Using space more efficiently in the port of Amsterdam with our smart tool for tracking berth use
Port of Amsterdam
AI agency in Amsterdam
From Westerdoksdijk, we start with the decision you need to improve, then build and manage the solution, as we did for Port of Amsterdam.
Together with Port of Amsterdam, we developed a tool to measure berth usage
Port of Amsterdam manages and develops Amsterdam’s port, where space is scarce. Together, we used Databricks, real-time data and Power BI to develop a tool that measures historical berth usage, so policymakers can make decisions based on figures rather than gut feeling alone.
Where in our organisation are we still making decisions based on experience when the data is already there?
The data is often already there: ship movements, reports from public spaces, audience numbers for each performance. Yet someone is still making decisions based on experience. Together, we identify one decision, test a model against your own historical data and only then build on it.
From the first question to a solution in operation, for organisations across the city
Data to underpin a policy decision
For Port of Amsterdam, a tool we developed together measures historical berth usage, providing a basis for decisions by ship type.
A path for growth across the organisation
At Nationale Opera & Ballet, we assessed data maturity and developed a growth plan that links data to full auditoriums and a broader audience.
On a platform that fits
We build on Databricks or Microsoft Fabric. As a partner of both, we choose based on your IT landscape, without a preference either way.
Monitored after launch
Data changes constantly. That’s why, after delivery, we monitor whether a model remains accurate and agree on who will adjust it.
Every project starts with the decision, not the model
-
Choose one decision
Which decision needs to improve, who makes it now and what data do they use? That’s where the project starts.
-
Test against your own historical data
An initial version runs on your own data, so you can see whether AI adds value.
-
Live and managed
The application goes into the system you already use, with monitoring and a designated owner.
Is a dashboard enough for our needs, or do we need an AI consultancy?
A good fit if
There’s a recurring decision currently made based on experience, the relevant data already exists, and someone in your organisation is willing to take ownership of the application.
Not a good fit if
You first want to understand where your organisation stands with data. A data maturity assessment is a better fit in that case, as it was for Nationale Opera & Ballet. Or a dashboard may be enough, and we’ll tell you if it is.
What determines the scope
How many source systems need to be connected, the quality of the data, whether the application needs to work in real time, and whether your team or ours will manage it.
Will we be tied to one platform afterwards?
We assess Databricks and Microsoft Fabric against your organisation’s needs and existing IT landscape. If you already use Microsoft, Fabric is often a logical choice; we’ll make the decision together.
An assessment, a first application or a specialist on your team: where do you start?
Start with a maturity assessment
Choose this if you’re not sure where your organisation stands. The assessment gives you a report with the next step for each area, as it did for Nationale Opera & Ballet.
Build one application
Choose this if the decision you need to make and the data are already clear. Our advice: pick a recurring decision so you can quickly see whether the model helps.
Bring a specialist into your team
Choose this if you’re building it yourself but lack capacity. Through secondment, one of our AI engineers or data engineers works as part of your team.
Our office is on Westerdoksdijk, close to the organisations we build for
Data Science Lab has worked with data and AI since 2016 and is a Databricks and Microsoft partner. In Amsterdam, we’ve worked for Port of Amsterdam and Nationale Opera & Ballet, among others. Drop by, or we’ll come to you.
At Westerdoksdijk, the first conversation starts with one decision
Stef Stoelwinder
Business DevelopmentStef is your first point of contact. In our first conversation, we’ll choose one decision that needs to be made better, either at our office on Westerdoksdijk or online.
E-mail info@datasciencelab.nl
Phone +31 (0)20 3636 724
Which recurring decision would you like help with from an AI consultancy?
Describe the decision and the data you already have. We’ll reply by email or arrange an initial conversation.
Message sent
Thanks for your message. We’ll be in touch as soon as possible.
Frequently asked questions
An AI agency builds applications that support a decision or task in your organisation: a predictive model, an assistant based on language models or an agent that handles a process. Data Science Lab is an AI company in Amsterdam and has been here since 2016.
We start with what you want to improve, not the technology. You can read about the types of applications we build on our AI applications page.
Our office is at Westerdoksdijk 599/601 in Amsterdam. You can call, email or drop by. We can also meet at your office or online.
We work with organisations across the region and beyond. In Amsterdam itself, we’ve worked with Port of Amsterdam and Nationale Opera & Ballet, among others.
There’s an overlap. AI consultancy helps you decide which application will deliver value, what it requires and in what order to proceed. An AI agency also builds the application, connects it to your systems and manages it. We do both, often within the same project.
You can read how we prepare those decisions on our AI consultancy and AI roadmap pages.
Three types come up most often: generative AI that supports employees with knowledge work, predictive models that help organisations look ahead, and AI agents that take over steps in a process.
If your question is mainly about predictive models built on your own data, you can also read about data science.
It often starts with insight. For Port of Amsterdam, we made historical berth usage accessible so policymakers can decide whether to increase or decrease the number of berths for certain types of vessels. We only add a model if looking ahead brings something useful.
Episode 6 of our podcast De Databar covers the opportunities and challenges of data in the Port of Amsterdam. You can read more about shipping and AIS data on our AI in ports and maritime page.
The AI Act is being introduced in phases. According to the European Commission, the prohibited practices have applied since 2 February 2025, and following the AI omnibus, the requirements for high-risk applications listed in Annex III will apply from 2 December 2027. If you work for or with Amsterdam City Council, it publishes descriptions of its algorithms in the government’s Algorithm Register.
Your legal counsel or compliance department should interpret the legal requirements. We can help you identify your applications, classify their risks and set up governance; see AI Act.
Agree on that before it goes live: who owns it, who gets an alert when the input changes and who approves a new version. Data changes constantly, so a model that works well now may become less accurate later.
Your own team can manage it, or we can do so through management and continuous improvement. You can read how model monitoring and retraining work on our MLOps page.
If you mainly want to know what happened last month, a Power BI dashboard is enough. The same applies if no one in your organisation is willing to own the application. We’ll tell you that, too.
If you’re not sure where you stand, start with a data maturity assessment. For Nationale Opera & Ballet, this led to a plan for growth. “Data Science Lab’s expertise immediately gave us the confidence to take a broad approach to the project.” Leon Backbier, Head of I&I
Ports, data science or all applications: what will you read next?
- AI applicationsFrom generative AI to predictive models: what we build and how we keep it working in practice.
- AI in ports and maritimeHow AIS data and port reports become useful for planners and policymakers.
- Machine learningPredictive models built on your own data, from prototype to production and management.