Productivity management via app: capturing and understanding machine data

Efficient work only works when everything runs smoothly – also on the shop floor. But how do I know how busy and productive my machines are at the moment? This requires data that reflect reality, but which is often still lacking today. ENLYZE has developed an analysis platform that can close this gap and make production more efficient.

In an interview with K-MAG, Julius Scheuber talks about data-driven productivity management and how it can be used to achieve higher performance, less downtime and fewer rejects.

You have developed a solution for data-driven productivity management. What problem are you addressing with it?

Julius Scheuber: Most companies today are "blind" when it comes to their shop floor. They often have no live information about utilisation and productivity. If this data is available, it is usually based on manually recorded data by the workers. But this does not reflect what is really happening on the shop floor.

Fast, efficient work and the clarification of important questions that arise in production are not possible in this way. By this I mean questions like: Which plant had the highest productivity this week? What was the top reason for a shutdown and how many hours of downtime does this reason allocate? Which product causes the most frequent production problems?

ENLYZE provides insights into all these questions based on machine data. In our app, the most common problems can thus be quickly identified and rectified so that production is always running at a top level.

How does that work?

Scheuber: We connect our edge device to all systems from which data is needed. Typically, these are the PLC and, if necessary, other sensors such as the inline quality measurement. Afterwards, the relevant data points are identified and transferred into a data model. We also connect to the MES/ERP system to obtain order and product data.

The data is then merged in our analysis platform "Shopfloor BI". Some selected features of this platform are:

• The automatic downtime recording: here, all downtimes of your plants are automatically recorded based on machine data and KPIs are displayed in order to remedy them.

• The co-pilot for workers: Like a digital setting sheet, the ideal setting parameters for the product to be manufactured are displayed and compared with the values currently applied to the line.

• Search: With one click, the entire production history of an order can be made accessible with all its data. Questions of traceability can thus be clarified quickly and easily.

From which sectors do your customers come?

Scheuber: Our app, especially the out-of-the-box analyses (OEE, process analyses, standstill detection) are designed for continuous production – be it the production of noodles, wire, paper, plastic films et cetera. Wherever a long, stable process is to be achieved, we have our expertise. That's why no industry can be so firmly assigned.

Schüco is one of our biggest customers. However, the majority (around 60 %) are from the plastics extrusion industry, for example Storopack, Bischof+Klein and Duoplast. They are all united by the motivation to manufacture excellent products, and for this they rely on reliable data from their production.

What does a concrete application look like?

Scheuber: Our customers realise the most diverse use cases as soon as they have access to their data. However, we always start by identifying the biggest losses in production. After this step, our customers know exactly where the problem areas are and what the causes are.

Our customers achieve on average 5% more output, save 3 hours of downtime per month and can reduce their scrap by about 2%. Of course, the latter makes us doubly happy in terms of sustainability!

In addition, there are other very special but enormously interesting cases that make us proud:

• One customer was able to perfect its quality management to such an extent that delivered defective goods could be traced back to a specific batch, thus reducing returns many times over.
• In our latest analysis of 1,328 products across 12,413 customer production orders, we were able to uncover 8.9% performance variations for the same product that would have gone undetected without systematic recording.
• An adhesive tape manufacturer was able to save about 30 % of scrap by detecting a production error ("rewinding").

In general, it is evident that use cases only become clear once data collection has been set up and the biggest losses have been identified.

You also participated as an exhibitor at K 2022. How did the trade fair go for you?

Scheuber: K was the first trade fair ever for us, as we entered the market virtually at the beginning of the Corona crisis. Our expectations weren't huge, but we naturally hoped to learn a lot about the market through discussions, to talk about our product and to win over one or two potential customers.

K exceeded our expectations by far – and not only because of the amount of interested parties, but also because of the great insights into the state of the art at the stands, the in-depth exchange of experiences with other companies, especially start-ups, and a general aura of innovative spirit and technological progress. For us, the fair was a complete success, and our founders are still talking about the fair party today. We are already looking forward to the next K!