Data scientist Gabriele Compostella uses AI and human expertise to retrieve and combine relevant data using big data. Part 6 of a series about IT jobs at Volkswagen.
There is as much data out there as there is sand at the seashore. Yet the rapidly growing amount of data is about as profitable as a pile of sand if it only lies around and isn’t utilized. But in order to gain knowledge from data that can be used to make forecasts and optimize processes, you need artificial intelligence and a data scientist who manages the algorithms.
Big data as a puzzle
Gabriele Compostella from the Volkswagen Data:Lab Munich is one such data scientist. He works at the Data:Lab of the Volkswagen Group’s IT department. “I love solving puzzles,” the 40-year-old says with a smile. And not only at work: Compostella has around 130 different board games at home. That makes sense, because data science is all about trying out new scenarios over and over again in the team, and analyzing those scenarios scientifically – much like working together to solve puzzles for fun. His education also makes Compostella perfect for taking on the responsibilities of a data scientist: “I always wanted to be a scientist, even as a child,” he says. He succeeded: university studies in Padua, a doctorate from Trento, work at the Fermilab particle accelerator near Chicago in the U.S., a postdoc at the Max Planck Institute for Physics in Munich, work at the CERN particle accelerator in Switzerland. “I essentially made big data with particles, but I wanted to work more with data that have a more direct bearing on the world human beings live in. My decision to go into industry with data science was an obvious choice,” he says.
Volkswagen at CEBIT 2018
Volkswagen is a digital company that drives modern information technology forward. During the CEBIT in Hanover, we present a series of portraits of people in the Group with exciting IT jobs. At CEBIT (June 12-15), the Volkswagen Group is located in the Future Mobility Hall (Hall 25), offering a forum for interested parties and experts alike – with stimulating presentations and first-class exhibits as well as interesting panel discussions and talks. The range of topics and highlights is considerable and includes not only new forms of digital automotive design, quantum computing and test projects with Blockchain, but also applied artificial intelligence in the company and data-supported traffic optimization in European metropolises. There is also a world premiere at the exhibition stand.
Analyzing huge amounts of data and bringing them together in a meaningful way is not something any one individual is able to do on their own. Therefore, at the Data:Lab, data scientists like Compostella work closely together with artificial intelligence experts. “A person cannot put together a puzzle with hundreds of thousands of pieces,” the 40-year-old says. “Self-learning systems, which we specifically develop for the purpose, take over that work for us.” The teams feed these algorithms with data, have them analyzed, combined and use them to draw conclusions – and make corrections where there are errors. That’s called Supervised Machine Learning.
Seeing the big picture
So what is all this for? For Compostella and his colleagues it’s not about personal data, but rather the huge amount of information that the Volkswagen Group itself with its complex corporate processes produces on a daily basis. This includes, for example, logistics and flow of goods, key financial figures, requirements and usage down to the smallest levels. “Just like with a puzzle, you need a systematic approach to be able to recognize the big picture behind it all,” Compostella explains. The analysis of data with advanced statistical methods is therefore just the first step.
“Data can help answer questions correctly and in a fact-based way,” he says. And that means questions predicting the future. The technical term for that is “predictive analysis”. Compostella gives an example: How will market demand develop into an equipment line, and how will the supply situation evolve? Which components and parts will have to be where and when? Can trends be extrapolated?
“Our company has always had this information and data, but it’s only been in the last few years that we had the technological capability to put different big data sources side by side,” says Compostella. The specialists at the Data:Lab are also experimenting with data analysis of traffic flows. In a joint effort with cities, they want to test how to optimize urban traffic using intelligent data analysis – a puzzle in motion, so to speak. It’s an exciting challenge for the team: “The more complex the puzzle,” says Compostella, “the more fun it is for us!”