This page maps each Axomem Digital Twin Platform component to the Digital Twin Consortium's Capabilities Periodic Table (CPT) - an architecture and technology-agnostic framework for defining digital twin capabilities. For a portfolio-level view of each component and what it does, see the Products page.
The six CPT categories below colour-code every module and capability on this page. Click any module to expand its supported and roadmap capabilities.
Capabilities Periodic Table
Six capability categories from the Digital Twin Consortium framework.Data Services
Integration
Intelligence
User Experience
Management
Trustworthiness
Status legend: Supported Moderate Partial Backlog Future
Platform Modules
Click any module to expand its capabilities.DS.AI · Data Acquisition and Ingestion Supported
Configure and acquire data from control systems, historians, IoT sensors, smart devices and enterprise systems.
High-speed CSV ingestion via C++ and API-based ingestion. ODBC and many other ingestion paths via embedded Python.
DS.ST · Data Streaming Supported
Transfer large volumes of data continuously and incrementally.
High-speed inbound streaming via API. Outbound streaming possible via customer-developed C++ modules.
DS.TR · Data Transformation and Wrangling Supported
Clean, structure and enrich raw data to make it suitable for further processing.
Embedded Python provides flexible transformation. xBase C++ custom dynamic modules deliver high-speed processing.
DS.CX · Data Contextualization Supported
Add metadata to enrich real-time or transactional data.
Graph database contextualises any Thing (vertex) through typed relationships (edges) with their own data fields.
DS.BP · Batch Processing Supported
Execute against previously collected data in bulk form.
Task system supports short or long-running batch jobs within the server.
DS.RT · Real-time Processing Supported
Manage and act on captured data with minimal latency.
Soft-realtime via separate threads and memory-speed latency. Hard realtime possible via C API on an appropriate OS.
DS.DS · Temporal (Time-series) Data Store Supported
Store, organise and retrieve data relating to time instances.
Circular time-series buffer (configurable size) or continuously expanding time-series to ensure no data loss.
DS.AS · Data PubSub Push Backlog
Package filtered data to different services based on a publish/subscribe model.
Currently achievable via a custom C++ subscriber/publisher built on common pubsub frameworks. Native support is on the backlog.
DS.SA · Data Storage and Archive Services Future
Store, organise and retrieve data based on access frequency and retention.
IC.PR · Prediction Moderate
Estimate that a specified event will happen in the future or as a consequence of other events.
Basic engine to predict via risk aggregation. PyTorch training and inference supported via Python.
IC.AI · Machine Learning Moderate
Algorithms that improve through experience, building models from training data.
PyTorch and Scikit-learn via Python. Embedded C++ PyTorch interface available for higher performance.
IC.AI · Artificial Intelligence Moderate
Perform actions and take decisions like humans, including NLP, reasoning, inference and generative AI.
Integration with GenAI via APIs. Older AIML-style virtual assistance supported. Other AI capabilities available via Python.
IC.IC · Command and Control Partial
Execute work instructions without human interaction, limited to IoT devices and non-plant controls.
External commands triggered via APIs. Related to orchestration; a more complete command/control service is planned.
IC.PS · Prescriptive Recommendations Backlog
Create recommendations based on business rules and AI logic for optimal next actions.
IC.FL · Federated Learning Future
Train algorithms across multiple decentralised edge devices or servers without exchanging local data.
UX.BV · Basic Visualization Supported
Simple charts, graphs, dashboards, tables and basic 3D views of assets.
In-client visualisation of graph, relationships, valueset results and graph-based data.
UX.RM · Real-time Monitoring Supported
Continuously updated information streaming at zero or low latency.
Real-time visualisation of newly arrived data in the client. xBase can trigger processing on arrival of new data in soft real-time.
UX.GE · Gaming Engine Visualization Supported
Immersive virtual worlds and interactive experiences using gaming engine technology.
Native game engine visualisation via Unity-based client.
UX.3R · 3D Rendering Supported
Render 3D visualisations from point cloud data sets generated by LiDAR and other scanning technologies.
Native 3D rendering via Unity-based client.
UX.ER · Entity Relationship Visualization Partial
Present digital twin entities and their hierarchical or graph-based relationships interactively.
Some graph display already in xScape; backlog includes design-time editing and schema enforcement.
UX.AV · Advanced Visualization Partial
Complex charts, multi-system dashboards, animations and overlaid data visualisations.
Available via custom extensions for xScape or third-party tools integrated via APIs.
UX.XR · Augmented Reality (AR) Supported
Interactive experiences of real-world environments enhanced by computer-generated perceptual information.
HoloLens 2.0 supported. Any device supported by Unity can be targeted. Magic Leap 1.0 proof of concept completed.
UX.XR · Virtual Reality (VR) Supported
Simulated experiences that can be similar to, or completely different from, the real world.
Meta Quest 2 supported. Any device supported by Unity can be targeted.
IC.AA · Data Analysis and Analytics Moderate
Study and present data to create information and knowledge - charts, tables, dashboards, filters.
Basic spatial analysis in C++ services (e.g. outbreak analysis and spread maps). Wider analytics via embedded Python.
IC.MA · Mathematical Analytics Partial
Mathematical and statistical calculations to enable physics-based and other models.
Supported via Python or C++ library integration. Examples of integration with common packages on backlog.
IR.IO · OT/IoT System Integration Moderate
Integrate directly with control systems and IoT devices/sensors, SCADA.
Device management and integration with control systems and sensors. Full Linux systemd integration; EdgeX integration under consideration.
TW.DS · Device Security Partial
Authenticated and authorised access to IoT device data through identity management and policies.
Dependent on device setup. Integrates with Linux systemd; newer OS versions add native encryption and protection.
MG.DM · Device Management Partial
Provision, authenticate, configure, maintain, monitor and diagnose connected IoT devices.
Lifecycle management of edge devices including Thor-class hardware for video analytics deployments.
IC.AI · Edge AI and Intelligence Moderate
Make decisions at the device level based on real-time data, distributing analytics to the edge.
Edge components for video-based analytics now run on the Thor device - the underlying platform for the EGH video analytics work. Decisions are made at the device level on real-time video data rather than transporting all data to the cloud. xBase can also run part of the graph and most server services on the edge.
DS.RP · Digital Twin Model Repository Moderate
Store, manage and retrieve metadata that describes the digital twin model - formal data names, definitions, structures and integrity rules.
Built on experience with 1000+ bed hospitals, the repository is reusable across large-scale buildings, manufacturing facilities and comparable complex sites. Includes a library of pre-built open-source 3D FBX objects and full building models for Hospital, Hotel and Office Building scenarios, packaged via upm.axoverse.io.
IR.EG · Engineering Systems Integration Moderate
Integrate the digital twin with existing engineering systems such as CAD, BIM and historians.
Core function: extracting assets and structured digital twin data from any large Revit (BIM) project, including geometry, classifications and metadata - the basis for accelerating digital twin builds at hospital scale and beyond.
TW.PR · Privacy Supported
Control over what personal information is collected, stored, disclosed and to whom.
One-way hashes or encrypts PII data. Client supports per-user unlock keys for PII viewing. Using the Axomem Gateway server enables on-premise hashing/encryption so no PII is directly accessible in public cloud or via GenAI public services.
IC.SM · Simulation Partial
Approximate imitation of a process or system using historical information, physical models or animation.
Spatial simulation directly within the Unity editor. Push-back to xBase for logging, visualisation or replay on backlog. Data simulation on xBase possible via open-source C++/Python frameworks (e.g. SOFA, SimPy).
DS.SG · Synthetic Data Generation Partial
Generate synthetic data based on patterns and rules in existing sources.
Synthetic data generation framework with a "Random Patient Data Generator" example powering the AMB_Hospital demo.
UX.DB · Dashboards Moderate
At-a-glance views of key performance indicators relevant to an objective or process.
Standard dashboards based on computed values within the platform.
UX.BI · Business Intelligence Backlog
Analyse stored data to derive insights and actions in a business-focused interface.
Planned: Jupyter Notebook-style functionality on server with embedded integration, plus API-based integration to PowerBI, Tableau and similar tools.
IC.BR · Business Rules Backlog
Create, manage and use business rules that influence digital twin behaviour throughout its lifecycle.
Planned: integrate rules services based on JSON Decision Model (JDM), e.g. GoRules.
UX.BP · Business Process Management & Workflow Future
Execute a sequence of actions as a process flow to achieve business outcomes.
TW.RL · Reliability Supported
Perform required functions under stated conditions for a specified period of time - including performance, QoS and accuracy.
Achieved via underlying platform engineering. Axoverse IaC templates support multiple reliability options with cost trade-offs based on AWS cloud services. Azure platform support on roadmap.
TW.RS · Resilience Supported
Maintain an acceptable level of service in the face of disruption and recover lost capacity in a timely way.
Based on the same platform engineering as reliability capabilities.
TW.SF · Safety Moderate
Operate digital twins without unacceptable risk of physical injury, health damage or environmental harm.
Automated test frameworks for major components. Full rebuild and test of the platform performed every evening. Additional test framework templates available for client functionality.
Supported Environments
Where the platform can be deployed and the devices it supports.The Axoverse Digital Twin Platform supports a wide range of client end-user devices, including browsers, desktop executables, iOS, Android and selected VR and AR headsets. Any device supported by Unity 3D can be used as a client device for the platform.
The platform can be self-hosted on a wide range of server-side environments, including regular x86 servers, virtual machines, Docker/OCI-compatible containers and Raspberry Pi for IoT. The server also runs on desktops, laptops and NUC-type devices for development, proofs of concept and small-scale deployments.
The platform currently focuses on AWS for its cloud-based server option, using Terraform for full Infrastructure as Code (IaC) deployment and full container-based deployment.
CI/CD services run on Jenkins, including SonarCloud SAST and DAST scanning with OWASP ZAP. AxoMem can host client CI/CD environments or provide support to customer-hosted environments.