Automated enforcement Environmental zones

Municipality of Amsterdam

The environmental zones for cars, trucks, vans, taxis, buses and mopeds and motorcycles in the municipality of Amsterdam have been set up to keep out the most polluting vehicles.

Last change on 4th of December 2024, at 9:43 (CET) | Publication Standard 1.0

• Publication category: Impactful algorithms
• Impact assessment: Field not filled in.
• Status: In use

General information

Theme

• Economy
• Space and Infrastructure

Begin date

Field not filled in.

Contact information

[e-mail]

Link to publication website

https://www.amsterdam.nl/verkeer-vervoer/milieuzone-amsterdam/

Responsible use

Goal and impact

Automated enforcement Environmental zones, heavy traffic zone, clipping Munt and Sloterweg, other similar entry bans (ban on mopeds and light-mopeds Cuyperspassage and Rijksmuseum, busable Twiske, Montelbaansbrug, temporary bridge Berlagebrug, enforcement of the ban on mopeds and light-mopeds on the cycle path and the use of license plates to calculate travel times in the centre of Amsterdam.

An environmental zone is an area in a city where certain means of transport are not allowed to drive. Residents and visitors to Amsterdam are entitled to clean air and a liveable climate. The air that visitors and residents currently breathe is not clean enough and this is largely due to city traffic. The environmental zones for cars, trucks, vans, taxis, buses and mopeds and motorcycles in the municipality of Amsterdam have been set up to keep out the most polluting vehicles. By doing so, the municipality of Amsterdam aims to improve air quality in the city.

Other entry or transit bans have been introduced by the municipality to improve road safety and quality of life, for example by preventing cut-through traffic or congestion.

There is also automated enforcement of the 7.5-tonne zone to protect bridges and quay walls in the historic centre.

The aim is to enforce environmental zones for passenger cars, trucks, vans, taxis, buses and mopeds and motorcycles in the municipality of Amsterdam established by traffic orders, so that the most polluting vehicles are kept out.

The purpose enforcing the heavy traffic zone is to protect bridges and quay walls in the old centre.

For the other zones, the aim is to enforce the underlying traffic orders of the municipality of Amsterdam.

Within Amsterdam municipality's environmental zones, ANPR (automatic number plate recognition) cameras with algorithms are used to read license plates of passing vehicles. The license plate is then linked to information, such as vehicle information from the RDW (Dutch Road Transport Authority), to a date, time and location.

Environmental zone enforcement is supported by several other rule-based algorithms. All these algorithms contribute to a decision with legal effect in case of violation of the environmental and other zones (penalty decision).

Considerations

The impact for the use of the algorithm is proportionate to the goal, which is the right of Amsterdam residents to clean air, a liveable climate, traffic safety, an accessible city and a pleasant living environment. Environmental zones keep the most polluting vehicles out of the city, thereby improving the city's air quality. In the other zones, in particular, vehicles are barred that cause relatively high nuisance, affect road safety, reduce accessibility, e.g. due to their weight damaging bridges and quay walls that have to be closed off as a result, etc., thus contributing to reduced liveability.

Human intervention

Sampling by special investigating officers (boas) is an integral part of this automated decision-making process. A spot check is the manual assessment of whether the registration number of the vehicle in the violation photo matches the registration number as identified by the recognition software. This is done by a BOA. If an objection has been lodged, a violation photo is always assessed again (manually) by a BOA. In addition, those involved can view the violation photo themselves via https//Mijn.Amsterdam.nl and DigiD.

Risk management

There are two types of risk namely:

1. The risk is that breaches are wrongly identified.

2. Another risk is that violations are missed.

These risks are further mitigated as follows: License plate recognition algorithm: minimal risk. If a registration number is misread and the misread registration number turns out to exist and belongs to a vehicle that does not have access to the environmental zone, the wrong person will be fined. This happens very rarely because the recognition software meets the tough requirements mentioned above and a check is made (randomly) by a BOA, and the chances of the incorrectly identified vehicle not having access to the environmental zone are small. In the single case that this does happen, an objection by the owner of the wrongly fined vehicle against the fine will of course be upheld. If the registration number of a vehicle that does not have access to the environmental zone is misread by the camera, the owner is wrongly not fined. This rarely happens. Other than by measures related to the reliability requirements for the recognition software in the cameras, this risk cannot be mitigated. Reliability estimation: this algorithm mitigates the above risks. Vehicle recognition algorithm: no risks because this algorithm does not lead to a fine. Decision access to a zone: no risk subject to a correctly read license plate (see algorithms 1 and 2).

Legal basis

- Preventing or limiting nuisance, annoyance or damage caused by traffic as well as the consequences for the environment referred to in the Environmental Management Act (art. 2 subsection 2 sub a WvW);
- Promoting efficient or economical use of energy (art. 2 subsection 3 under a of the Act);
- Traffic orders issued by the municipality of Amsterdam to promote road safety and quality of life based on the Wegenverkeerswet 1994 (Road Traffic Act 1994) and in the further regulations based on that Act, such as the Reglement verkeersregels en verkeerstekens 1990 (RVV 1990) and the Besluit administratieve bepalingen inzake het wegverkeer (BABW) (Road Traffic Administrative Provisions Decree).
- An offence can be established "in an automated manner" on the basis of Article 3(2) of the Administrative Enforcement of Traffic Regulations Act ("Mulder Act").

Operations

Data

1. licence plate recognition algorithm.

When a vehicle passes by, an ANPR camera takes a picture and in that picture a registration number is searched based on shape (plate finder). Image recognition techniques are then used to search for characters. For this, the algorithm uses dots and lines. In this way, an image of numbers and letters is created. This algorithm uses a reference database filled mainly with Dutch license plates, which is the best way to recognise them. The majority of vehicles (around 97%) are vehicles with Dutch license plates.

2. An algorithm that assesses the reliability of license plate recognition.

The system assesses the probability of the license plate being read correctly. The result of recognising up to 5% of the license plates of a vehicle with a Dutch license plate may be categorised as unreliable. This data is then not processed.

3. Vehicle recognition algorithm

This is an algorithm that is only used if it has been established by Algorithm 1 that a non-Dutch license plate exists. Algorithm 3 uses image analysis to determine whether the passing vehicle is a passenger vehicle, van, truck or bus. The numbers of vehicles by category, location and time of passing, without registration number, are used for policy purposes, such as analyses for traffic research.

4. Decision algorithm access to a zone

To determine whether a passing vehicle has access to the zone as defined in the traffic decision, the relevant vehicle data belonging to the detected license plate are automatically requested from the RDW. If, based on this, it is found that the right of access is present, the registration number is irrevocably removed from the passage control system. If the right is not present, the system automatically checks whether an exemption is known to be linked to the licence plate. If so, the badge is irrevocably removed from the passage control system. All these steps follow the traffic decision one-to-one. If no right and no exemption are found, a fine file is compiled automatically and forwarded to the CJIB, which sends the decision. A BOA randomly checks whether the detected registration number matches the one in the photo and whether there are reasons to dismiss a fine.

Technical design

There are links with

- RDW for retrieving vehicle data, exemptions issued by RDW and for owner data in the event of an offence

- the BRP (Basic Registration of Personal Data) for private vehicles and with the Trade Register (Chamber of Commerce for commercial vehicles) to check whether the address details of the owner of a vehicle in violation are correct.

- NPR (National Parking Register) for exemptions in car-free areas.

- DigiD for applying for an exemption for environmental zones and viewing photos of a violation,

- E-recognition, idem

- TNO for linking the emission factor to passed vehicles for the environmental zones.

External provider

Connection Systems