Mindset in Agile

As a leading global manufacturer of pumps, systems and systems for liquid dosing, Lewa GmbH focuses on digital products and services as well as on agile work. The software service provider generic.de is providing important impetus to this. In an interview, the partners explain their approach.

Interview: Peter Trechow

What is the added value of digitized pumps?

Moritz Pastow: Our pumps often move dangerous, sensitive and sometimes expensive substances through the systems of our customers, who therefore attach great importance to process safety. As central drive units, digitized, sensor-monitored pumps not only provide data about their own status, but also important process parameters from within the systems. This makes it possible to monitor plant operation in real time and to use the collected data to identify trends and make predictive statements — for example about valve wear. At Lewa, we had the necessary diagnostic expertise even before the project. What we lacked, however, was the know-how to derive a digitally connected and Industry 4.0 compatible product from this. That's why we turned to generic.de.

How and when did the collaboration come about?

Pastov: As a VDMA member, we asked the Software and Digitalization Association whether they could recommend IT service providers to us. We were given a list of potential partner companies, from which we selected generic.de in a casting process.

Rene Paschold: The first prototype projects began in 2019. At that time, Lewa had carried out initial tests with hardware and software prototypes. We then helped them to develop marketable solutions from this, both in terms of cost, size, reliability and connectivity.

“Agile development starts a process of reflection on the product.” (Moritz Pastow)

‍So it was about the transition from planning the digital product strategy to implementing it?

Paschold: Yes, it was about defining the appropriate hardware, PLC solution and industrial PCs as well as developing a sustainable software architecture. Thanks to our experience, we were able to assist Lewa in this regard.

Pastov: It was only with generic.de's expertise that we found the current hardware setup. And they have also provided decisive impetus for software architecture. In working together, we rely on agile development, which was initially unusual for us — which led to a steep learning curve in several respects. Our first prototype did not have nearly as many diagnostic functions as the final product. The agile process is divided into two-week sprints, for which we agree on clear work goals in advance. This starts a process of reflection on the product, in which we have learned more and more about our products and about the goals and possibilities of their digital networking. This resulted in functional enhancements that would not have originally occurred to us.

Lewa has been building up mechanical engineering know-how since 1952. How can such a wealth of experience be translated into digital form?

Paschold: Lewa had defined digitization goals and prepared teams from various specialist and management levels for agile work when we started the project. They know their pumps and the diagnostic needs of their customers as well as the relevant parameters in detail. We were able to build on this when engineering requirements and agreeing on sprint goals. It is crucial to keep an eye on the added value of digitized pumps for Lewa customers at all times. This shows which key figures are relevant for the final product and which storage, processor, cloud connections or IT security requirements the solutions must meet.

What is the case for adopting agile methods from the software world in mechanical engineering?

Pastov: That is an exciting topic. We had to learn to work in an agile way. Mechanical engineers are used to implementing specifications. They work through predefined requirements step by step up to the start of production. In agile development, I initially look into a black box of content and only vaguely know what my goals are. We had to train ourselves to formulate our requirements in such a way that they did not directly lead the way. The help from generic.de was important for this. It is important to formulate and structure sprint goals or user stories in such a way that they build on each other in a meaningful way. And we had to convince colleagues in other departments as well as management and our customers of the agile process. They are not used to a finished pump receiving updates with new functions. This requires a new mindset. But agile work is absolutely suitable for mechanical engineering. Through the project, we have created an interdisciplinary “Digital Services” team and thus initiated an agile transformation. We are now involved in new pump projects right from the start in order to contribute our know-how about digital hardware and software as well as agile work.

Looking back, what were the most difficult hurdles — and how did you overcome them?

Paschold: If I remember correctly, we missed our sprint goals in the first few attempts and only achieved a solid, effective working mode after the fourth sprint. We had to find each other, develop a feeling for the scope of the work packages and learn that it is sometimes worthwhile to start over. Dead ends and changes of direction are part of the agile world — they contribute to clarity about the end product. At Lewa, it was also necessary to understand that in the work packages they were no longer responsible for how, but only for what. After all, the concrete implementation of software technology was our task.

Pastov: It proved to be absolutely right that our management also underwent the agile training courses and understood the new methodology. Because we need a new error culture and the insight that every course correction involves learning that moves us forward. Of course, despite everything, we had to justify our budgets, formulate goals and show results.

“Dead ends and changes of direction contribute to clarity about the end product.” (René Paschold)

Keyword budget: How is the market accepting their digital pumps, including the value-added services in the service?

Pastov: Persuasion is definitely necessary. Operators are reluctant to make changes to systems that frequently run 24/7. There is also sometimes a lack of strategic clarity on the goals of digitization in the market. The need for advice is therefore higher than expected. But the message that our digital pumps provide the necessary characteristics for smart factory operation is now being received. However, the challenge remains that digital pumps are not enough. They must be integrated into digital networks and IT security measures and connected to control centers in a meaningful way. And it needs clarity as to who can access which process data and when. It takes time.

Final question: What advice do you give machine builders who want to digitize but don't really dare to tackle the unusual subject matter?

Paschold: Lewa did a good job. They considered how digitization could improve their products, experimented with prototypes and sought help when their IT expertise was reaching its limits. It makes sense to start with a clearly defined project in which machine manufacturers and IT service providers can test the options, get to know each other and gradually deepen their cooperation. Agility can help with that. But it is not a panacea.

Pastov: It is important to understand your own expectations beforehand, to communicate them openly and clearly during the project — and to accept that the path to the goal is not always straight. A good digital product matures as it develops.

‍

‍

‍Published in: VDMA Magazine #01 /02 February 2022

Interview: Peter Trechow/ www.textschmiede.com
‍photos: Horst Rudel & Ines Rudel GbR/ vor-ort-foto.de

‍

More information:

Vanessa Koller, VDMA: vanessa.koller@vdma.org
‍Alexander Weber, generic.de: alexander.weber@generic.de