Why model before you build?
Infrastructure changes are expensive and difficult to undo. A calibrated model makes the effects of options visible in advance: queues, delays, travel times, route changes, public transport reliability and safety risks. It also helps to explain choices to decision-makers and residents.
Types of traffic models
- Strategic (macroscopic) models describe a whole region and estimate how many trips are made, where they go, which mode is used and which routes are chosen.
- Mesoscopic models add the dynamics of congestion over time, for example for evaluating diversions or roadworks over a network.
- Microscopic models simulate individual vehicles, cyclists and pedestrians, including car-following, lane changing and traffic signals. They are suited for junctions, corridors and roundabouts.
Well-known tools include commercial packages such as PTV Visum and Vissim or Aimsun, and open-source software such as SUMO. The right level of detail depends on the question to be answered.
Data collection
- Traffic counts: manual, video, radar, inductive loops or ANPR, classified by vehicle type.
- Origin-destination surveys and ANPR-based matching to learn where traffic comes from and goes to.
- Travel-time measurements and queue observations.
- Counts of pedestrians, cyclists and public transport passengers where relevant.
Calibration and validation
A model is only useful if it reproduces observed reality. During calibration, parameters are adjusted until modelled flows, travel times and queues match the measurements; validation then checks the model against independent data. Recognised criteria, such as the GEH statistic for traffic flows, make the quality of the model transparent.
Typical questions a model answers
- Does a junction work better as a roundabout or with traffic signals?
- What happens to traffic in the surrounding streets when a new circulation plan or school street is introduced?
- How should traffic be diverted during roadworks or an event?
- Which signal timings or coordination give the best result on a corridor?
- What is the effect of public transport priority on general traffic?
Presenting results
A model is a means, not an end. Results are presented as clear comparisons between scenarios — travel times, queue lengths, delays per mode, levels of service — together with the assumptions and limitations of the model. Animations of microscopic simulations are often the most effective way to explain a proposal to decision-makers and residents.
How Urbiflow helps
We plan and carry out traffic counts and surveys, build and calibrate strategic and microscopic models, test scenarios and report the results clearly. Because our modellers work alongside design and integration engineers, recommendations stay technically feasible.
Relevant for: Cities & municipalitiesRoad & highway authorities