A Cyclist Traffic Simulation and Analysis Tool †
Abstract
:1. Introduction
2. Related Works
3. System Model
3.1. City Representation Model
3.2. Cyclist Mobility Model
- Length: The length of a journey is defined as .
- Frequency: The frequency of a journey , denoted , is defined as the amount of times that the journey has been made and is independent of whether it was done by one or several cyclists. This is,
- Density: The density of a journey , denoted , is defined as the amount of distinct cyclists that made that journey. This is,
- , which is the amount of centralities of the same type as in the zone Z.
- , which is the amount of centralities of the same type as in the zones next to the zone of .
- , which is the amount of centralities of the same type as in the whole city.
4. Simulation Parameters
- Allowing the simulation of a large number of routes at a time.
- Generation of realistic routes, that is, not perform unnecessary turns or excessively long roads.
- Allowing the selection of origin and destination neighborhoods for the simulation. Within these neighborhoods, allowing the selection of the following modes:
- -
- Random points within the neighborhood.
- -
- Random centralities within the neighborhood.
- -
- Centralities selected by importance within the neighborhood.
- Allowing the saving of the selected data to run an experiment again.
- Simulation of trips between all the neighborhoods of the city in a synthetic way.
4.1. Parametrization
- Origin Zone: Neighborhood where the trip begins.
- Destination Zone: Neighborhood where the trip ends.
- Destination Zone Mode and Origin Zone Mode: The simulation mode from each zone:
- Random point: A random corner is selected within the zone.
- Random centrality: The corner closest to a random centrality within the zone is selected.
- Weighted centrality: The corner closest to a weightingly selected centrality is chosen. (Centrality weighting is explained in the following subsection.)
- Number of Trips: It is the number of different trips that want to be simulated between both zones. For each trip, two randomly selected origin and destination points within each zone are chosen.
- Time Slot: It is the time slot between which the trip is made. The exact time will be selected randomly within this time slot. The options are: 6 a.m.–9 a.m., 9 a.m.–1 p.m., 1 p.m.–4 p.m., 4 p.m.–9 p.m., 9 p.m.–6 a.m.
- Daily Frequency: It is the maximum daily frequency in which the trip is made. A random value is selected between 1 and the maximum selected (up to 4), indicating the number of times per day a user makes the trip.
- Density: It is the number of different users that make the same trip. A random value between 1 and the maximum is selected (up to 20).
- Weekly Frequency: These are the days of the week in which the trip takes place.
- Start date and End date: These are the dates that identify the start and end period of the simulation. The trips are created accordingly to the selected days.
Centrality Weighting
Algorithm 1: Randomly selecting a weighted centrality. |
|
4.2. Simulation Outputs
- Total number of simulated routes.
- Mean and variance of the frequency.
- Maximum and minimum frequency values
- Mean and variance of density.
- Maximum and minimum density values.
- Total number of cyclists used for the simulation.
4.3. Full Simulation
- (A)
- Surveys conducted to cyclists of Puerto Madryn in 2017.
- Does not cover all neighborhoods or the population of the city.
- Provide source-destination data.
- The number of surveys reaches a barely minimum number of samples required for the survey to have statistical validity [18].
- (B)
- Population data of Puerto Madryn neighborhoods according to the 2010 census [19]
- The information is rather outdated.
- (C)
- Data collected through the mobile application PMYLap [20], which is a complementary project to the one presented in this article.
Algorithm 2: Full Origin-Destination Matrix Generation. |
|
5. Simulation Analysis
5.1. Journey Analysis
- weight: Current journey weight
- opac_importance: a value between 0.01 and 0.1 that deepen the difference between the largest and smallest weights. The closer this value of 0.1, the more difference there will be between small and large weight values. Thus, making the low weight journeys almost imperceptible and highlighting those of high weight. This range of values was obtained experimentally.
- max_weight: Highest weight among all calculated trips.
Journey Analysis Parametrization
- Zone: It allows the selection of a zone and only calculate the journeys generated by trips that cross it. It is possible not to select any area and calculate the journeys over the entire city.
- Weighting: It allows the visualization of only those journeys that have a weight between the requested range.
- Minimum Length: It allows to calculate the journeys taking into account only the trips with a distance traveled greater than the selected one.
- Time Slot: It allows to calculate the journeys taking into account only the trips taken during the selected time slot.
- Frequency and density values: They allow the analyst to distribute the density and frequency values according to his own criteria. Between both values it must add 1.
- Opacity Level: Once the journeys have been calculated, this option allows to repaint them with a new value in the opacity level. The higher this value, the greater the difference between the most important journeys and the lesser ones.
6. Experimental Results and Discussion
6.1. Specific Trips Generation
6.2. Journey Analysis Experiment
6.3. Full Case Experiment
6.4. Discussion
7. Conclusions
References
- Soledad, M.; Paz, D.; Vázquez, D. La bicicleta en la movilidad cotidiana: Experiencias de mujeres que habitan la Ciudad de México. Rev. Transp. Territ. 2017, 16, 112–126. [Google Scholar]
- Orellana, D.; Hermida, C.; Osorio, P. Comprendiendo los patrones de movilidad de ciclistas y peatones. Una síntesis de literatura. Rev. Transp. Territ. 2017, 16, 167–183. [Google Scholar]
- Aón, L.C.; Giglio, M.L. Patrones modales de movilidad y desarrollo urbano no planificado en la ciudad de La Plata. Rev. Transp. Territ. 2017, 17, 117–144. [Google Scholar]
- Díaz, R.; Rojas, F. Mujeres y Ciclismo Urbano: Promoviendo PolíTicas Inclusivas de Movilidad en América Latina; Technical Report; Banco Inter-Americano de Desarrollo: Washington, DC, USA, 2017. [Google Scholar]
- Damant-Sirois, G.; Grimsrud, M.; El-Geneidy, A.M. What’s your type: A multidimensional cyclist typology. Transportation 2014, 41, 1153–1169. [Google Scholar] [CrossRef]
- Félix, R.; Moura, F.; Clifton, K.J. Typologies of Urban Cyclists. Transp. Res. Rec. J. Transp. Res. Board 2017, 2662, 125–133. [Google Scholar] [CrossRef]
- LINVI. Cycleyzer. 2018. Available online: http://ciclovias-if012.ddns.net:32000/ (accessed on 18 September 2018).
- Ordinez, L.; Barry, D.; Torrico, A.; Mallon, E.; Devia, J.L.; Abella, L.; Alamo, N. Modelo de Análisis para el Dominio de Tránsito Ciclista. Available online: http://sedici.unlp.edu.ar/handle/10915/63863 (accessed on 18 September 2018).
- Ordinez, L.; Barry, D.; Torrico, A.; Mallon, E.; Devia, J.L.; Abella, L.; Alamo, N. A Bike Traffic Analyzer for the Determination of Cycle Lanes. In Proceedings of the Conferencia Internacional de la Sociedad Chilena de Ciencia de la Computación (SCCC 2017), Arica, Chile, 16–20 October 2017. [Google Scholar]
- Blokpoel, R.; Joueiai, M.; Blokpoel, R.; Joueiai, M. Bicycle Modelling in SUMO for Accurate Traffic Simulation. Available online: http://www.xcycle-h2020.eu/Resources/Files/Bicycle_SUMO.pdf (accessed on 18 September 2018).
- David Gerstle, D.M.; Yang, Q. Microsimulation of Bicycles for Planning and Design. Technical Report, Caliper Corporation. 2013. Available online: https://www.caliper.com/Library/micro-simulation-of-bicycles-for-planning-and-design.htm (accessed on 18 September 2018).
- Biking & Walking. Available online: https://www.ptvgroup.com/en-us/solutions/topics/biking-walking/ (accessed on 18 September 2018).
- Ziemke, D.; Metzler, S.; Nagel, K. Modeling bicycle traffic in an agent-based transport simulation. Procedia Comput. Sci. 2017, 109, 923–928. [Google Scholar] [CrossRef]
- Abrahamsen, R.N. Simulating Cyclists in a Simulator with the Use of Behaviour Trees. Master’s Thesis, Norwegian University of Science and Technology, Trondheim, Norway, 2016. [Google Scholar]
- Mereu, A. Analyzing the Behavior of Cyclists at Intersections to Improve Behavior Variability within Micro-Simulation Traffic Models. Master’s Thesis, University of Waterloo, Waterloo, ON, Canada, 2015. [Google Scholar]
- Ma, X.; Luo, D. Modeling cyclist acceleration process for bicycle traffic simulation using naturalistic data. Transp. Res. Part F Psychol. Behav. 2016, 40, 130–144. [Google Scholar] [CrossRef]
- Kwigizile, V.; Oh, J.S.; Ikonomov, P.; Hasan, R.; Villalobos, C.G.; Kurdi, A.H.; Shaw, A. Real Time Bicycle Simulation Study of Bicyclists’ Behaviors and Their Implication on Safety; Technical Report; Western Michigan University: Kalamazoo, MI, USA, 2017. [Google Scholar]
- Perfil Ciclista 2017. 2018. Available online: http://transporteativo.org.br/ta/?page_id=9762 (accessed on 18 September 2018).
- Argentina, I. Unidades Geoestadísticas. 2018. Available online: https://www.indec.gov.ar/codgeo.asp (accessed on 18 September 2018).
- Aguirre, A.; Ordinez, L. PMYLap. 2018. Available online: https://play.google.com/store/apps/details?id=com.linvi.ariel.pmylap (accessed on 18 September 2018).
a Descriptive statistics | |||
---|---|---|---|
Descriptors | Formula (5) | Formula (6) | Formula (7) |
Mean | 0.08160 | 0.17652 | 0.27475 |
Std err | 0.00431 | 0.00932 | 0.02122 |
Median | 0.05263 | 0.125 | 0.1 |
Std dev | 0.10123 | 0.21871 | 0.49768 |
Asymmetry | 1.57334 | 2.26073 | 3.71781 |
Interval | 0.6 | 1 | 4.13043 |
Maximum | 0.6 | 1 | 4.13043 |
Sum | 44.88419 | 97.090 | 151.11496 |
b Pearson correlation coefficient | |||
Correlations | |||
(1) | 0.60599 | 0.23677 | −0.156959 |
(2) | 0.51936 | 0.03498 | −0.406773 |
(3) | 0.39417 | 0.51343 | −0.14812 |
Origin Zone: Gobernador Fontana |
Origin Zone Mode: Random point |
Destination Zone: Parry Madryn |
Destination Zone Mode: Random centrality |
Number of Trips: 5 |
Time Slot: 9 a.m.–1 p.m. |
Daily Frequency: 4 |
Density: 5 |
Weekly Frequency: M–T–W–TH–F |
Start date: 28 May 2018 |
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Ordinez, L.; Savarro, M.; Ascagorta, O.; Cura, R.R.; Buckle, C.; Barry , D.; Stickar, R. A Cyclist Traffic Simulation and Analysis Tool. Proceedings 2018, 2, 1220. https://doi.org/10.3390/proceedings2191220
Ordinez L, Savarro M, Ascagorta O, Cura RR, Buckle C, Barry D, Stickar R. A Cyclist Traffic Simulation and Analysis Tool. Proceedings. 2018; 2(19):1220. https://doi.org/10.3390/proceedings2191220
Chicago/Turabian StyleOrdinez, Leo, Mauricio Savarro, Octavio Ascagorta, Rodrigo René Cura, Carlos Buckle, Damián Barry , and Romina Stickar. 2018. "A Cyclist Traffic Simulation and Analysis Tool" Proceedings 2, no. 19: 1220. https://doi.org/10.3390/proceedings2191220