Many factories have more than enough data. Sensors measure conditions, and controllers deliver values ​​every second. Nevertheless, every new analytics or AI service becomes its own IT project: clarifying interfaces, building data pathways, coordinating infrastructure, and documenting everything. This is precisely where the it's OWL project I4.0AutoServ comes in. Developed within the it's OWL network, this one-stop-shop ecosystem helps companies bring data-driven value-added services into production faster, more reliably, and with significantly less integration effort. The core of the solution: machines, data-driven services, and the appropriate computing power are automatically matched. This is based on standardized asset administration shells (AAS). Put simply, this AAS is a digital profile that describes a component in such a way that other systems can automatically understand it.

I4.0AutoServ views the shop floor as a modular system. Every machine, every data-driven service, and every compute resource is assigned an administration shell. This shell defines the component's capabilities, the data it provides or requires, and the conditions under which it can be used effectively.

Users select, for example, a plant, a robotic system, or a laser cutting center via a graphical interface. They then choose one or more data-driven services, such as condition monitoring or anomaly detection. The ecosystem uses information from the administration shells to identify suitable combinations. It establishes the data flows and controls the execution via the connected components for the data platform and deployment.

Many sophisticated individual solutions thus create a structured service landscape. Components can be reused and combined in different scenarios, instead of having to start from scratch for each application.

"Our Industry 4.0 ecosystem for the automated deployment of data-driven services enables companies to implement product-service systems faster and more cost-effectively."

Dr. Marc Hesse, Team Leader Cognitronics at the Faculty of Technology of Bielefeld University

For companies, this means: New data-driven services can be selected more precisely and integrated into existing production with less effort.

“Our Industry 4.0 ecosystem for the automated deployment of data-driven services enables companies to implement product-service systems faster and more cost-effectively,” says Dr. Marc Hesse, Team Leader Cognitronics at the Faculty of Engineering at Bielefeld University. “This is of interest, for example, to component manufacturers, machine builders, or manufacturing companies.”

Automatic matching in two steps

The core of the ecosystem is the matching service. It maps the interaction between the shop floor and IT in two steps. The following diagram illustrates how this matching works, showing the matching of shop floor assets and data-driven services.

In the first step, I4.0AutoServ compares the machine descriptions with the service requirements. Capabilities, available data, sampling rates, and interfaces are stored in the administration shells. Based on this information, the system checks which services can be used effectively on which machines and which machines are suitable for a particular service.

Matching shop floor assets and data-driven services.

In the second step, the matching service considers the required computing power. It determines which infrastructure meets the latency, performance, and platform requirements of the service. The spectrum ranges from edge devices directly on the machine to edge cloud environments and the public cloud.

Ultimately, users see a selection of technically validated combinations of machine, service, and compute resource. They decide which combination they want to use for testing, training, or operation. The system then automatically handles the necessary data flows and deployment.

In practice, this means: less manual coordination between production, IT and external service providers, more clarity about what fits together technically.

Administrative shells as a common language

For this automatic matching to work, a common language is needed for all participants. I4.0AutoServ therefore relies on defined AAS sub-models. These sub-models describe the identity, interfaces, technical data, and capabilities of a component in a structured format.

This includes digital nameplates, technical data, handover documentation, hierarchical structures, and interface descriptions. The capability submodel is particularly relevant. It defines what a component can do and what constraints apply. Based on this, it can be verified whether a machine provides the data required by a service and whether an environment supports the execution of that service.

With I4.0AutoServ, we describe machines, data-driven services, and computing resources uniformly using administration shells. This allows an algorithm to transparently check which combinations are technically compatible. In many companies today, this task still involves a significant amount of manual work.

Dr. Magnus Redeker, Head of Mathematical Optimization at Fraunhofer IOSB-INA and Project Manager of I4.0AutoServ

“With I4.0AutoServ, we describe machines, data-driven services, and computing resources uniformly using administration shells,” says Dr. Magnus Redeker, Head of Mathematical Optimization at Fraunhofer IOSB-INA and project manager of I4.0AutoServ. “This allows an algorithm to transparently check which combinations are technically compatible. In many companies today, this task still involves a lot of manual work.”

The sub-models unify identity, interfaces, capabilities, and requirements. They form the basis for automatic matching, the orchestration of data flows, and traceable deployment throughout the entire lifecycle.

For decision-makers, this means that investments in data and services are based on a clear, standardized framework rather than on individual, isolated solutions.

https://youtu.be/SSEnqeaYVng Video can't be loaded because JavaScript is disabled: App Store for Machines? How Software and Services Match | Inside it's OWL (https://youtu.be/SSEnqeaYVng)

Experience autonomous robots live at the Hannover Messe trade fair.

An intralogistics application with autonomous robots demonstrated how I4.0AutoServ can be used in practice at Hannover Messe 2024. Autonomous mobile robots (AMiRo) moved through a sample environment. Three different data-driven services, including those for anomaly detection and condition assessment, were connected via the ecosystem, automatically matched, and executed live.

The robots provided sensor data such as accelerations and motor settings. This data was transferred to the data platform via connectors. Training and inference of the models were performed using a DDS Toolbox, a toolkit for data-driven services (DDS) based on the it's OWL project ML4Pro².

The demo demonstrated how I4.0AutoServ integrates shop floor assets, data flows, and services. For companies, this setup shows how multiple data-driven services can be operated in a controlled manner within a networked system.

A look at the interface for controlling the I4.0AutoServ system.

Predictive maintenance in the laser cutting center

A second application comes from ongoing production at Remmert. The drives of the SortFLEX sorting robot in the company's laser cutting center deliver high-resolution measurement data. This data, provided as oscilloscope traces from Lenze inverters via MQTT, serves as the basis for a vibration-based monitoring system. The goal is to detect wear on mechanical components such as timing belts at an early stage.

“For us, the benefits are quite pragmatic: We detect wear and tear earlier and can plan interventions better, instead of only reacting when something breaks down. Because data paths and services are cleanly described via administration shells, we don't have to start from scratch every time. This allows us to implement such improvements more quickly and with less effort in other areas as well,” says Rafael Schroeder, software developer at Remmert.

The system, services, and data paths are described in administration shells. I4.0AutoServ uses these descriptions to support matching, integrate the monitoring service, and control execution. The evaluation is based on frequency analysis. Changes in vibration patterns can indicate the onset of wear.

This provides Remmert with a practical entry point into condition monitoring and predictive maintenance. Existing data is used to enable more targeted planning of maintenance and service.

What advantages can companies use?

From a business perspective, the added value addressed by I4.0AutoServ can be summarized into several points.

The introduction of new data-driven services is accelerating. Companies can move from individual lighthouse projects to scalable rollouts. Standardized descriptions and automated matching reduce manual integration effort and facilitate the transition from pilot to regular operation.

Efficiency increases because data flows, services, and computing resources are automatically orchestrated. Fewer manual interventions mean fewer potential sources of error. This supports stable processes, especially in complex, interconnected systems.

Flexibility increases because machines, services, and computing resources can be dynamically combined. When production conditions change or new applications emerge, suitable combinations can be identified and tested more quickly.

Standards-compliant approaches ensure interoperability and future-proofing. Industry 4.0-compliant standards provide a framework for digital business models and reduce dependence on isolated, stand-alone solutions.

Small and medium-sized enterprises (SMEs) in particular benefit from the fact that low-code approaches, automated processes, and assistance functions reduce the need for specialized knowledge. Production and maintenance professionals can use data-driven services without having to become data experts themselves.

Who should take a look at I4.0AutoServ

I4.0AutoServ addresses two key groups in industry. Component manufacturers and machine builders can use the ecosystem to develop product-service systems, such as condition monitoring, remote services, or configuration services for their products.

Manufacturing companies use the platform to build production service systems. These include applications in quality assurance, throughput and capacity optimization, and predictive maintenance based on existing machine data.

In both cases, the ecosystem provides a structured basis for gradually testing data-driven services and standardizing them if necessary.

Book a free demo

Entry into the ecosystem can be done gradually. A pragmatic approach is to first describe and match a single shop floor asset with suitable data-driven services and monitor its operation. This could be a machine, a production line, or a clearly defined process.

The building blocks developed within the it's OWL environment demonstrate how administration shells, data platforms, matching, and deployment interact. Companies can use this experience as a guide to plan their own use cases and systematically develop data-driven services towards a smart, data-based shop floor.

“Let your first asset be automatically matched with suitable data-driven services and run live. Book a demo, test the ecosystem in your environment, or bring a specific use case; we support you from matching to productive rollout,” says Redeker.

The article "Less integration effort and real added value: How I4.0AutoServ automatically connects machines with data-driven services" first appeared on it's OWL .