Digital twins (DT) represent a transformative paradigm with broad applicability across various domains, including manufacturing, healthcare, and infrastructure. While their potential spans many sectors, this study focuses on the application of DT technology within the context of smart cities and urban planning. Despite growing interest, adoption remains slow due to implementation complexity, such as data integration and system scalability challenges. This study introduces DigiTDevOps, an open, modular platform designed for DT modeling, deployment, and operation in a hybrid edge-cloud environment. By analyzing the DT ecosystem and defining key functional and non-functional requirements in alignment with ISO/IEC 25010 software quality standards, the platform ensures scalability, security, and interoperability. Initial findings emphasize the importance of open-access datasets, the reuse of configurable DT fragments as building blocks for digital twins, and the advantages of a standardized DT platform in improving efficiency and reducing development complexity. By addressing integration barriers and fostering scalable, high-performance DT solutions, this work contributes to the broader adoption of digital twin technology in urban applications.
Digital Twin (DT) technology, first conceptualized by Michael Grieves [
Through stakeholder engagement, the project has identified 131 functional and 23 nonfunctional requirements, ensuring that the platform addresses practical needs while aligning with ISO/IEC 25010 software quality standards. This standard guarantee core quality attributes such as functionality, security, maintainability, and performance. Central to the approach is a structured DT data ecosystem that facilitates scalability and reuse and is modeled by Capability
(platform-wide), meso (domain-specific), and micro (use case–specific) – following principles
outlined in digital ecosystem research [
The remainder of this paper outlines the project overview, describes how the structured DT data ecosystem was developed, and details the functional and technical requirements that enhance scalability, adaptability, and efficiency. The findings highlight the potential to overcome integration barriers and accelerate the adoption of DT technologies in smart cities by ensuring a flexible and standardized development framework.
The goal of this project is to develop DigiTDevOps, an open, modular platform for DT modeling, deployment, and operation, grounded in the latest advancements in scientific research. Designed for a hybrid (edge-cloud) computational environment, the platform facilitates the full life cycle management of urban-oriented digital twins. By enabling efficient knowledge aggregation, data interoperability, and modular integration, it aims to overcome key technological barriers associated with DT development and operationalization.
A key innovation is the modular approach, achieved through the introduction of reusable, configurable DT fragments. The platform includes a repository of unique fragments, allowing for the efficient composition of new DTs, reducing development time, and enhancing scalability. Unlike existing solutions, which often face rigid architectures and fragmented data ecosystems, this system is designed for flexibility and automation. Its structured multi-scaling approach ensures seamless interoperability across diverse urban applications, while CDD enables adaptive and goal-oriented DT solutions.
Additionally, the platform enhances automation and decision-making capabilities by facilitating real-time data integration and supporting open-access datasets for more informed urban planning. Future enhancements may explore AI-driven optimizations, further improving the ability to automate processes, predict trends, and enhance operational efficiency.
By addressing these challenges, the project increases accessibility and accelerates the adoption of digital twin technologies in smart city applications. It is undertaken within the framework of a European initiative of common interest (IPCEI-CIS).
To foster collaboration and accelerate innovation within the research community, the DigiTDevOps platform will be released as open-source software. This approach ensures that the broader community can freely access, modify, and build upon the platform. In particular, the platform will feature a repository of reusable, configurable DT fragments – modular components that serve as building blocks for constructing digital twins across various domains. By making these resources openly available, the project not only supports transparency and knowledge sharing, but also provides concrete, actionable tools for RCIS researchers and practitioners to advance their own digital twin initiatives.
The current phase of DigiTDevOps has focused on defining the DT data ecosystem and identifying key requirements for the platform. The ecosystem has been structured using a multi-scaling approach to capture relevant stakeholders, context elements, data sources, and adjustments based on simulations. Additionally, the core functional and non-functional requirements have been outlined to ensure scalability and efficiency.
The following sections provide a concise overview of these results, highlighting the key aspects of the DT data ecosystem and the essential requirements for implementation.
The DigiTDevOps project has focused on identifying application cases, priority verticals, and key
stakeholders essential for scaling DT solutions. A key aspect of this effort involved defining a data
ecosystem capable of supporting DT deployment in smart cities and beyond, ensuring both
contextual relevance and future scalability. A digital twin data ecosystem is a networked system
where multiple digital twins are interconnected, facilitating seamless data exchange and
collaboration across various platforms and stakeholders. This interconnectedness enables real-time
data sharing, comprehensive analysis, and coordinated decision-making, enhancing the efficiency
and adaptability of complex systems [
The DT data ecosystem was modeled using a structured set of interrelated concepts that describe the goals, contextual factors, and stakeholder-driven actions shaping Digital Twin operations. At the center of the model is the notion of Capabilities that adapt to dynamic environments. These are influenced by changing contextual situations and are pursued by specific stakeholders through simulation-informed actions known as Adjustments. Key Performance Indicators are used to evaluate the achievement of goals, while measurable data properties ensure traceability and grounding in real-world conditions. Rather than representing the DT as a standalone entity, the model positions it as the enabling mechanism behind these interrelated elements. The conceptual overview in Table 1 provides the basis for structured modeling of the DT data ecosystem, supporting a consistent and adaptable approach to platform development.
Figure 1 illustrates a meso-level data ecosystem model example developed for the transport and mobility vertical, serving as a representative example of how the conceptual approach is applied in practice. While this case focuses on a single vertical, the overall project targets three key smart city domains: transport and mobility, energy, and civil defense. The model highlights how data sources are linked to contextual elements and how simulation-driven insights lead to targeted adjustments, supporting data-driven decision-making. The modeling is based on the concepts described in Table 1, ensuring consistency and traceability across different verticals and levels of abstraction. This example demonstrates how data ecosystems are structured to reflect real-world complexity and enable adaptive, goal-oriented actions within the platform.
real service, process or equipment based on the simulations performed with the DT. Each adjustment has an owner or Stakeholder who executes the adjustment.
Measurable attributes and data used to describe the context situation.
representation
Further analysis of the data infrastructure underscores the importance of ensuring reliable data availability, efficient retrieval, storage, and processing throughout the Digital Twin lifecycle. To support the modular and reusable design of DT components, open-access datasets have been identified as valuable resources for constructing interoperable DT fragments. This approach addresses a critical challenge in DT development—reducing dependency on case-specific data pipelines and enabling faster adaptation across domains. Ensuring access to high-quality and sustainable data sources remains a key priority for upcoming development phases. The requirements and insights gathered during the current phase form the basis for evolving the DigiTDevOps platform into a scalable, efficient, and generalizable solution, aligned with software quality standards and open innovation principles.
The requirements for the DigiTDevOps platform have been defined to cover all core components necessary for delivering an integrated, modular, and scalable environment for Digital Twin development and operation. Rather than focusing on visual representation alone, the project prioritizes data structures, analytics, and automation to support the full lifecycle of DTs. The platform design places strong emphasis on orchestrating systems, services, and processes to ensure both technical depth and operational flexibility.
The identified requirements are grounded in insights gathered from the end-user workshop,
where a diverse set of DT use cases across three priority smart city verticals were analyzed and
modeled. Additionally, rapid literature review [
Compared to existing DT platforms such as Microsoft Azure Digital Twins or Eclipse Ditto,
DigiTDevOps distinguishes itself through its capability-driven foundation, focus on DT fragment
reuse, dynamic resolution adaptation, and integrated simulation and scenario management features
[
To date, a total of 131 functional requirements have been identified, covering essential platform components such as User Entity, Core Entity, Data Collection & Device Control, Fragment Repository, Data Management, and Security Management, along with analytics and orchestration engines. In addition, 23 non-functional requirements have been grouped into six critical quality areas: Security, Interoperability, Dependability, Predictability, Reliability, and Sustainability. This structured approach ensures that the platform is not only feature-rich but also robust and production-ready.
To guide the development roadmap, five overarching epics have been defined, articulating the principal objectives the platform must achieve: enabling users to develop and operate Digital Twins; providing real-time views of smart city systems — with the proposed DT platform deliberately designed to be extensible beyond the smart city domain through the incorporation of additional DT fragments; supporting DT-driven analysis and decision-making; ensuring high-performance, cloudbased execution; and automating deployment and operational workflows. These epics function as a high-level coordination mechanism, thereby aligning technical development efforts with stakeholder value and broader system objectives.
All requirements have been aligned with the ISO/IEC 25010 Software Product Quality Standard to ensure coverage of key quality attributes such as performance efficiency, usability, maintainability, and portability. This alignment provides a shared reference for platform evaluation and reinforces the focus on delivering a trustworthy and extensible solution for Digital Twin deployment across diverse application domains.
Following the gathering of initial requirements for the Digital Twin (DT) platform, the project has advanced to the next phase, during which the development of the initial platform architecture is essential. To achieve this objective, a preliminary conceptual architecture diagram has been created (Figure 2).
The UML component diagram in Figure 2 provides a high-level representation of the DT system architecture, detailing the main components and their interactions. At the center of the system is the DigitalTwin component, described using the "DTDL" metamodel, and composed of modular Fragment elements for reusability. The development process is supported by DTDevelopment (a "view") and DTControllers, which collectively enable Digital Twin creation and lifecycle management. The DTMonitoring component oversees runtime observation and integrates with visualization (DTVisualization, DTDashboard) and layout components (VisualizationTemplate, ReportingLayout) via VisualizationControllers.
The system supports scenario-based experimentation using the DTExperimentation view, which is connected to Scenario entities expressed in "DTSL", and managed by the ScenarioManagement controller. Scenarios are executed through DTSimulation, using the SimulationEngine and deployed via DeployableEntity instances managed by a ContainerizationSystem. Analytical insights are derived through DTAnalytics, leveraging both Simulation and DTMachineLearning. Reporting is handled by a dedicated Reporting component, coordinated by ReportingControllers, and relies on ReportingData and DataIntegration for data acquisition and presentation. The overall data lifecycle is maintained by DataManagement, with hardware interfacing enabled through DeviceIntegration.
Given that the development of the detailed architecture is ongoing and will adhere to the C4
model standard [
This paper has presented the development of DigiTDevOps, an open and modular platform designed to support the modeling, deployment, and operation of DTs in smart city environments. A key contribution of the work lies in the structured definition of a multilevel DT data ecosystem and the systematic derivation of functional and non-functional platform requirements from real-world use cases. By integrating enterprise modelling and CDD principles, the platform fosters adaptability and supports simulation-based, context-aware decision-making.
In contrast to existing DT solutions, DigiTDevOps emphasizes reusability through modular DT fragments, scenario-based experimentation, and a focus on aligning system behavior with dynamically changing context conditions. These design choices contribute to the platform’s scalability and interoperability, while compliance with ISO/IEC 25010 standards ensures attention to core software quality attributes.
Future work will focus on refining platform functionalities, improving integration with openaccess datasets, and expanding its application across smart city domains. Additionally, further research is needed to address potential challenges, such as computational scalability, data privacy concerns, and integration with AI-driven decision-making. Next steps include refining system requirements, enhancing detailed architecture design, and evaluating real-world deployment scenarios. Platform prototyping will help identify additional architectural challenges, which must be continuously revised and optimized to ensure a robust and scalable solution.
Acknowledgements This research is conducted as part of the project "Development of the DigiTDevOps Digital Twin Development and Operation Platform"2 under the European Union's Recovery and Resilience Mechanism Plan. It falls within Reform and Investment Direction 5.1: "Increasing Productivity Through Investment in R&D," specifically under Sub-action 5.1.1.r (Reform): "Innovation Management and Motivation for Private R&D Investment" and Sub-action 5.1.1.2.i (Investment): "Support Instrument for Research and Internationalization" (4th round). The project number is 5.1.1.2.i.0/4/24/A/CFLA/001. The project is developed by Ltd. DATI Group as the lead developer and implementer in collaboration with Riga Technical University. 2 https://www.datigroup.com/en/projects/ipcei-next-generation-cloud-infrastructures-and-servicesdigitdevops During the preparation of this work, the author(s) used X-GPT-4 in order to: Grammar and spelling check. After using these tool(s)/service(s), the author(s) reviewed and edited the content as needed and take(s) full responsibility for the publication’s content.