PowerPoint Presentation
The importance of interactions in determining service measures
for bicyclesChris [email protected]
www.osowski.co.ukInstitute for Complex Systems Science, University
of SouthamptonBen [email protected]
Research Group, University of Southampton
Institute for ComplexSystems SimulationInstitute for
ComplexSystems SimulationIs this good?From: Warrington Cycle
Campaign Cycle Facility of the Month (http://goo.gl/EiBnDv)
Institute for ComplexSystems SimulationAre these value for
money?
From: Warrington Cycle Campaign Cycle Facility of the Month
(http://goo.gl/EiBnDv)Institute for ComplexSystems Simulation
How good is this?Institute for ComplexSystems SimulationCapacity
of Motor Infrastructure
From: Transportation Research Board. (2010). Highway Capacity
Manual.Institute for ComplexSystems SimulationCapacity of Motor
Infrastructure
From: www.flownz.comInstitute for ComplexSystems
SimulationCapacity of Pedestrian Infrastructure
From: Halcrow Group Ltd. (2005)From: Colin Buchanan Ltd.
(2010)From: Fruin, J. J. (1971). Pedestrian Planning and
DesignInstitute for ComplexSystems SimulationCapacity of Cycle
Infrastructure
From: Navin, F. P. D. (1994). Bicycle Traffic Flow
Characteristics: Experimental Results and Comparisons. ITE Journal,
(March), 3136.Institute for ComplexSystems SimulationA Quote from
Fruin[...] many authorities are using maximum capacity ratings for
dimensioning pedestrian space. No evaluation or consideration of
human convenience has been made in developing these design
standards. The flow curves [... demonstrate] that the maximum
capacity of a pedestrian traffic scheme is attained only when there
is a dense crowding of pedestrians.From: Fruin, J. J. (1971).
Pedestrian Planning and DesignInstitute for ComplexSystems
SimulationSourceUseTypeCapacity(bicycles per metre per hour)Design
Std: UK?CROW (2007)Design Std: NetherlandsEmpirical75 - 187.5
(width dependent; NB. Comfort not absolute capacity)Botma
(1995)Design Std: USA (HCM)Theoretical650Vejdirektoratet
(2012)Design Std: DenmarkTheoreticalPath up to 2.0m: 1000Path over
2.0m: 1500Navin (1994)NoneEmpirical (with theoretical
extrapolation)4000Capacity of Cycle InfrastructureInstitute for
ComplexSystems SimulationWhat about Quality of Service?Botma (1995)
assumed no impedance; i.e. fixed speed.Does this matter?
Can we use simulation tools, take the best practice of the
simulation of other modes and cross-apply it to find out?Institute
for ComplexSystems SimulationUse Pedestrian Modelling?Often
continuous space (2D) models.Successful tools at variety of scales
PedrouteLegionVisWalk
From: Halcrow Group Ltd. (2005)From: Colin Buchanan Ltd.
(2010)Institute for ComplexSystems SimulationMacroscale
speed-flowMicroscale interactionsSFM
12The Social Force ModelHelbing, D., & Molnr, P. (1995).
Social force model for pedestrian dynamics. Physical Review E,
51(5), 42824286.Motivation to act
Institute for ComplexSystems SimulationOur Model PurposeDo
interactions matter?
From: CXMagazine.comInstitute for ComplexSystems
SimulationInteractions are obviously intuitive but are they
non-trivial?Is it a reasonable simplification?14
Our Model ImplementationAgent Based Model2 dimensional
continuous spaceUnidirectional flowFixed path widthParallel
boundaries
Institute for ComplexSystems SimulationOur Model Force
Generation
Institute for ComplexSystems SimulationSFM Directional
PerceptionIn Helbing and Molnr (2005):
Institute for ComplexSystems SimulationOur Model Directional
Perception
Institute for ComplexSystems SimulationSFM vs Our
ModelPedestrians (SFM)CyclesFuture forces and squishinessPresent
forces and NOT squishyLimited directional perceptionMore complex
directional perceptionContinuous bounded speed Bounded speed with
step minimumUnbounded accelerationBounded accelerationInstitute for
ComplexSystems SimulationOur Model Parameters and OutputsBicycle
behavioural parameters:Generally from CROW (2007)Exploratory
variables:Path Width and Bicycle Arrival RateOutput data:Average
speed, average crashing proportion, average distance to the nearest
bicycle, etc.Institute for ComplexSystems SimulationResults Speed
vs. FlowHigh QualityVariable QualityPoor QualityInstitute for
ComplexSystems SimulationResults Crashes(No impedance)Institute for
ComplexSystems SimulationInteractions as crashesCrash graphFixed
speed quasi-linear; variable non-linear22Results CrashesFlow
breakdown with speed-selection occurs primarily as multi-bicycle
collisions at the entry:
Without speed-selection:
Institute for ComplexSystems SimulationMBC indicates a regime
change. Replicates results seen empirically in Hoogendoorn.General
crunching without.Speed-flow bimodal breakdown.
23SourceUseTypeCapacity(bicycles per metre per hour)CROW
(2007)Design Std: NetherlandsEmpirical75-187.5 (width dependent;
NB. Comfort not absolute capacity)This ModelSimulationca. 500Botma
(1995)Design Std: USA (HCM)Theoretical650Vejdirektoratet
(2012)Design Std: DenmarkTheoreticalPath up to 2.0m: 1000Path over
2.0m: 1500Navin (1994)NoneEmpirical (with theoretical
extrapolation)4000Results Literature ComparisonInstitute for
ComplexSystems SimulationConclusionsLack of robust quantitative
measures for capacity or quality of service.
Inherent assumption that service quality is satisfactory up to
capacitybicycles do not interact in a meaningful way.Institute for
ComplexSystems SimulationSources are often historic, mathematical
and/or not empirical.Suspiciously low precision or conspicuous
precision which happens to equate to low precision imperial
measure.
Non intuitive but does it matter?25Conclusions2D microsimulation
model using the SFM.Basic behavioural traits applied.
Data show qualitatively different result...sudden collapse in
quality of service.
Modelling confirms intuition that we cant ignore service quality
for non-trivial flow rates.Institute for ComplexSystems
SimulationAcknowledgementsEPSRC Doctoral Training Centre grant:
EP/G03690X/1
University of Southampton:Institute for Complex Systems
Simulationwww.icss.soton.ac.ukTransportation Research
Groupwww.trg.soton.ac.ukIRIDIS High Performance Computing
Facility(and support services)Institute for ComplexSystems
Simulation27Questions?Paper at: https://db.tt/8APJ8Vrk
Institute for ComplexSystems SimulationMaths
Institute for ComplexSystems Simulation
Our Model AlgorithmInstitute for ComplexSystems SimulationOur
Model Bicycle Operation
Institute for ComplexSystems SimulationBicycle Arrival Rate
ProximityInstitute for ComplexSystems SimulationBarriers to
CyclingTfGM, 2011
Institute for ComplexSystems Simulation
Barriers to CyclingFrom: City of Copenhagen. (2011). Copenhagen:
City of Cyclists - Bicycle Account 2010.Institute for
ComplexSystems Simulation