Efficient traffic planning and prediction
The geographical location of Pilsen and the layout of the population and industry bring transport problems. Pilsen lies at the confluence of four rivers and there are very few bridges. Approximately one third of the population lives in the north of the city, while most employers are on the south.
The city of Pilsen is planning essential roadworks during the coming years which will have a significant impact on city life and transport. Pilsen will implement other essential mobility related works according the Sustainable Mobility Plan (SUMP) to improve traffic situation in the city.
Traffic experts do not have tools/applications, that can predict what will happen to traffic flows across the city when a new road will be added, or an existing road will be partially or completely closed.
The outcome of this case study will be a visual application for traffic experts. With this tool they will be able to test, model and plan new roads in the traffic model of the city and plan the best way to improve traffic situation in Pilsen. This application will help with planning new routes and also with making decisions for events that affect traffic (coordination of roadworks, closures, etc.).
For example, if the city wants to build new city road, in the application you can draw it in and visualize different variants of the project (such as the capacity of the road, connections to surrounding traffic). In the application will be possible to see the impacts of the new measures on surrounding traffic and policy makers will have more information for their decisions.
Challenges encountered:
The first challenge was improving the existing traffic model which is the cornerstone of this case study and which will be used in other future case studies. The improved traffic model will be bidirectional and will be based on real traffic data, because it will be calibrated by traffic sensor data.
In the calibrated traffic model various restrictions can be entered, such as closures and roadworks, changing the capacity of the road, entering a next section.
A second challenge was using traffic sensor data, which is new open data not used before. These datasets will be used also in other case studies. The team learned how to aggregate real-time traffic sensor data for the calibration of traffic model and for visualization of traffic volumes and their changes over time to create a real-time traffic map.
As the first example of use was created an application based on WebGLayer technology, which displays traffic intensity according obtained data from detectors (https://dopravaplzen.innoconnect.net/). The application allows you to view data from March 2017 to the previous day. Traffic intensities are displayed only in places where traffic detectors are located. They are not calculated in other places as in Traffic Modeler technology.
As the main application was created application Traffic Modeller. The public part of application (https://intenzitadopravy.plzen.eu/) shows the traffic intensity in the past in the places where traffic detectors are located and enables modeling the expected traffic development in the future based on enhanced traffic model and scheduled closures or SUMP measurements. The non-public part of application allows to enter experiments, model traffic based on selected closures and publish approved closures. This application should be usable for modelling and visualising short term traffic restriction (from SUPERDIO) and also for simulating the long term traffic measurement from SUMP.
Stakeholders:
- SITMP – Information Technology Administration of the City of Pilsen: Responsible for the project (application, visualizations etc.);
- SVSMP – Organization for the management and maintenance of the publicly accessible property of the city: Responsible for traffic sensor data and traffic model;
- OD MMP – Traffic department of the City of Pilsen: Responsible for planning of traffic roadworks and new roads;
- UKRMP – The Urban Planning and Development Institute of the City of Pilsen: Responsible for Land use map;
- Plan4all association (which brings together organizations working to develop the results of EU research projects)
- ZČU: Katedra geomatiky – University of West Bohemia: Department of Geomatics
Future Actions:
- Decide how to visualize traffic volume in map application (historical data from sensors, predicted situation from the traffic model, comparison of traffic intensity in place and time according to ongoing traffic measures). There may be a problem with locations without traffic sensors.
- Creating a functional application for visualizing planned traffic measures over a new transport model;
- Testing the usability of the application.
- Disseminating the application
Lessons learned:
- There is a problem with traffic sensor data sets from datacenter – it provides two data formats – DATEX and CSV – and there is a difference between them – it is necessary to convert and aggregate the data to get right results.
- Traffic sensors occur only at several crossroads in the city. There may be a problem with visualization of historical traffic volume in locations without traffic sensors.
- Aggregated data from traffic sensors over a longer period represent greater computational demands for visualization.
Outcome impact:
- The application will be useful for traffic experts planning major transport constructions in the city. The application is able to accurately illustrate how the planned closure actually loads the surrounding traffic and can also be used by designers who take into account the results of modelling when designing new roads and service systems;
- Created tools/visualizations will be able to simulate potential planning impacts (modification and construction new infrastructure) and also verification of the impacts of the measures already implemented ;
- The application will be useful for crisis management. They can simulate possible consequences of traffic constraints and revise existing or design new crisis plans in traffic for different situations (floods, explosions, accidents, evacuations, terrorists, …).
Visualisations
https://intenzitadopravy.plzen.eu/
https://dopravaplzen.innoconnect.net/