Connected machines and IoT enable real-time production monitoring

In order to be able to help companies in their transition to an Industry 4.0 production, Sirris decided to set up its own pilot environment of an Industry 4.0 factory: the 4.0 Made Real Pilot Factory. Various innovative technologies will be implemented here and the production can be followed digitally from order to delivery. This thanks to connected machines and IoT.

As is already common practice in the industry, an order typically arrives via an ERP system. This results in production orders which are then translated into the various production steps. However, once the order has reached the production floor, everything is still done on paper (via the production folder with the production orders or instructions), and therefore the production status is unclear until the order is ready for delivery. However, customers increasingly want to be able to follow up on their orders in real time, while a (digital) overview is also desirable within the company, in order to be able to plan in the shorter term, to set priorities, to have a view of the production planning from the front office (which makes planning meetings superfluous), to better deal with unforeseen circumstances and to take pro-active action when delays are likely to occur. It is also getting more important to guarantee the traceability afterwards, when problems arise or to make sure processes can be adjusted.

Monitoring from ERP to production floor

To set up a digital monitoring process involving minimal manual actions, your first need to digitise the entire production equipment, connect the various machines and collect and use the production data. Not so obvious in practice, it seems.

For our pilot plant, we opted for an ERP system from Priority Software which receives orders online via a Twikit web-frontend and converts these into product orders. The ERP system is responsible for managing the typical order flow and production control processes. It defines routes, cells and products. Ready-to-use APIs and full web-SDK development tools also enable the connection of external data, IoT devices and third-party applications. This is particularly important since our ‘4.0 platform’ must be able to connect and communicate with the various machines, sensors and software.

We opted for the production of a watch, which requires some sub-assemblies, including sub-parts, which in turn require picking and work orders. It should also be possible to include all of this in the ERP. Challenges included: how to register this, and how to link the ERP system with the production? In order to find an answer to this question, different paths were followed, compared and combined.

Product tracking via unique code

A first method is providing an entry and exit station for kitting trays at each workstation. When an order comes in, a tray is put in active position and this is recorded via an RFID code that accompanies the tray during the production. The status of the order can be closely monitored by reading the tag.

A second way to follow what happens in the production cell: each object in the kitting tray gets its own position and barcode that is linked to the tray. For example, before a casing is placed in a milling station, the code is scanned. The processing data and conditions from the machine can now be linked to the product. When a problem arises, for example, it can be checked afterwards whether a malfunction occurred during the process.
However, not all machines can be connected yet, but registering a minimum of data for an operation (start-stop-failure) should also be able in this case.

Process monitoring at different levels

Digital product monitoring can take place at different levels. First of all we aim to follow-up information at a high level (machine status, settings, etc.), but we also want to be able to achieve this at other levels (in terms of inspections and their results, process parameters, etc.).. The machine controls do not always make the desired data available and especially with legacy machines, communication can be a challenge. We connect these types of machines in our pilot plant by adding sensors, where we test the advantages and disadvantages of communication with MQTT and OPC UA in practice.

Data storage and visualisation

We want to store the flow of data that is collected during the production in a structured way. To this end, we are investigating various possibilities for managing and using the data, for example via classic SQL databases and via NoSQL, time-series databases. The goal is for a manufacturing company to control the management of production data, to be able to change things as desired, to work via modules that can be linked to a database that is under its own management, without being completely dependent on one system or supplier. 

Wondering how far we have come with this? Come and see at the launch of our 4.0 Made Real Pilot Factory on 7 November in Hasselt and Liège!

This article is part of our ‘4.0 made real by Sirris’ campaign, illustrating the feasibility of 4.0 technology in industry. Want to know more? Visit our 4.0 Made Real page or LinkedIn page!