Fokker X Data Science Lab

This series is for and by data professionals. We talk to data professionals, managers and founders at our partners about the data challenges their organisations face. We look back and ahead. Where are the opportunities for data science? And how does Data Science Lab help? You can read about it here. We kick off the series with none other than Wouter van Dis. Wouter van Dis has worked for 29 years at Fokker Technologies, a Dutch aircraft components manufacturer and a source of national pride. For the past 4.5 years, he has been manager of Advanced Analytics.

Hi Wouter, you’ve worked at Fokker for 29 years, including the last 4.5 as manager of Advanced Analytics. How have you seen the organisation change over the years?

I have indeed worked at Fokker for 29 years. It’s a great organisation! There’s a world of difference between when I started at Fokker Aircraft and where we are now. Back when we built aircraft, manufacturing planes was our core business. In recent years, our work has mainly involved manufacturing aircraft components for companies such as Boeing and Airbus. With the arrival of the new owner, logistics and services have become part of our core business for the first time.

Can we go back to the beginning for a moment? How did Fokker get started?

Do you want me to go all the way back to Anthony Fokker? Anthony Fokker was an aviation pioneer. He flew the Spin around a church in Haarlem. That was remarkable at the time (1910). Interestingly, Fokker became the largest aircraft manufacturer at one point. He died in 1939. Although Anthony Fokker was an aviation pioneer, he was not in favour of using aluminium. Even so, aluminium remained the standard for a long time. Aircraft are now made from composite materials.

You’re currently responsible for the data & analytics department. What are the most important changes and steps you’ve made, and what topics do you see emerging in the coming years? When I took on this role 4.5 years ago, I didn’t quite know what I was getting into. I’d always had a thing for data — that’s typical of people at Fokker. Microsoft Office was everywhere, and it still is. We’ve made a start in our department. Some of the data comes from the aircraft themselves, and we need data engineers and data scientists to write code to process and analyse it using Python. We’re making good progress on that now. It’s clear across our organisation that this is the future. Four years ago, things were different: I encountered quite a bit of resistance from our IT organisation. Back then, the focus was more on managing office applications; now it’s shifting towards how we work.

"I want to know which component in my client’s aircraft will fail tomorrow"

The former manager had already said, ‘we need to do more with data and digitalisation so we can work more intelligently’. We can move further towards predictive and prescriptive analytics. I want to know which component in my client’s aircraft will fail tomorrow. That’s a topic I find particularly interesting for the coming years.

What we really need to do is give users as little raw data as possible and provide them with conclusions and assistance to make better decisions instead. How will we do that? We have a number of ideas.

Where are the data science opportunities for Fokker?

I see an opportunity in reliability management. We want to know which components will fail in our customers’ aircraft in the future. Predictive maintenance plays a major role in that. It’s very interesting to us, and we need a lot of IoT data for it. We also see opportunities in logistics, where we need to use control towers to make key decisions. Imagine choosing between three options, with the consequences of each laid out, such as the lowest gross margin or the best customer service. That lets you make a more informed choice. Of course, there’s a lot of intelligence and automation behind this. Reliability management is where my heart lies, but I think it will remain somewhat in the background for the next few years. Logistics is the priority right now. That means bringing in data in a structured way so we can analyse it and make decisions. That’s our focus for the coming year.

How have you found working with Data Science Lab.?

At the start of our collaboration, I realised that our organisation didn’t yet have the knowledge we needed. I needed data professionals with expertise in areas including machine learning. I looked for organisations with a development programme for data professionals. You can’t simply pick up this profession; you have to invest in people’s development. That’s how I found you, and it felt right from the start.

We have an arrangement where the data professionals work for us four days a week and spend one day a week studying and developing their skills at the Lab. That’s how it started, and it works very well for me. Expanding our analytics capacity was a logical step after the acquisition of Pantha Holdings. The team now consists of two data scientists, a data engineer and a BI specialist. After a year at Fokker, they’ll become employees. As well as helping me find, retain and engage data professionals, Data Science Lab. gives me a source of knowledge, experience and expertise to draw on.

What was the biggest lesson you learned as the manager of a data analytics department?

You need to be a hands-on manager to understand what the field involves. Some knowledge of programming is useful. But at some point, you need to step back from the programming itself. In my experience as a manager so far, I’ve learnt to take small steps and, above all, not to try to move too fast or do too much at once. Change takes time.