lifecycle phases of DTs, as well as managing different kinds of assets associated with DTs, such as models, data, tools, etc. Nevertheless, a few issues remain to be addressed, such as (i) how to build the models underlying DTs, and thereafter how to build DTs themselves, and (ii) how to support building DTs at the low level, i.e., enabling discovery, implementation, and reconfiguration of existing models. In order to support resusability of assets across DT projects, it is crucial to provide storage and discovery mechanisms that enable users to locate those assets that are of interest to them. The DTaaS platform provides a storage mechanism that allows users to store their assets locally, or on a remote server, such as a git server. The latter is important, as it allows users to share their assets with other users across multiple instances of DTaaS. In this setting, users have access to a git server from both within and outside of the DTaaS platform. The regular git workflows can be used to support (i) synchronisation of the DT assets

Read the complete text