Pedestrian simulation in signalised intersections in Vissim

Student thesis: Master thesis (including HD thesis)

  • Line Degn Andersen
  • Morten Andersen
4. term, Transport Engineering, Master (Master Programme)
For several years simulation of traffic has been a popular tool to predict consequences of future changes in the road network. In Denmark Vissim is the most commonly used program for this. Vissim is frequently used to visualize future changes in the road network, and these visualizations are an important part of the political decision-making. Until recently the focus has primarily been on making sure that motorized vehicles are simulated correctly, but now focus has increased on being able to simulate other transportation forms as well.

This report aims to determine how involving pedestrians in simulations in Vissim influences the description of the capacity of a signalized intersection. The report also seeks to answer how the use of the new pedestrian module, Viswalk, in Vissim affects the description of the capacity of a signalized intersection.

This is done by selecting two intersections with different amounts of pedestrians and vehicle traffic and creating three different scenarios for each intersection. The first intersection chosen is the intersection between Dannebrogsgade and Kasetvej in Aalborg which has a fairly normal amount of pedestrians and vehicles. The second intersection is the intersection between Vesterbro and Urbansgade in Aalborg, which has a high volume of both pedestrians and vehicles making left or right turns.
For each intersection video cameras have been placed to film the traffic in the intersection in order to do a count of vehicles, pedestrians and cyclist, and also to be able to build two separate models for the intersections in Vissim. After this, three scenarios are made for both intersections:

Scenario 1: A simulation of the intersection without any pedestrians.
Scenario 2: A simulation of the intersection with pedestrians but without the use of Viswalk.
Scenario 3: A simulation of the intersection with pedestrians and with the use of Viswalk.

It is important to point out that the model for each intersection is exactly the same in all scenarios apart from the way pedestrians are simulated. To compare these three scenarios, data for travel times for all possible routes and queue lengths at every stop mark are collected in Vissim for each intersection. These are then compared and statistical tests are performed to determine whether there is a statistically significant difference between the medians of the travel times (data are not normally distributed) for the three scenarios.

The conclusion is that it does influence the capacity of the intersections whether pedestrians are included or not. The difference is smallest in Dannebrogsgade/Kastetvej where travel times in average are 10\% lower if pedestrians are not included. The biggest difference is in Vesterbro/Urbansgade where travel times in average are 27\% lower if pedestrians are not included and on some routes the difference is as high as 56\%. This means that including pedestrians in intersections with a high volume of pedestrians combined with a large amount of vehicles making left or right turns makes a big difference.

The other conclusion is that it does not influence the capacity of the intersection whether Viswalk is used to simulate pedestrians or not. The travel times differ only 3\% in average in Dannebrogsgade/Kastetvej and 6\% in Vesterbro/Urbansgade. This means that regardless of the amount of pedestrians and vehicles in these two intersections, there is no remarkable difference in travel times between simulating pedestrians with or without Viswalk.
LanguageDanish
Publication date10 Jun 2014
Number of pages114
ID: 198661948