Conquering the world of data
Topics in this Article
- Data scientists - in greater demand than ever
- A chemist and data scientist
- The goal: Understanding and linking data
- Making the right connection is key
- Accelerating product development - with artificial intelligence
- Using artificial intelligence at LANXESS
- Tracking delivery - in real time
- Protecting the climate with data science
- The team makes the difference
Data rule the world
Data-driven corporations like Google, Apple or Facebook are the most successful in the world. That is hardly surprising as a study by the management consultancy Capgemini suggests: Companies that make targeted use of data achieve 22 percent more profit and 70 percent more revenue (per employee) than their competitors. Data is therefore becoming a success factor for companies. But what data does a company need and how should it be used?
Data scientists - in greater demand than ever
LANXESS also generates data. In research and development, in the production facilities, in sales or in the logistics chain. But data alone is of no value. Why? Data must be read, understood and linked. And that is why we, as a specialty chemicals company, also need data scientists.
Data scientists can read data. They gain insights from it and make it usable for others. As a company, we can use it to make production processes and research more effective, shorten test phases, optimize logistics routes and even make a contribution to climate protection.
Within a short time, a data team with more than 20 colleagues was created at LANXESS. It includes data scientists, data engineers and front-end developers. The team members are young, have international backgrounds and work interdisciplinary. Our data scientists are being deployed where they can use real data to solve concrete challenges from our operational business. A mobile task force with data expertise!
Waldemar Czaplik had already dealt with programmatic data processing when he was working on his doctorate in chemistry. “But back then it was more of a hobby,” he admits.
In his position as head of the global production processes in our Inorganic Pigments business unit, data suddenly became important for the 40-year-old again. In order to gain more detailed insights into the processes and optimize them, he spent hours compiling and processing data from various systems. “The work was terribly inefficient. I had far too little time besides my actual job to make proper us of the data,” Czaplik recalls. And he was not the only one to face this problem.
The only way to remedy the situation was to get to grips with the data. And so, at the beginning of 2018, a team was formed, initially consisting of a handful of data scientists. Waldemar Czaplik seized the opportunity to make data his profession: He took over the leadership of the team, which quickly grew in size.
The primary goal of our data scientists: to process the data in such a way that as many employees as possible can understand it and work with it. “This enables us to link data, even across multiple business units, and create a high level of transparency across the entire value chain. We can get a lot out of it if we keep an eye on the big picture,” says Czaplik.
“If we increase the output of a plant by five percent with the help of data, this is of course of little use to us if we cannot purchase the corresponding raw materials or bring the additional products to market. The entire value chain has to be connected.”
Samreen Hassan ensures the right connection. The 32-year-old works on many projects as an intermediary between the data team, the business units and third-party providers.
She is responsible for connecting different systems and thus for integrating data. And that is where she is a real expert. The computer scientist worked as a developer for large software corporations before joining LANXESS in the summer of 2019. “I think it’s great that I can use my knowledge to solve real problems here. That’s where I can put my problem-solving skills to good use,” she says.
Samreen Hassan’s passion for computers and programming was sparked at school. She wrote her first computer programs when she was just 11 years old. “Back then, I saved them on floppy disks,” she says, laughing. What she likes most about her job is the variety.
“Every week looks different: sometimes we have an intensive exchange with the business units to understand the problem, then we look at the facilities in our production plant. The following week I may prefer to work in quiet when I collect, process and analyze the data."
The project teams are therefore always staffed interdisciplinary. Lynn Ferres heads one of them. Ferres, who holds a doctorate in chemistry, has always been fascinated by numbers and formulas. At university, the 30-year-old preferred to explain her observations using mathematical methods rather than deriving them experimentally: “I’m more at home in theoretical and physical chemistry. I also did my doctorate on a topic that focused on measured data.”
At our company, she is currently helping the High Performance Materials (HPM) business unit to solve a very specific challenge. HPM develops plastics that replace metal parts in cars. They therefore have to be lightweight, but still very strong.
“The glass fibers that are incorporated into the plastic play a major role here. Only if these have the right coating do they also bond ideally with the high-performance plastic,” explains Ferres. The whole process is very complex as more than 500 raw materials can be considered for the coating. Countless tests and examinations would be necessary to find the ideal combination of raw materials.
And this is where Lynn Ferres comes into action. Together with her team, she developed an artificial intelligence that can make predictions about coating formulations based on existing data with regard to measured properties of the glass-fiber-reinforced plastics.
“That saves the application engineers a lot of trial and error,” she says. The experts can narrow things down further because they have the empirical data to determine whether a formulation can also be put into practice.
“Combining expert know-how with data science and artificial intelligence have led us to the decisive breakthrough here.”
The first successful projects are whetting the appetite for more - not only among data scientists, but also among colleagues in the business units.
“I think it’s great that more and more colleagues are coming up with their own ideas for projects now that they can see what’s possible.”
Oliver Tebeck has been at LANXESS for around three years. He heads the team that is responsible for data projects in the area of purchasing and procurement. He “fell in love” with data during his studies, he says. Yet the 31-year-old did not originally intend to go into data science. But during his master’s degree in statistics, he got hooked.
At LANXESS, he and his project team scored a big hit with their first project. For our Polymer Additives (PLA) business unit, he developed a tool that allows the team to track where its bromine tanks are - in real time.
Bromine is the business unit’s most important raw material and has to be shipped around the world. “We have linked a large amount of data from different sources and made it visible and analyzable in our tool,” he explains. This enables PLA to control the tanks more easily and in a more targeted manner, thus optimizing transport routes and shortening delivery times.
For example, an alarm is triggered if the tanks remain in one place for too long - the business unit can then intervene directly. This improves the utilization of the tank fleet and the customer is always up-to-date on the delivery status. “The bottom line is that we save time and money,” says Tebeck.
The tool has certainly proven its worth - but especially at the end of March 2021, when a container ship got stuck in the Suez Canal and blocked this important transport route for several days.
“It was great to see the impact of the project. We knew immediately where the bromine tanks were and the business unit was able to take the appropriate measures quickly.”
Waldemar Czaplik believes that data scientists can achieve even much more. He is certain: “Without data science, we will not be able to solve the world’s big questions. Or at least not at the speed we need.” Climate protection is at the top of the list.
“With our simulations or mathematical models, we often help to make processes more efficient, reduce energy consumption and thus emissions.”
Waldemar Czaplik is proud of his team: “Everyone here has their own superpower. But the most important thing is that everyone is a real team player!” Samreen Hassan also agrees.
“Everyone here is very open and has a positive attitude. I'm encouraged to contribute my own ideas - and they’re often taken up and implemented.”
New insights are shared directly within the data team. “We regularly present our projects to each other. That way, we learn a lot from each other. And it’s not unusual to get the one or other suggestion for your own project,” says Lynn Ferres. “I felt at home in the data scientist team right away. I am always motivated by the fact that we can really make a big difference here,” adds Oliver Tebeck.
