Slow traffic monitoring system Amsterdam

Municipality of Amsterdam

The various algorithms and calculation rules within LVMA work together so that information from video images can be converted to interpretable census data as decision-supporting information for public space steering and policy.

Last change on 25th of April 2024, at 12:27 (CET) | Publication Standard 1.0

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

General information

Theme

Traffic

Begin date

Field not filled in.

Contact information

[e-mail]

Link to publication website

https://data.amsterdam.nl/datasets/PnrmHN-YvSqjhw/langzaam-verkeer-monitoringsysteem-amsterdam-lvma/

Responsible use

Goal and impact

The purposes of processing personal data and for the deployment of calculation rules within the Slow Traffic Monitoring System Amsterdam are:

- Avoiding and controlling unsafe situations (safety);

- Good pedestrian accessibility of public functions and good traffic flow in a wider area around busy locations (accessibility and flow);

- A high-quality public space in which pedestrians feel welcome, safe and comfortable (quality and comfort).

Considerations

Deploying cameras with algorithms is the only way to effectively achieve the goal, as it allows monitoring pedestrian and bicycle traffic in a large area from 1 location.

Human intervention

The algorithm indicates how busy somewhere is, and based on this, it can: 1) generate an automatic script decision for dynamic traffic management (DVM-exchange), per minute. Think of DVM-exchange as controlling information and matrix signs about congestion. 2) by human intervention, a decision to take traffic measures. In this case, the algorithm provides part of the information for that decision. Regular validations are performed that compare numbers from the measurement systems with the actual numbers on the street.

Risk management

If numbers are under/overestimated along a sensor, spikes/decreases in numbers in the database and dashboard can occur. This can give the crowd manager a distorted picture of the actual situation. The system is decision-supportive in operations; this means that visual checks are always made on location before a measure (e.g. setting one-way traffic) is deployed. Location data /Combining datasets: At very low numbers along sensors and when combined with other datasets, it could theoretically be possible that a trajectory of a passer-by could be tracked along several sensors.

Legal basis

The municipality of Amsterdam carries out Slow Traffic Monitoring in public spaces under Article 2 of the Road Traffic Act 1994 (Wvw 1994). The municipality of Amsterdam is road manager. In this role, it is responsible for ensuring the safe and smooth flow of traffic as well as keeping public functions accessible. This also applies to slow traffic and pedestrian flows. The municipality processes these data for the performance of a task of general interest, namely promoting a safe, accessible and comfortable public space.

Operations

Data

Census data from the Langzaam Verkeer Monitoringsysteem Amsterdam (LVMA) provides information from public spaces on numbers of passers-by. Slow traffic is mainly pedestrians and cyclists. This information is used strategically, tactically and operationally to support traffic management. Analyses are also carried out to obtain a picture of how congestion develops over time. On the sensor, images are translated into counting data. - Numbers per direction - Speed - Density in measuring section - Modality The various calculation rules on the near real-time counting data then take care of this data processing, including: - Storing - Aggregating (summarising over a longer period of time) - Enrichment and calculations on the count data - Short-term prediction - Limit values - Dashboard - Making count data publicly available Short-term prediction Aggregated count data from the past 8 weeks From the same quarter of an hour on the day -> what has been the increase/decrease in the number of passers-by at the same time Development current day The City of Amsterdam does not store any personal data in the LVMA. All LVMA sensors are on: https://maps.amsterdam.nl/lvma/ and in the Sensorenregister of the municipality of Amsterdam: https://sensorenregister.amsterdam.nl The aggregated data is aggregated on hourly/daily/weekly level and published as open data on https://data.amsterdam.nl/datasets/PnrmHN-YvSqjhw/langzaam-verkeer-monitoringsysteem-amsterdam-lvma/

External provider

Amsterdam municipality for own calculation rules