USE CASES
What customers actually solve with connubes
The tasks differ widely – from a container terminal to a spinning mill. What they share: it is almost never about new technology, but about letting existing systems talk to each other.
AUTOMOTIVE · POLAND
Stopping the wrong load before it happens
On this production line a barcode scanner decides whether a basket of material may go to the machine. connubes checks the scanned identifier against the running order in the MES and reports the result back to the Simatic S7 controller. If the basket does not match the order, the controller does not release it.
A side effect: every loading is archived. Traceability thus arises as a by-product of the safeguard, not as an extra task. All that stands at the line for it is a simple cabinet with a scanner and a status lamp.
DAIRY · DMK GROUP, ALTENTREPTOW PLANT
Nothing is pumped off until the milk has been tested
Milk collection tankers arrive at the plant around the clock. The driver carries an RFID chip on the key ring; the vehicle is thereby recognised as soon as it pulls up at the gatehouse, and the same identifier carries it on over the weighbridge. All the tanker’s data is available at once at every station.
Before pumping comes the test: a rapid test for antibiotic inhibitors following the HACCP concept, plus the pH value. The readings run from the test system into the laboratory system and into the plant systems – and release for pumping only follows if the milk is sound. Otherwise it is disposed of as the rules require.
After cleaning, weighing again and checking out, a message goes to the head office that plans the routes. Here the middleware does not only move values, it controls a release – and documents the chain without gaps, as the food regulations require.
An installation by inray Industriesoftware at the DMK Altentreptow plant, from a publicly documented user report.
ASSEMBLY · ANDREAS STIHL AG & CO KG
Sixteen lines, one architecture
Every assembly line communicates over its own middleware instance with a local SQL server; above them sit central databases that bring all the lines together. Sixteen installations in Germany work this way, plus plants in the USA and China.
Order data runs from SAP into the local server and from there to the controller; feedback goes the same way back to the shop floor data collection and the MES. Because parts, bills of material and production periods are available locally long before an order starts, the sentence that carries this architecture holds: If the connection fails, the current order and the one after it can still be produced.
At the first station every product gets its serial number, to which all process values and status codes are attached as it passes through. At the end of a line up to nine test cells work in parallel; each test run takes up to five minutes and is documented over its whole course.
An installation by inray Industriesoftware at STIHL; first published in Maschinenmarkt, 5 April 2019.
Further installations
Shorter examples from various industries – each with the core of the task.
Container logisticsA terminal operator visualises its equipment in real time. Crane, conveyor and scale report continuously to the control room – the basis for more autonomous operation.
Life scienceAt a chemical and pharmaceutical site, connubes has become the central supplier of production data. The documentation arises without gaps within the process.
Tool makingProcess values are read from the controller on events rather than polled at a fixed rate. That lowers the load and makes the data say more.
Pharmaceutical siteEvery department had its own island solution; two hours a day went on rounds of coordination just to establish a shared picture of the data. Today it is simply there.
Bicycle manufacturingconnubes sits between the database and the ERP system and keeps both sides in step – the heart of order processing.
Cutting toolsMachine data goes into the cloud and is brought together there with SAP data – as the basis for dashboards and machine learning.
PATTERNS
What runs through every case
| Starting point | What connubes takes on | Result |
|---|---|---|
| Values arise at the machine but are not analysed | Picking them up at the controller or sensor, passing them to database or cloud | A data base you can analyse, without touching the machine |
| Two systems do not know about each other | Reconciliation both ways, converting the data formats | One state instead of two, with no manual transfer |
| A mistake is only noticed late | Checking at the moment of the action, reporting back to the controller | The mistake is prevented instead of corrected later |
| Records are collected together afterwards | Archiving every action while operation continues | Traceability with no extra effort |