The Role of Data in Car Manufacturing
Hi, I am Johannes — Data Scientist at Porsche Digital.
A lot has happened since the first assembling line was created in 1913 by Henry Ford. It was the breakthrough of the mass-production of automobiles and reduced the production time for one car from 12 hours to under three.
Mass-production has evolved ever since and nowadays it strongly relies on data and automation. In modern factories, the number of robots exceeds humans as specialized computers put together the car on the assembly line. Even though this process has been perfected throughout the years, issues still occur. And that’s where my job starts.
Since January 2020, I work as a Data Scientist at Porsche Digital in Berlin and use technologies like machine learning and AI to improve car manufacturing. Coming from physics, I am able to fill the gap between manufacturing and data science. After working as a software developer for a large mechanical engineering company, I later joined a startup to work with software and data in the context of Industry 4.0. Now, I and my team try to bring the Porsche assembly line to the next level with data.
You can’t build a Porsche without data
As stated above, car manufacturing is in constant change and new technologies evolve at rapid speed. Porsche wants to stay ahead of this change and create a production that is smart, lean and environmentally-conscious. At Porsche Digital, we explore future technologies such as AI, blockchain and IoT in order to approach business challenges from a new angle and make business processes within Porsche — but also other companies — more effective and efficient.
As a Data Scientist, I try to support this shift in production by finding solutions for issues that occur along the way with data. This is actually very special because we don’t work on “possible problems”. Instead, we are searching for industry solutions to solve real problems and to improve corporate processes.
So, how do we do this? With every new project, we kick-off our solid process ⎼ and it works quite well.
- At first, we discuss the problem and how data and software could help to solve it. What is our starting position? Is the necessary data already available or do we need to collect it? Is the customer looking for a web-based solution or a desktop application?
- After understanding the problem and gathering some data, we develop a mathematical solution in a proof of concept phase. This involves experimenting with different models and the data at hand.
- We bring the best solution into production by using the expertise of our skilled Software Engineers and DevOps within Porsche Digital
We work with agile software development and short iteration cycles. Therefore, we continuously evaluate if our solution really solves the problem or if we’re heading in the wrong direction. Due to the sprints, we get quick results, quick feedback and work closely together with the client, which brings us much faster to a better solution.
In comparison: When working with classic waterfall methods, you often develop in the wrong direction for a long time and end up with a solution that misses the actual problem because you never put it to the test in between.
Beyond technical skills
What I love about the work at Porsche Digital is that we, as Data Scientists, are involved in the whole lifecycle of a project. Often, in the corporate world, you are only responsible for a single aspect like coding or project management. Here, it’s not “either-or”, it’s “and”. Starting from identifying the problem in the production to evaluating a mathematical solution, software development, communication, and re-evaluation up to the final presentation of the solution — in my role, I am part of all these. I love that I can bring in more than just my mathematical and technical skills, I can also develop my communications, management and presentation skills. Meetings with different stakeholders are a big part of my job and I’ve learned to ask the right question in order to understand the issues as a whole — and thus ultimately to develop the right solution.
At Porsche Digital, we do this in a technical yet creative way — we try, test and learn along the way. Looking for new digital solutions for existing challenges requires an open mind and a progressive inventive spirit, which is more than present here. I think that’s one reason why my job is so much fun: The close communication and the agile working atmosphere builds up a strong team spirit. Even though I just started a few months ago, I already feel that I am a valuable part of it.
Do you want to learn more about Data Science at Porsche Digital? Read about my colleagues Murali and Sophie, who shared their stories recently.
Johannes Märkle is a Data Scientist at Porsche Digital in Berlin.
About this publication: Where innovation meets tradition. There’s more to Porsche than sports cars — we’re tackling new challenges, develop digital products and think digital with a focus on the customer. On our Medium blog, we tell these stories. It’s about our #nextvisions, smart technologies and the people that drive our digital journey. Please follow us on Twitter (Porsche Digital, Next Visions), Instagram (Porsche Digital, Next Visions, Porsche Newsroom) and LinkedIn (Porsche AG, Porsche Digital) for more.
