The Digitalization Journey of DSM
We sat down with two DSM process experts to discuss their experiences using TrendMiner’s advanced analytics platform for Vitamin E production
Here at TrendMiner we pride ourselves on the relationships we build with our customers – from our Coach Program to our yearly user event, TrendLab, we love chatting with them about their experiences, what current challenges they may be facing, or what new features they’d love to see that can make their jobs a little easier.
Recently we had the chance to (virtually) sit down with DSM’s Head of Production Support Vitamins, Marco Tubolino, and DSM Plant Chemist Ulf Conrad, two process experts who are part of the team leading the Production Support Department at their plant in Sisselen, Switzerland, to hear what they had to say about how they use TrendMiner self-service analytics to drive digitalization and improve Vitamin E production.
What has been your company’s journey on the road to digitalization and using TrendMiner self-service analytics? What was it like before and after its adoption?
Marco: Before TrendMiner, we worked with Excel, getting our data right from our historian database. This was quite complicated, and we were also restricted by the amount of data that we could recreate from the database. Nowadays, it is a lot faster and easier with self-service analytics. The application is usable for everyone from the shop floor to management.
Do you find that everyone is using TrendMiner easily with no resistance to change?
Marco: That depends on the functionalities. It’s very easy to make people use it with the basic functions, for instance, for doing trend observing and trend searching for parameters. But if it comes to analyzing the process, then of course, many of the personnel have different backgrounds, so the use depends on their jobs and workflows.
What makes it unique for you?
Ulf: Well to visualize the data in, let’s say, seconds, is one of its biggest values. Otherwise, we just had raw data which require substantial time to find and look at. It’s so easy to visualize your trends and have a smart and fast overview of your processes and to create a graph in TrendMiner.
Has using a self-service analytics tool impacted your daily work?
Ulf: A lot. I’m using it every day right now. So, we first visualize information and then share it with other team members and the management level to explain where some issues are or where there is potential for improvement. I would say this is the most advanced aspect of TrendMiner.
Marco: From my point of view, it’s also being able to use the advanced search algorithms which are normally quite difficult to use. Before you needed data scientists to do this. Now these algorithms are usable for everyone, so that makes TrendMiner unique.
Speaking about data scientists, from that perspective, how can data scientists use TrendMiner?
Marco: Well, the huge advantage of TrendMiner is that you don’t need to be a data scientist. Everybody can use it, but you still need to understand the process. Before, these algorithms were only accessible for data scientists to analyze huge amounts of data with them using programming languages such as R, Python and so on. And now with TrendMiner these functionalities are available for every end user. I think that’s the huge advantage of TrendMiner.
What are the main use cases you find yourselves using it for?
Ulf: We use it to improve our yield of Vitamin E. For example, we used it also for root cause analyses for some issues with the rectification column. We also used it to improve our plant production and operation, and we used it to focus on problems as well.
What would you say is the most useful feature of TrendMiner?
Ulf: Um, the most useful? I’d have to say, I really like the layer function.
Marco: Yes, me too. I like the layer function. It’s super cool.
Would you say that’s your favorite feature?
Marco: That’s difficult. There are a bunch of favorite features.
Marco: Well for Ulf, it’s very helpful because he analyzes the process and finds differences in these processes. Therefore, the layer function is essential, I think, right?
Ulf: Yeah. Exactly. But also, the search function is quite helpful as well to detect some issues and shifts in our process.
Why would you recommend TrendMiner to other industry companies?
Marco: Well, the support is amazing. It’s great. The team is great and very fast and very responsive. And the application itself, I mean, it’s a browser application so it’s very much self-explanatory, at least the basic functionalities, so it’s very easy to use. And it’s super fast. It will bring you to a new stage regarding digitalization.
What advice would you give to other companies wanting to scale with TrendMiner?
Ulf: Use it.
What would you say to companies thinking about getting started with self-service analytics?
Marco: Give it a try. That’s what we did. We started off with trial. And the great thing is that TrendMiner, and that’s also from a customer’s perspective, sent a Customer Success engineer who worked with us for three days to help us and give us an introduction. We got a trial system together and we received great support. And whenever we need help, we just click on the little heart*, and they will help you.
*The in-application support chat button that can be found inside the platform’s Help Center. It opens a direct chat line with TrendMiner Customer Success engineers so you can get your question answered and get back to work as quickly as possible.
What value does TrendMiner bring to your company?
Ulf: Well, it brings us together. That’s the main point here. As I mentioned before, you can share your information, whatever you find in TrendMiner, something that happened to your process. You can share it with all operators on the shop floor and also with people on the management level and bring them together, bring the information together. Therefore, you have a faster way to better results as well.
Marco: Yes, it’s TrendMiner’s speed and it’s collaboration with speed. And it’s easy. It’s just simply easier. You don’t need to analyze the raw data initially, so it speeds up the whole process.
So far what is the best use case that you’ve been able to use TrendMiner for?
Ulf: Well, with the help of TrendMiner, we were able to increase the yield significantly in 2019. I am not sure if it would have been possible without TrendMiner or if we would have come to the same result. But with the help of TrendMiner, the way to the result was way, way faster than without it.
Do you have anything else you’d like to add?
Marco: It’s great. There is great support from the TrendMiner team – they are very, very helpful people. The software is accepted within our company, and it’s being widely used. And all of our personnel are interested in it. Also, TrendMiner is expanding the software’s functionalities. I don’t know if a company could be closer to their clients. We like it.