Agent-based simulation of bicycle traffic - Background information

Simulating Bicycle
Traffic in Salzburg
- Background Information
Martin Loidl
Department of Geoinformatics, Z_GIS
University of Salzburg
martin.loidl@sbg.ac.at | http://gicycle.wordpress.com

Bicycle Promotion
 Urban agglomerations suffer from negative effects of -
still growing! - car traffic
 Environmental impact
 Financial/economic impact (externalities!)
 Social impact
2
reuters.com salzburg24.at salzburg.orf.at

Bicycle Promotion
 Bicycle as efficient, urban mode of transport
 Cheap
 Accessible
 Environmentally friendly
 Healthy
 Flexible
 Fast within 5-10km
3

Investments
 Infrastructure, events, information
4
stadt-salzburg.com
Benchmarking?
salzburg24.at
radlkarte.info

Accidents
 Accidents reported by police are geocoded
5
Statistical Population?

Lack of Data
 “Such a lack of reliable and high-resolution data
about bicycle-specific factors is hence one of the most
central issues in research into cycling, as it often
hampers to get in-depth knowledge on the factors that
significantly influence both bicycle use and cycling
accidents.” (Vandenbulcke-Plasschaert 2011: 20)
 “Without information about how much cycling is being
done, statements about how many cycle crashes occur
are of limited use.” (OECD 2013: 63)
VANDENBULCKE-PLASSCHAERT, G. 2011. Spatial analysis of bicycle use and accident risks for cyclists. PhD, Université catholique de Louvain.
OECD 2013. Cycling, Health and Safety. In: TØRSLØV, N. (ed.). Paris: ITF-OECD Working Group on Cycling Safety.
6

Data needed
 To know [estimate] when + where + how many
 Benchmarking („What effect do my initiatives have?“)
 Demand-based planning („Where is infrastructure most
needed?“)
 Traffic management („How can I guide bicyclists to
efficiently distribute all traffic participants in the network?“)
 Routing information („At which point of time are certain
connections suitable/to be avoided?“)
 Accident analysis („What is the risk to be involved in an
accident?“)
7

What we have so far I
 Epidemiological analysis (prevalence, risk) based on
mostly weak, highly aggregated data
 Study from 2012 (ISI-indexed) uses data from study from
2008
 Study from 2008 refers to data from the European
Comission, published in 2002
 This publication is an annual report for 2000
 The annual report uses data which where nationally
collected between 1970 and 1997
8
https://gicycle.wordpress.com/2014/02/11/where-do-the-data-come-from

What we have so far II
 Traffic models for motorized individual
traffic and partly for public transport
 Obligation to register cars (know exact
number of vehicles)
 Large intrest in data  telematic systems
 No traffic models for bicyclists and
pedestrians
 No registration
 Lower demand (better: direct economic
pressure)
 Many influental variables
Passenger car
Public transport
9
mvv-muenchen.de

What we have so far III
10
 Traffic flow simulation
http://www.mtreiber.de/MicroApplet_html5

What we have so far IV
 Single counting systems along major routes
 Don‘t necessarily allow for global information
11
Stadt Salzburg

What we want to have
 Estimation of flows per segment per time interval
 Different scenarios for environment and agents
 We provide the following data:
 Road network as routeable graph with all available
attributes
 Socio-demographic data in 1.000/250m grid
 LULC data (digital cadastral map)
 POIs
 Data from 6 counters
12

Agent-based simulation of bicycle traffic - Background information

More Related Content

What's hot (20)

Viewers also liked (6)

Similar to Agent-based simulation of bicycle traffic - Background information (20)

More from Martin L (13)

Recently uploaded (20)

Agent-based simulation of bicycle traffic - Background information