Please note: The algorithm descriptions in English have been automatically translated. Errors may have been introduced in this process. For the original descriptions, go to the Dutch version of the Algorithm Register.

Risk analysis inland navigation

The Environment and Transport Inspectorate (ILT) has created a model to determine which inland vessels are more likely to be infringed upon. This model helps the inspector choose which vessels to inspect or not.

Last change on 18th of December 2025, at 13:59 (CET) | Publication Standard 1.0
Publication category
Impactful algorithms
Impact assessment
DPIA, AIIA
Status
In use

General information

Theme

  • Traffic
  • Nature and Environment
  • Economy

Begin date

2024-04

Contact information

https://www.ilent.nl/service/privacy/privacyverklaring-artificial-intelligence-en-machine-learning

Link to publication website

https://www.ilent.nl/onderwerpen/scheepvaart-algemeen/toezicht-binnenwateren

Responsible use

Goal and impact

Safety, environment and fair market conditions are paramount in the supervision of the inland navigation sector. Violations can give unfair competitive advantage. Although regulators work together, there are not enough people and resources to inspect all vessels. The risk model helps inspectors make smart choices. They try to spare skippers who comply nicely with the rules. Ships with a higher risk of violations are instead inspected. The aim is a reduced inspection burden for ships that are in order, fairer competition in the sector by reducing non-compliers, and increasing safety and reducing environmental impact in this sector.

Considerations

More insight into the entire sector is needed to achieve more effective supervision. In the first place to support the inspector's expertise in making choices, but also to play a directing role in inland navigation supervision. There are very few inspectors and there are very many inland navigation vessels in the Netherlands, some of which have not been inspected before. To be able to estimate risks in advance, the ILT opted for a predictive model. The ILT has included information-driven work in its strategy, with the aim of better identifying and addressing societal risks.

Human intervention

The inspector decides whether or not to inspect a ship. In doing so, he/she takes into account the work instruction, the risk score, knowledge and experience and his/her own observation. The risk score is an aid and not mandatory advice.

Risk management

Changes in the industry and supervision are discussed with a user group. If necessary, the model is adjusted accordingly. Changes in input data are monitored and the reliability of the model is regularly tested. If necessary, measures are taken and the model is re-trained with adjusted characteristics.

Legal basis

The ILT checks whether ships comply with laws such as the Inland Shipping Act, the Transport of Dangerous Goods Act, the Shipping Traffic Act, the Working Hours Act and the Environmental Management Act. Additional regulations and decrees also apply.

The ILT processes personal data because this is necessary to carry out its statutory duties. This is permitted under Article 6(1)(e) of the General Data Protection Regulation (GDPR).

Elaboration on impact assessments

No IAMA test was conducted. Instead, an AIIA test was conducted, covering similar questions and topics.

Impact assessment

  • Data Protection Impact Assessment (DPIA): https://www.autoriteitpersoonsgegevens.nl/themas/basis-avg/praktisch-avg/data-protection-impact-assessment-dpia
  • AI Impact Assessment (AIIA): https://www.rijksoverheid.nl/documenten/rapporten/2022/11/30/ai-impact-assessment-ministerie-van-infrastructuur-en-waterstaat

Operations

Data

The model was created using data collected during the monitoring of inland navigation. This data comes from the ILT and from organisations the ILT cooperates with: the port authorities of Amsterdam and Rotterdam, the Dutch police and Rijkswaterstaat. It includes information such as inspection data, permits, vessel characteristics and company information. Together, these data give the most complete picture of the sector.

Links to data sources

Inspection View: https://www.ilent.nl/onderwerpen/inspectieview-voor-inspectietaken

Technical design

The inland navigation risk model uses computational rules to predict whether a vessel is in breach. Such a model is also called an algorithm. In creating the model, a form of artificial intelligence was used: machine learning. The computer has learned from a lot of data from the past. This makes it clear which characteristics of a ship or company say something about the likelihood of a violation. So the calculation rules are not devised by an inspector, but are derived from the data and the type of algorithm. The model works for all inland vessels, even if they have never been inspected before.

Similar algorithm descriptions

  • Whereas ordinary traffic lights register by detection loops (sensors in the road) that traffic is passing over, 'smart' traffic lights additionally register via apps what kind and how much traffic is coming. This allows traffic lights to see traffic "arriving" earlier and thus determine how long the traffic light is on green.

    Last change on 2nd of April 2025, at 14:10 (CET) | Publication Standard 1.0
    Publication category
    Impactful algorithms
    Impact assessment
    Field not filled in.
    Status
    In use
  • Whereas ordinary traffic lights register by detection loops (sensors in the road) that traffic is passing over, 'smart' traffic lights additionally register via apps what kind and how much traffic is coming. This allows traffic lights to see traffic "arriving" earlier and thus determine how long the traffic light is on green.

    Last change on 14th of March 2025, at 9:08 (CET) | Publication Standard 1.0
    Publication category
    Impactful algorithms
    Impact assessment
    Field not filled in.
    Status
    In use
  • Based on measurements, sewage discharge is controlled. The algorithm determines whether a valve for the transit is open or closed. This makes it possible to control where this water goes during periods of high rainfall.

    Last change on 5th of January 2024, at 14:22 (CET) | Publication Standard 1.0
    Publication category
    Other algorithms
    Impact assessment
    Field not filled in.
    Status
    In use
  • Detect risks in customs declarations regarding compliance with import, export procedures of chemicals covered by the PIC Regulation.

    Last change on 2nd of April 2025, at 12:35 (CET) | Publication Standard 1.0
    Publication category
    Impactful algorithms
    Impact assessment
    Field not filled in.
    Status
    In use
  • The Halt-Signaleringsinstrument (Halt-SI) is a scientifically based questionnaire (risk assessment instrument) that is used with all young people during a Halt intervention. The Halt-SI is used to determine the risk of repetition of criminal offences, to design the intervention appropriately and to assess whether referral to assistance is necessary.

    Last change on 28th of November 2025, at 12:28 (CET) | Publication Standard 1.0
    Publication category
    Impactful algorithms
    Impact assessment
    DPIA
    Status
    In use