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.
WIBON
- Publication category
- Impactful algorithms
- Impact assessment
- DPIA, IAMA
- Status
- In use
General information
Theme
Begin date
End date
Contact information
Link to publication website
Responsible use
Goal and impact
The National Digital Infrastructure Inspectorate (RDI) oversees the Information Exchange Above Ground and Underground Networks and Networks Act (WIBON). The RDI monitors compliance with regulations aimed at a careful excavation process to prevent excavation damage during mechanical excavation work.
The CROW 500 guideline is an important starting point for implementing the careful excavation process. This guideline was drawn up by all those involved in the excavation chain and details the conditions for careful excavation.
One component of the CROW 500 is the submission of an excavation notification to the Land Registry via the KLIC system. Without this excavation notification, excavation work may not be started. Information is provided from the Land Registry about the cables and pipelines in the subsurface. This information is called the area information (KLIC). The area information must always be present at the excavation site.
The aim of this project is to develop an algorithm capable of predicting where, when and by whom excavation damage may occur. These predictions will serve as guidelines for WIBON (Wet Informatie-uitwisseling Bovengrondse en Ondergrondse Netten) inspectors, enabling them to be information-driven. By focusing on locations where damage is likely to occur, inspectors can operate more efficiently.
Considerations
The advantage of using an algorithm that calculates the probability of excavation damage is that inspectors can make targeted visits to certain excavation sites. This is because it is not possible to visit all excavation sites. The algorithm can then help to quickly choose between visiting excavation sites. However, a high probability of digging damage does not automatically mean a high risk, because the impact of the damage is not calculated. So inspectors do not just look at probability of damage, but also take other factors into account. Thematic inspections are also carried out. For all excavation reports, the probability of excavation damage is calculated in the same way. A disadvantage is that the model does not take into account the impact of the damage and any other factors influencing the probability of excavation damage that are currently not included in the model.
Human intervention
The outcome of the algorithm is a probability of digging damage. Inspectors can use this for prioritisation, but always decide where to go on inspection. All excavation reports are available to inspectors & besides the algorithm's predictions, other external factors also influence the decision to visit an excavation report. These include practical consideration (location), type of excavation activity, experience and expertise of the inspector and signals from society that may justify an inspection.
Risk management
The quality of the model's predictions of the probability of excavation damage is evaluated annually. The Land Registry delivers the actual excavation damage data once a year, during which the model is re-trained. Furthermore, any legal and social changes are examined and, if necessary, discussed and adjusted in consultation with the users of the model.
Legal basis
Above-ground and underground grids and networks information exchange act
Links to legal bases
Impact assessment
- Data Protection Impact Assessment (DPIA)
- Impact Assessment Mensenrechten en Algoritmes (IAMA)
Operations
Data
The primary source for the model is KLIC reports from the Land Registry (https://www.kadaster.nl/producten/woning/klic-melding?gad_source=1&gclid=EAIaIQobChMI5bTblYibjAMVeqWDBx2cZghmEAAYASAAEgLnDfD_BwE). Cable owners and network operators have to provide information about underground cables and pipes to the Land Registry. The Kadaster aggregates this information. Those carrying out excavation work are obliged to report excavation work to the Land Registry with a KLIC notification. With this notification, performers receive area information about the location where they want to carry out an excavation. This exchange of information can help prevent damage during excavation work. These KLIC notifications are provided to the RDI by the Land Registry and contain details of the excavation notification and excavator used in the model. A second source is the excavation damage figures provided annually from the Land Registry. Finally, additional sources used include: Land use (CBS), soil type (BRO) and tree data (RIVM)
Links to data sources
- Bodemgebruik (CBS): https://www.cbs.nl/nl-nl/dossier/nederland-regionaal/geografische-data/bestand-bodemgebruik,
- Basisregistratie Ondergrond(BRO): https://basisregistratieondergrond.nl/inhoud-bro/registratieobjecten/modellen/bodemkaart-sgm/,
- Bomen in Nederland (RIVM): https://data.overheid.nl/dataset/14394-bomen-in-nederland
Technical design
The algorithm is an XGBoost machine learning algorithm, this algorithm is particularly suitable for structured data in tables. It combines multiple weak models to eventually arrive at a better model where residuals (errors) of the current model are iteratively corrected with training a new model. The algorithm uses excavation notification data (e.g. date, client, excavator, polygon, cables and pipes theme), excavation damage data (e.g. date, location , repair cost, type of net), cables and pipes characteristics, soil data & RIVM's tree map. The target (what will be predicted by the algorithm) is a score indicating the probability of damage per excavation report.
The model works by finding relationships between the features and the target. Based on these calculations, the model determines which excavation reports have an increased probability of excavation damage.
External provider
Similar algorithm descriptions
- The algorithm provides a prediction of the probability of a violation of the Metrology Act for the small weighing equipment category. The algorithm gives a probability of a violation and this is used to design part of the supervision in a risk-based way.Last change on 18th of June 2025, at 14:23 (CET) | Publication Standard 1.0
- Publication category
- Impactful algorithms
- Impact assessment
- IAMA
- Status
- In use
- The algorithm uses sensor data to make a calculation for predictions of slipperiness risks. These predictions are used to determine where to grit (preventively), in consultation with meteorologists. In addition, the algorithm can issue various alarms, for example when there is a chance of wet road sections freezing.Last change on 19th of June 2025, at 12:38 (CET) | Publication Standard 1.0
- Publication category
- Other algorithms
- Impact assessment
- Field not filled in.
- Status
- In use
- The algorithm uses sensor data to make a calculation for predictions of slipperiness risks. These predictions are used to determine where to grit (preventively), in consultation with meteorologists. In addition, the algorithm can issue various alarms, for example when there is a chance of wet road sections freezing.Last change on 10th of July 2024, at 8:18 (CET) | Publication Standard 1.0
- Publication category
- Other algorithms
- Impact assessment
- Field not filled in.
- Status
- In use
- The algorithm calculates which absence reports made by schools can be assigned to which compulsory attendance officer.Last change on 5th of July 2024, at 10:32 (CET) | Publication Standard 1.0
- Publication category
- Other algorithms
- Impact assessment
- Field not filled in.
- Status
- In use
- Automated assessment of the admissibility of an incoming notification based on the submitted file format.Last change on 13th of March 2024, at 8:00 (CET) | Publication Standard 1.0
- Publication category
- Impactful algorithms
- Impact assessment
- Field not filled in.
- Status
- In use