Predicting travel times

Continuous and automatic traffic counting

Continuous and automatic traffic counting

Since 2000 the City of Delft has been continuously, and automatically, counting traffic in the city. There are already more than 120 counting points spread over 40 controlled intersections. Delft has for some time been a leader in this field and is using the figures to monitor traffic flows and calibrate transportation models. But, Delft is going further in developing this field of expertise.

Real time

The consultancy firm Cityflow has stated, that on the basis of the data gathered (number of cars waiting at traffic lights); they are able to predict travel times within urban areas. Predicting travel times on urban road network is more difficult that on freeways. On an urban road network many factors can interfere with the flow of traffic. Congestion can arise quickly and just as quickly be resolved. This is especially the case in Delft where the construction of a new railroad tunnel causes short term delays.

Under the umbrella of the National ‘Beter Benutten’ project (‘better use’: increase road capacity without adding lanes) Delft, together with Stadsgewest Haaglanden (The Hague region coordinating body) and the municipality of Rijswijk in 2011 launched a research project to predict travel times on an urban road network.

Service

Cityflow has developed an algorithm that calculates travel times. It uses real-time and historical data from 16 controlled intersections in Delft and 4 in Rijswijk and combines this data with calculations from a dynamic and a static traffic model. At the moment, the travel times to the Zuidpoort parking garage from the Prins Claus interchange can be calculated. Travel times can now be given for those heading into the center of Delft on the Dynamic Route Information Panels (DRIPS) on the A4 or A13.

Further investigation

TNO (an independent research organization) has now been commissioned to assess whether the estimated travel times match the actual travel times. TNO will use blue tooth technology to gather travel data. They will measure how fast cars equipped with a car kit for mobile phones cover the recommended route.

Results

This project has already produced the following results:

  • It has been shown that it is possible to use data of traffic lights to predict real-time travel times on an urban road network.
  • Traffic data from various types of traffic control devices have been standardized (using V-log) and can be shared with the traffic control centers of Delft and The Hague.
  • Providing the National Data Warehouse for Traffic Information (NDW) with the calculations of real-time traffic flows.

In the future, the system will be expanded to:

  • Include bridge openings on the Schie and Vliet. The province of South Holland is busy adjusting their ICT communication accordingly using V-log.
  • Extend the network to all of Delft and then throughout Haaglanden.
  • Communicate results via apps, urban DRIPs, car navigation systems and parking lots.

The project has also indicated a new research direction:

  • Automatic analysis of the quality of performance of traffic lights.
  • Detection of poorly timed traffic lights.

Share this page