Metal and Electronics

IoT in metal and electronics manufacturing means presses, melting furnaces, pickling lines, lasers and pick-and-place machines report their own condition. Steel mills spot wear before a line stops, PCB makers track every board through 30 or more process steps, and foundries see the fill level of their furnaces in real time.

Here you will find more than 40 implemented solution examples from users such as Salzgitter Flachstahl, Aurubis, Würth Elektronik, Bosch and CERATIZIT, each with the provider who delivered the project. No marketing, just what actually runs in the plants.

IoT in Metal and Electronics Manufacturing: What Connected Plants Achieve Today

In steel mills, foundries, sheet metal shops and electronics plants, micrometres, seconds and degrees decide between a good part and scrap. At the same time, machines from several decades run side by side, with controllers from different vendors. IoT brings both together: sensors and gateways pull condition and process data out of existing equipment, a platform combines it, and maintenance, quality and production work from the same numbers.

The examples below come from implemented projects in the IoT Use Case network. Each one names the user, the provider and the technology used.

Typical IoT Use Cases in Metal Processing and Electronics Manufacturing

Predictive Maintenance and Condition Monitoring in the Steel Industry

The continuous pickling line at Salzgitter Flachstahl processes around 2,500 tonnes of steel per shift and only stops every two weeks for planned maintenance. More than 100 measuring points along the 330-metre line now report their condition over a second, read-only fieldbus channel to the Endress+Hauser Netilion platform, without touching the automation. Technicians know where the fault is and which spare parts they need before they head out. Details in High availability in flat steel production.

Copper producer Aurubis connected its entire production with azeti, from smelting furnaces and electrolysis to wire drawing machines. Out of 1,500 process parameters, the team picked the truly critical ones and built dashboards and machine learning models for predictive maintenance on top (Aurubis solution example). A steel producer with more than 10 million tonnes of annual capacity monitors conveyors, motors and fans with Honeywell LoRaWAN vibration sensors (Fan condition monitoring in steel manufacturing). More on the approach in the condition monitoring use case.

IoT in Sheet Metal Fabrication: Machine Data as the Basis for Billing

Sheet metal shops often do not know exactly which jobs are profitable because actual machine hours and cost per part are missing. The Riguzzi Group therefore uses a pay-per-part model with TRUMPF, Relayr and Munich Re: the connected machine counts the parts produced, and payment follows usage instead of a large upfront investment. See On the growth path with pay-per-part.

Machine Data Acquisition and Tool Monitoring in Machining

With the autinity software from Schaeffler Digital Solutions, even older machines from different vendors can be connected. Continuous data capture detects tool breakage, chatter during machining and excessive tool wear, and provides the basis for OEE figures (From machine data acquisition to predictive maintenance). On the supply side, CERATIZIT and Materna connected around 1,000 tool vending machines in 22 countries to the cloud and to SAP, each holding up to 840 items (Tool dispensing from the IoT cloud).

Electronics and PCB Manufacturing: Data from Every Process Step

A printed circuit board goes through about 30 production steps, multilayer boards up to twice as many, and pressing repeatedly erases markings. Würth Elektronik introduced the FabEagle MES from Kontron AIS to evaluate machine states, alarms and process data across three sites (More efficient PCB production thanks to a solid database). KSG GmbH and Schaeffler Digital Solutions connected a heterogeneous machine landscape through standardised IT/OT integration, because in the chemical processes every deviation quickly costs entire batches (IT/OT integration in PCB manufacturing).

At the Bosch semiconductor fab in Reutlingen, the journey from raw wafer to chip takes up to 14 weeks. Staff used to walk the corridors to find faults; today Peakboard displays show the status in real time in one central place (Data visualisation in semiconductor production).

Foundry and Metallurgy: Monitoring the Melting Process

Norican Group connected its dosing furnaces to the azeti platform. The Refill Monitor shows foundries in real time how full each furnace is, so the process never runs short of molten metal (Optimising the melting process in the foundry). In vacuum metallurgy, ALD Vacuum Technologies records every melt with pressure, temperature and gas flow in melt review sheets and hands the data over to ERP and MES (Documentation and traceability in metallurgy).

Quality Assurance with Cameras and AI

When coating titanium turbine blades, an ALD operator used to watch the vapour cloud through a small viewport for an entire eight-hour shift. Today a camera-based system monitors the process and documents it for aerospace customers (Automated quality assurance for coating systems). How process data reveals the causes of scrap is shown in Process data analysis with a digital twin. More examples in the quality management use case.

Traceability of Metal Parts and Material Documentation

If you have to prove in a complaint which batch, which parameters and which machine a part came from, you need a digital identity per part or lot. RFID links production orders to their position in the plant in real time (Track production orders in real time). Solder maker Stannol uses blockchain with Minespider to prove which mine the tin in its solders comes from (Tamper-proof supply chain evidence for fair-trade tin). More in the traceability use case.

The pressure is growing: steel and aluminium are among the first product groups for which the EU Ecodesign Regulation (ESPR) is preparing requirements and a Digital Product Passport, and since 2026 the Carbon Border Adjustment Mechanism (CBAM) requires reliable emissions data for imported steel and aluminium. Both can only be met efficiently with data from production.

Energy Monitoring in Energy-Intensive Processes

Furnaces, pickling lines, electroplating tanks, reflow ovens and compressed air are among the largest consumers in a plant. Sensors on compressors reveal idle running and leaks (Detect unused compressor runtimes and leaks, Cut energy costs with compressed air monitoring), and energy data can be analysed live in the plant without coding (Evaluate energy data live in the plant). The same data feeds the carbon footprint and ESG report. More in the energy monitoring use case.

How IoT Improves Efficiency in Metal Manufacturing

The gains almost always come from the same four places: fewer unplanned stoppages because wear is detected early; less scrap because process deviations show up immediately rather than at final inspection; shorter maintenance jobs because technicians know the fault location and spare part in advance; and lower energy use because idle running and leaks become visible. Progress is usually measured through overall equipment effectiveness (OEE), which only becomes reliable with automatically captured machine data.

What Makes IoT in Metal and Electronics Different

Heavy vibration at presses and stamping machines, electromagnetic interference at welding systems, heat at furnaces and aggressive media in pickling and electroplating call for rugged, industrial-grade sensors. On top of that comes the mixed machine park: OPC UA is far from standard everywhere, and many machines speak Profinet, Profibus, Modbus or EtherNet/IP. Platforms that connect these protocols directly or unify them through an edge layer save the most time here.

It is also essential not to put the control system at risk. Salzgitter deliberately reads its instrumentation only through a second channel and transmits data encrypted via TLS, so nobody can interfere with operations from outside.

How to Start an IoT Project in Metal or Electronics Manufacturing

1. Start with a specific question, not with technology. Würth Elektronik first clarified which alarms should be captured at all and how they would be evaluated.

2. Pick one critical asset whose failure really costs money, such as a pickling line, a press or a pick-and-place machine.

3. Use the data already in the controller and only retrofit sensors where values are missing, as Aurubis and Norican did.

4. Define the few metrics that matter: Aurubis narrowed 1,500 parameters down to the critical ones.

5. Run it as a pilot, measure the result, then roll it out to further lines and plants.

IoT Providers with References in Metal and Electronics

The projects on this page were delivered by providers including Endress+Hauser, azeti, Schaeffler Digital Solutions, Kontron AIS, Honeywell, Peakboard, ALD Vacuum Technologies, Materna, Minespider and TRUMPF. Below you will find all solution examples for the industry, with users and partners. If you are looking for a provider for a specific project, get in touch and we will connect you with the network.