We spoke to Frank Moos, Vice President Global Process Management, about the role that robust business process management plays in the company’s success and how assistance systems support this.
KSB’s holy grail is configuration and complexity management. Our product catalog contains around 360,000 variants that can be ordered via our web store. We can currently generate 2.5 times 1047 variants in our configurators. Even in the case of the most standardized of our pumps, we have a repetition rate of 1.8, which is immense. The fact that we can offer so many variants presents us with a variety of challenges, ranging from procurement of parts to documentation. And it only works if the business processes behind it are set up properly. That’s where our central global process management comes in.
We provide the methods and tools in my team. We use MO²GO, a model-based process assistant developed by Fraunhofer IPK, as well as a process mining software from Celonis. We also have so-called global process owners who define the framework for all processes and the process architecture. For standard products in our web store, the sales process is always the same: there is a customer inquiry, a quote is created, the order is produced, delivered and paid for.
For large industrial pumps, things become much more complex, because here we have an additional stakeholder management. We have to coordinate technical and commercial requirements, and consider geopolitical conditions as well. To do this, we have global process owners who keep an eye on different markets, products and business types and set up a framework that takes country-specific regulations and market conditions into account. Below this level, local process owners manage or shape the requirements at a local level, for example where there are different local laws and regulations. In addition, we have established our lean management team. These levels must all interact operationally. The idea we implement at KSB is this: We provide a central process tool as an architecture, in our case MO2GO, and use it to create uniformly harmonized processes and standard tools from top to bottom as far as is necessary. If more details are needed, we use procedures and guidelines to further specify them internally.
With its object-oriented modeling approach, the Fraunhofer IPK process assistant makes it possible for us to structure processes in a transparent way – both horizontally and vertically. For example, I might say: I create an offer, followed by ten or twenty steps in different sequences and I’ll know exactly what happens at the end. But I can enrich this with every resource I need to carry out this process. So I don’t just look from left to right, I also see what happens in each process frombottom to top. Which roles, which documents are involved? Which guidelines affect a process, which machines? I can connect any number of different variables to a process and check: What happens if I adjust a particular variable and how does that affect my process model? This contextual view of processes and resources that MO2GO provides is an immense advantage over any other process documentation.
This also benefits our large international pump projects, which often run for several years. Based on the MO2GO process architecture, we use a project process assistant that is generically structured and supports us in quality assurance with milestones for process-based projects. This really gives us an up to date picture of a project situation and allows us to identify: Okay, in 9 out of 100 projects, we have pending issues that pose a risk, and at the same time we can see what actions can be taken to address them. Our project managers now use this project process assistant regularly and perform weekly checks: Where do individual projects stand? Which steps in the processes have we completed? Which quality gates have we achieved? Are there lessons learned? We discuss this each week, then fill the project process assistant with the relevant feedback and work with it continuously. The highlight: If we realize at the end of a long project that we should adjust a process step, for example to improve quality, we can see the impact of this change across all ongoing projects. We see where else we can implement this improvement, where we can apply it retroactively or – when in doubt – where a risk could arise and how we can resolve it.
Here too, we rely on the process architecture and consider every process step that we model: What could go wrong, i.e., what are the potential risks? How significant could their impact be? The risk management web app that we are developing together with Fraunhofer IPK visualizes these risks. Similar to a fever thermometer, I can see how serious a risk is and how it changes over time. The app is based on the criteria of the PS 982 auditing standard of the Institut der Wirtschaftsprüfer in Deutschland e. V. (Institute of Public Auditors in Germany, IDW) and other applicable regulations that define principles for auditing and optimizing internal control systems for internal and external reporting. Both tools, the project process assistant and the risk management app, are continuously analyzed and optimized with the help of process mining, in our case with Celonis. This is practically an automated process-based control which helps us to verify that our integrated risk and control systems are working. Today, this can no longer be done manually, especially when it comes to regulations, if your company operates globally across multiple locations. Take, for example, the different anti-corruption laws that apply nationally and internationally – it is almost impossible to keep track of them without the support of assistance systems.
At the moment, we are actually discussing how to further optimize the mechanisms of our business partner risk analysis on a process basis. But I would also like to continue what we have started with Fraunhofer IPK and be able to carry out business impact analyses preventively. To do this we need a system similar to a neural network, with which we can connect individual process stages even more closely in order to generate not only a picture of the current situation, but also a kind of preview: what if ...? What happens, for example, if a machine breaks down in a German foundry? What impact would this have on the assembly in India, if the corresponding component is not delivered? What happens if IT systems fail? What critical fallback systems do I have? How far do I get in a worst-case scenario? A business process management system that covers the entire supply chain would be fantastic. It would also help us to respond more effectively to unforeseen events or crises. Figuratively speaking, we would be able to see the iceberg we are heading towards much earlier and steer around it more easily.
Fraunhofer Institute for Production Systems and Design Technology