Interview with Tamas Erkelens, Program Manager of Data Innovation at the Municipality of Amsterdam


By John Connolly

The Municipality of Amsterdam is on a mission to make the city more efficient, transparent and innovative for residents and visitors. Central to this is the creative and open use of data to improve government services and encourage citizens and the private sector to collaborate on social causes.

As part of the Amsterdam Smart City Initiative, the municipality’s open data program has released around 300 datasets and over 1700 files to the general public since 2012. It continues to leverage its own data collections to find new ways of responding to the unique urban challenges it faces as one of the most visited cities in the world. This involves sharing data with partner organisations such as TomTom, Google, Universities, and releasing data in machine readable format. Many organisations talk about the free-flow of data, but Amsterdam has truly embraced the concept, and visitors to data.amsterdam.nl can find information on everything from road traffic data to beehive locations across the city, while developers can reuse Amsterdam open source software on Github.

Tams Erkelens has been Program Manager of Data Innovation at Amsterdam Municipality since 2016 and is a true believer in creating public value through the craft of data analytics. His work focuses on using data to improve citizen services and connecting business, academia, and local government to citizen’s most pressing needs. We catch up with Tamas to hear his thoughts on data in Amsterdam, how data can be used for good and to hear how the city is using open data.

Hi Tamas, what makes the city of Amsterdam a good place to work in data?

A combination of social consciousness, creativity and working with motivated colleagues in the centre of the city. The cliché bureaucrat who finishes at 17:00 does not exist in our Datalab.

What are you currently working on at the City of Amsterdam Mayors Office?

At the moment I am managing a program on data innovation. I try to see my role as kickstarter of data innovation, setting up innovative projects, creating public private partnerships and setting the conditions for our 400 Data Analysts to be the linking pins of our organisation. Sharing data across departments contributes to a more societal impact driven government. In one project we’re combining scraped data, open data and city government information to try and improve the city’s crowdedness issues. Amsterdam’s streets were designed for a different era and we are constantly adjusting to cope with the needs of a modern city. Increasing visitor numbers means the centre is becoming even more crowded and more residents are becoming unhappy with the growing popularity of the city for tourists. Meanwhile, a lot of the data which can help us understand congestion is spread across multiple institutions and isn’t being properly applied to help fix this.

Our task is gather and combine better, real-time data on crowdedness so that we can begin to properly understand where, when and why crowdedness related problems occur. When we understand this better, we can then begin to properly address residents’ concerns. This goes beyond mobility challenges. If we know how many people are where, we can also clean the streets more dynamically instead of a weekly schedule. A modern smart city is a responsive city, in which the city government responds flexibly to civic needs.

What role does big data play in solving urban or social challenges?

Big data can only improve, but never solve urban challenges. Big data implementations always have to be a collaboration between man and machine. Machines can advise what we should do, but humans still have to be able to do what they think is right. This way humans can train models, and humans learn from machines.

In big data discourse you hear a lot that the model is right and that people are stupid. I don’t buy this. People always have reasons not to trust technology, and if you can’t convince them you either need to explain the technology better or you just haven’t offered them enough value.

Big data implementation should always be combined with a change management approach. The failure to acknowledge this in project designs that are too technical is one of the reasons why big data hasn’t delivered its promise yet to help improve urban or social challenges. 

Used correctly, there are multiple ways big data can help tackle social challenges though, for example though recommender systems that find services citizens are entitled to but haven’t requested yet or through big data models of video analytics to identify unsafe situations on the streets. Our open data on street panorama is available for solutions from the market by the way.

There is a lot of media focus on big data and machine learning and their potentially negative aspects, while positive impacts are often underdiscussed. Do you believe big data will have a positive impact on society?

I think it is up to us data analytics people whether big data can have a positive impact on society. For me the motivation to contribute to it is exactly the reason I am working at the city government and why I have founded the Data Mission. There are a lot of data scientists confronted with ethical dilemmas in their work. As long as we, in corporate or government environments, recognize these dilemmas, face them, make them clear to non-tech people and decide on it together, we can use big data for good societal causes.

What are the biggest challenges the city of Amsterdam faces, with regard to using data to improve citizen services?

A big challenge is how to transform governmental organisation for a new era. Operational staff who have been dedicated and creative in trying to serve people in roles like law enforcement and social work will be confronted with a more standardised way of working in most cases and require more specialised work for exceptions. On the other side, policy makers are faced with the challenge of making their policies more agile and test based. 

This will create a smaller divide between policy makers and implementation. Both policy makers and operational staff need to be served by a flexible and well managed core of databases and IT products and services built on top of it.

The ‘Open Data’ movement has gained traction in recent years, the idea being that a large amount of government and municipal data should be made available to the public. Do you support this, and does it have a role to play in Amsterdam’s development as a centre of innovation?

Yes it does. That said, the city of Amsterdam does not see Open Data as an end-goal, but more as the final step of a good information process across the city. A lot of the barriers to releasing more municipal data are data management challenges rather than the philosophy of open data itself.

Overall, I see the open data debate changing from the dichotomy of open and closed data to a more flexible layered model, with data kept internally for privacy reasons, being fully open to the public, and shared with trusted third parties, allowing the city to ultimately choose under which conditions we share data. For example, partner organisations only want to share data with the city government if it contributes to a cleaner city but are not looking to share it with their competitors, a flexible model allows this to happen. At the moment we are creating a data sharing marketplace, to overcome the many bilateral data sharing agreements.

Therefore, the current focus in Amsterdam is collaborating with these trusted third parties to create applications that can reach a large audience.

Tamas Erkelens will be speaking at Whitehall Media’s Big Data Analytics Europe conference on 8 March 2018 at the Postillion Convention Centre, Amsterdam.

More information about the event here:

www.whitehallmedia.co.uk/bdaeurope