Unlocking Shop‑Floor Intelligence: How Vorwerk Accelerated Digitalization with Insights Hub
Vorwerk has made digital transformation a core part of its manufacturing operations across two factories in Germany and France, supporting flagship product lines such as Kobold cleaning devices and Thermomix cooking devices. With a strong “data in the DNA” mindset, the company has long focused on analyzing processes and driving continuous improvement. Digitalization strengthens that culture by creating transparency across operations, enabling consistent KPI tracking, and giving teams a clearer, shared view of how production is performed in real time.
Behind that objective lies a clear operational intent: to make factory performance visible, measurable, and easier to improve without relying on spreadsheets. As Vorwerk’s production environment continues to evolve, systems such as MES and Insights Hub serve as the critical link between shop-floor activity and the overarching production plan. This connected approach enables improvements to be implemented more quickly and sustained more effectively over time.
We approached digitalization like an architecture problem: remove silos, establish MES as the backbone, then connect machines through a modern layer so order data and IoT signals meet in Insights Hub, because that’s where operational transparency and better KPI performance actually happen”
Björn Krückhans, Business Process Manager Production, Vorwerk
Building the Baseline: MES as the Foundation
Vorwerk’s modernization didn’t start with flashy AI demos or dashboards. It started with something more fundamental: dismantling siloed systems and creating a consistent baseline with MES. About five years ago, the focus was on getting the operational backbone right, standardizing how production is executed and measured so every improvement has a reliable reference point.
That foundation came together through a Siemens-centered stack: MES as the core layer, Opcenter Execution and Opcenter Scheduling for running and coordinating operations, and strong integration with Teamcenter to connect product, production, and enterprise data. The value here wasn’t just “integration” in theory; it was visibility into order data that production teams could use. Once that order-driven view was in place, Vorwerk achieved a single, end-to-end picture of production, bringing execution, scheduling, and order context into a single, clear source of truth.
Connecting the Machines: Edge, IoT Data, and Insights Hub
Once the baseline was stable, Vorwerk expanded the scope from order-related data to something even richer: Industrial IoT data from machines. This is where Insights Hub and edge devices became critical, as they enabled Vorwerk to bring together two key data streams, production order context and real-time machine and process signals, within a single environment. That shift is important because order data shows the intended flow, while IoT data reveals the true operating conditions. When those perspectives come together, teams can move from after‑the‑fact reporting to faster, more precise operational improvements.
One of the most compelling outcomes appeared in a KPI that directly reflects day‑to‑day performance on the shop floor: output pieces per hour. It is practical, and measurable for productivity. With the enhanced integration and transparency enabled through Insights Hub, Vorwerk recorded a significant improvement in this metric. This positive trend demonstrated that the initiative delivered tangible operational value.
Improving What Matters: “Pieces Per Hour” as the North Star
Output pieces per hour is the target KPI on the shop floor, not just a metric that Vorwerk tracks, but one that has improved significantly through better integration and the use of data-driven insights. The important detail here is how the improvement happened: not just collecting more data for the sake of it, but by merging previously separate worlds that were operating in parallel. Before, systems were siloed, and intelligence was limited. Over time, those layers grew together, connectivity, execution, scheduling, engineering context, enterprise systems, and IoT data, until performance improvement became easier to drive and easier to prove.
The key point is how this improvement was achieved. It was not simply a matter of gathering more data, but of unifying previously disconnected domains that had long operated in parallel. Historically, systems were siloed, limiting their collective intelligence. Gradually, however, these layers of connectivity, execution, scheduling, engineering context, enterprise systems, and IoT data became increasingly integrated. As this convergence took hold, driving performance improvements became both more efficient and more demonstrable.
Moving Toward AI: Quality Prediction and Maintenance Support
With the data foundation and integrations maturing, Vorwerk began exploring AI use cases, starting with quality prediction in collaboration with Siemens teams. The direction is straightforward: if data access becomes easier and more contextual, then formulating questions becomes easier too, and that’s where AI starts to feel practical rather than experimental.
One example is maintenance. Imagine a maintenance technician needing a fast answer in the middle of a task: how a repair was done last time on a specific machine, what steps were taken, what parts were used, and what outcomes followed. Instead of searching documents or asking around, the idea is to query an AI assistant that can respond immediately with the right history and instructions—helping technicians move faster and reduce downtime.
With a solid data foundation and increasingly mature integrations in place, Vorwerk is beginning to explore AI-driven use cases, starting with quality prediction in collaboration with Siemens teams. The rationale is straightforward: as data becomes more accessible, contextualized, and reliable, the questions that teams can ask become more precise. This is the point at which AI transitions from an experimental concept to a practical, operational tool.
A strong example emerges in maintenance, consider a technician who needs immediate clarity during a repair, how a similar intervention was performed previously, which steps were followed, what components were used, and what results were achieved. Instead of searching through documentation, an AI assistant could provide instant access to the relevant history and recommended actions. This capability has the potential to accelerate troubleshooting, support consistency, and significantly reduce downtime.
Lessons Learned: Knowledge Sharing and Choosing the Right Partner
Looking back, Vorwerk’s takeaway is grounded and practical. Digitalization isn’t just a technology decision; it’s an ongoing journey of standardization, learning, and scaling what works. The emphasis is on gaining knowledge, sharing it across teams, and selecting partners who are truly ready to support the full journey, not just the first phase or the easiest wins.
The bigger story is that transformation isn’t one big leap. It’s a series of deliberate moves: build the baseline, connect the machines, unify data streams, improve KPIs that matter, and then expand into advanced use cases like AI once the foundation can support it. Done that way, digital transformation stops being abstract and becomes measurable, scalable, and genuinely useful on the shop floor.
