In smart manufacturing, a scalable edge platform and the ability to quickly implement business-relevant use cases are crucial for success. With Azure IoT Operations and Microsoft Fabric, Microsoft offers two revolutionary technologies that provide real-time insights into production. Learn how these solutions unlock the potential of use cases such as predictive maintenance, process optimization and generative AI, transforming them from concepts into real value drivers.
With Microsoft Fabric and Azure loT Operations, the challenges of implementing smart manufacturing can be mastered with ease. Find out now what a possible solution architecture could look like and learn more about the exciting lessons learned from existing projects.
The right smart manufacturing architecture is the key to efficiently networking factories and the machines and systems within them and reacting flexibly to changes. A successful architecture is characterized above all by scalability with an increasing number of devices and Al capability at the edge and in the cloud. Based on this, we have created a reference architecture with the Azure loT Operations (AIO) service and Microsoft Fabric, which has been generally available since Ignite 2025.
At the edge, we rely on Azure loT Operations as a highly available and modular edge solution based on Kubernetes. Kubernetes has matured into the central technology at the edge, which uses the principles of GitOps to simplify the operationalization of the platform for IT and OT teams and enables zero-downtime deployments and traceability. For a deeper insight into the role of Kubernetes and GitOps in digital transformation, click here.
In the cloud, we rely on Microsoft Fabric, an Al data platform that offers comprehensive frameworks for data integration, analysis and the development of machine learning models. This allows data from different sources such as sensors, machines and controllers to be integrated via loT operations or various ERP systems such as SAP or many more and organized with manufacturing data models.
Azure IoT Operations offers a variety of data services for connecting assets from production, for contextualisation and for implementing machine learning applications:
Our experience with the implementation of use cases and the integration of factories via Azure IoT Operations has shown that there is still room for improvement in some areas. We encountered the following problems during project implementation:
With Real-Time Intelligence (RTI), condition monitoring can be implemented in the shortest possible time and with the Data Activator, findings can be converted into actions in no time at all. Anomalies in the sensor data can be recognized and alarms triggered. Direct integration into Teams or Outlook enables the right people to be notified directly.
The challenges often based in the fundamental design decisions when building a scalable real-time architecture. Two frequently asked questions are:
There are different answers to these questions depending on your preference, and we will try to answer them below based on our project experience.
Two properties have an impact on the architecture. Firstly, the possibility of carrying out transformations directly in the Eventstream using no-code or low-code approaches and, secondly, the fixed costs that are incurred per Eventstream. Due to the costs, it is advisable not to become too granular when creating the Eventstreams and to think about a possible grouping in advance. Experience has shown that grouping per factory is often a good choice in smart manufacturing. With regard to transformations, it should be noted that these can only be carried out in the event stream across all sources and the restriction from the previous section must be taken into account. Our recommendation in this context is to carry out the transformations in the Eventhouse on the basis of update policies and KQL.
The figure above shows an example of an event stream for a factory with two connected machines, the integration and processing of the data in the Eventhouse and a Data Activator for real-time reactions based on the telemetry data.
The implementation of smart manufacturing poses major challenges for many companies. It is therefore crucial to use the right tools and drive implementation forward with a holistic strategy. With Azure IoT Operations and Microsoft Fabric, in particular Real-Time Intelligence, companies have the tools they need to network their factories and implement use cases such as condition monitoring, predictive maintenance and many more efficiently and sustainably.
If you are also facing the challenge of implementing smart manufacturing in your company, we are happy to support you. Contact us to find out more about our b.telligent “Way to Fabric for Manufacturing”!
Domain Lead Cloud Transformation & Data Infrastructure
Do you want to replace the IoT core with a multi-cloud solution and utilise the benefits of other IoT services from Azure or Amazon Web Services? Then get in touch with us and we will support you in the implementation with our expertise and the b.telligent partner network.
Many users of Google's IoT Core are currently looking for a successor to this service which will expire in August 2023. This blog post shows how Stackable's data platform can be used to create a highly scalable open-source alternative to Google's IoT Core.
Lack of resources or technical challenges are often hurdles to establish the value and viability of IoT use cases, and present them later to project sponsors. Even for simple IoT use cases, sometimes weeks instead of days may be needed to produce tangible results. In this blog, we’ll present our IoT kick-starter platform that makes it possible to technically assess simple IoT use cases within a few days.
Microsoft describes its IoT Operations as a “Range of modular, scalable, and highly available services” that run on Azure Arc-enabled edge Kubernetes clusters. It includes native integration of other MS services like Event Hub or Fabric, whereby the latter’s been generally available since November 2023. Our blog summarizes the new Azure service’s features and deployment.
No previous post
No next post
