IIoT platform
Set up and operate IIoT infrastructure
For companies that want to integrate machine data faster, create a reusable database and use it to build applications and AI functions.
View platformBenefits and target image
Why companies are building IIoT infrastructure
A sustainable IIoT infrastructure does not just consist of connectors and databases. It is crucial that data sources, context, access and subsequent applications are based on the same technical and functional structure.
This reduces the integration effort, applications become reusable and analysis or AI functions can be created later on the same basis.
Typical challenges
Where infrastructure projects often stall
Heterogeneous system landscapes
Machines, controllers, brokers, databases and cloud services must be brought together, although data formats and operating models vary greatly.
High integration effort
The actual added value only arises above the basic infrastructure, while connectivity, data storage and operation already tie up a lot of time.
Lack of common data context
Without a consistent data model or a clear unified namespace, signals remain difficult to find and applications difficult to reuse.
Building blocks of the infrastructure
Together, these topics form the technical foundation for applications, analysis and AI.
Connectivity and Edge
S7, OPC UA, MQTT, REST and much more.Connect machine and IT data sources via standardized interfaces.
Edge-oriented and central operating modelsOrganize data acquisition close to the system or centrally across locations.
Reusable connector and rollout patternsRoll out similar machines, stations or lines in a structured manner.
UNS and messaging
Unified namespacePublish signals, events and context according to a consistent topic and structuring logic.
Event and streaming layerProvide live data for monitoring, further processing and downstream applications.
Connection to brokers and third-party componentsProvide downstream systems with the same event and context structure.
Data management
Historization and time series managementMake live data and histories available together for analysis, audits and queries.
Metadata and context modelLink assets, stations, products and processes with the data.
Data access via APIs and applicationsProvide data for dashboards, specialist applications or external systems.
Operation and expansion
On-prem, cloud and hybridAdaptable to existing IT and security requirements.
Roles, rights and governanceControl productive use across teams, locations and application layers.
Open source basisBased on Apache StreamPipes and expandable for your own modules.
Typical build-up path
The infrastructure is usually created gradually and provides a usable basis for initial applications at an early stage.
01Scope
Define data sources, operating model and target image
The first question is which assets, locations and target applications should be based on the same infrastructure.
02Structure
Set up connectors, data model and UNS
Sources are then connected, context structures defined and a reusable topic and data logic established.
03Utilization
Connect historization, access and applications
Historical data storage, APIs, dashboards and the first specialist applications then access the same infrastructure.
04Expansion
Add extensions and AI modules
Further analysis applications, AI notebooks and AI pipelines can be set up later on the same infrastructure.
Extensions on the same basis
Connectable products and AI modules
Both platform functions and AI modules can be based on the infrastructure without having to create a separate database.
Advanced modules
These pages take a closer look at how infrastructure is turned into concrete applications and AI functions.
IoT Data Hub
The product page describes the platform itself with its architecture modules, operating models and expansion options.
IoT Data Hub + AI
If additional AI-supported analysis and operating functions are to be based on the infrastructure, these are supplemented on the same database.
AI Notebooks
AI Notebooks access stored platform data and use LLM integration to generate explainable Python code for complex analyses, visualizations, tables and meaningful assembly analyses.
AI Pipelines
AI Pipelines supplements the infrastructure with standardized operating processes for the productive operation of models and data-driven processes.
Demo
Experience Bytefabrik live
Arrange a demo to get to know the platform, analysis and AI functions based on your questions.