=Paper=
{{Paper
|id=Vol-3282/icaiw_wits_1
|storemode=property
|title=Mobility Studies in Villavicencio: Situation and Prospects Towards a Smart Mobility
|pdfUrl=https://ceur-ws.org/Vol-3282/icaiw_wits_1.pdf
|volume=Vol-3282
|authors=Carlos Alberto Díaz Riveros,Alejandra Baena,Juan Camilo Ramirez
|dblpUrl=https://dblp.org/rec/conf/icai2/RiverosBR22
}}
==Mobility Studies in Villavicencio: Situation and Prospects Towards a Smart Mobility==
https://ceur-ws.org/Vol-3282/icaiw_wits_1.pdf
Mobility Studies in Villavicencio: Situation and
Prospects Towards a Smart Mobility
Carlos Alberto Díaz Riveros1 , Alejandra Baena2 and Juan Camilo Ramírez2,*
1
Corporación Universitaria del Meta, Villavicencio, Colombia
2
Universidad Antonio Nariño, Bogota, Colombia
Abstract
Congestion is a contemporary and growing problem in most urban areas around the world, and even
middle-sized cities are not exempt from this. In Colombia, this problem is especially accentuated in
large cities such as Bogotá, Medellín, and Cali; however, in mid-sized cities such as Villavicencio the
situation also tends to worsen and this stresses the need for a better understanding of traffic so that
appropriate measures can be taken to minimize its impact. In this city, there is a particular point that
attracts congestion near the roundabout called Grama, near Corporación Universitaria del Meta thus
affecting the nearby academic community and residents in the area. The objective is to determine
the main factors contributing to the increase in traffic and slow mobility throughout the day at the
Grama roundabout in Villavicencio, Colombia, through speed measurements. To achieve this, the license
plate method was used in order to measure vehicle speeds at the Grama roundabout during peak and
off-peak times. As a result, a characterization of traffic speed by vehicle type, namely public transport
and private vehicles, during both peak and off-peak times by segments, providing an insight into the
main contributing factors to congestion in the area under study.
Keywords
Spatialization, Transport, Travel Time, Modes of Transportation
1. Introduction
Mobility in a city is a basic need of society because moving from one point to another allows
individuals to carry out their daily tasks. On the other hand, the dynamics of modern cities
are characterized by high rates of mobility of people and goods, a dynamic that is associated
with the physical aspects of the community activities and their territorial distribution: the
more the city grows, the more the vehicle fleet, the number of people, increases the length
of displacement and other aspects that make a permanent diagnosis necessary to generate
improvement strategies [1, 2].
Currently, the political agenda on mobility in most cities has promoted numerous mobility
projects based on the concept of sustainability that encompasses a set of processes and actions
aimed at achieving the rational use of means of transport for individuals and professionals [3].
ICAIW 2022: Workshops at the 5th International Conference on Applied Informatics 2022, October 27–29, 2022, Arequipa,
Peru
*
Corresponding author
$ carlos.diaz@unimeta.edu.co (C. A. Díaz Riveros); alejandra.baena@uan.edu.co (A. Baena);
juan.ramirez@uan.edu.co (J. C. Ramírez)
0000-0003-4504-136X (C. A. Díaz Riveros); 0000-0002-2216-6816 (A. Baena); 0000-0003-0045-4144 (J. C. Ramírez)
© 2022 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).
CEUR
Workshop
Proceedings
http://ceur-ws.org
ISSN 1613-0073
CEUR Workshop Proceedings (CEUR-WS.org)
259
Carlos Alberto Díaz Riveros et al. CEUR Workshop Proceedings 259–270
Its primary focus is the reduction of traffic congestion, promoting road safety, decreasing air
and noise pollution, energy consumption, travel time reduction, and public transport service,
among other aspects. Therefore, the sustainable objectives for mobility are mainly focused on
improving the quality of life of citizens of current and future generations [4].
The mobility plans in regions such as Latin America and the Caribbean present significant
challenges associated with the uncontrolled growth of cities and the constant increase in
vehicles fleet, poor mobility culture, corruption, etc [4, 5]. However, efforts are being made by
the administration, the university, the state, and society to develop efficient planning processes
to optimize scarce resources to cover the demand for human development within the principle
of equity. Specifically, maximize the use of the existing infrastructure demanded by the different
modes of transport through traffic management processes. Therefore, planning should be a
continuous and permanent process of analyzing the current situation and forecasting future
scenarios in its development [6].
In Colombia, the city of Villavicencio, located in the east, is a principal commercial point
that connects the center with oil and economic zones [7]. Therefore, mobility is a dimension of
great concern, considering the challenges of many cities in LATAM and related to its features,
such as the roads with bad conditions and the high number of vehicles from different zones of
the country. Consequently, it impacts the travel times for displacement, road safety, and other
aspects of the community’s welfare.
Accordingly, it is crucial to diagnose the mobility dimension in strategic points of the City
using data for modeling and understanding the dynamics of mobility in the City [8]. Therefore,
this study starts with characterizing vehicular traffic on two Paths of the main Avenue of
Villavicencio, Alfonso Lopez. Then, it permitted the estimation of some traffic indicators, such
as the vehicular travel time and speeds at crucial moments of the displacements, well known as
peak and valley hours.
Finally, an overview of the traffic modeling is presented as a close future perspective to
include in the mobility studies in cities such as Villavicencio to reinforce the diagnosis phases
in the planning projects. Also, the importance of the data quality is emphasized to understand
the context deeply, guiding urban and transport decisions to mobility service improvement in
Cities.
2. Methodology
Based on a fieldwork design, the mobility study was developed using the license plate method
to analyze the travel time of vehicles on Alfonso Lopez Avenue in Villavicencio, Colombia.
The travel speed of each path was determined considering two types of mobility. The off-peak
(valley) hours (8:30 a.m – 11:30 a.m and 2:30 p.m – 5:00 p.m) and the peak hour (7:30 a.m –
8:15 a.m, 11:45 a.m – 12:30 p.m and 5:30 p.m. – 6:45 p.m). The fieldwork was designed using
counting and verification tools, and prospection work was done in situ with Waze and Google
Maps Apps.
260
Carlos Alberto Díaz Riveros et al. CEUR Workshop Proceedings 259–270
2.1. Study Zone
The Alfonso López Avenue was analyzed in two paths, one going from GRAMA to Carrera 33
with Calle 32 (VORAGINE) and the other one from Carrera 33 with Calle 32 (VORAGINE) to
GRAMA. Each path was divided into study segments, thirteen for GRAMMA-VORAGINE and
fourteen for VORAGINE-GRAMMA as presented in Table 1.
Table 1
Selected segments by each Path under mobility study
SEGMENT GRAMMA-VORAGINE VORAGINE-GRAMMA
1 GRAMA-CALLE 41 CALLE32-CALLE33
2 CALLE 41-CALLE 40 CALLE 33-PALACIOJUSTICIA
3 CALLE 40-CALLE 38 PALACIOJUSTICIA-CALLE 35
4 CALLE 38-CALLE 37B CALLE 35-CALLE 35A
5 CALLE 37B-CALLE 37A CALLE 35A-CALLE 36
6 CALLE 37A-CALLE 37 CALLE 36-CALLE 37
7 CALLE 37-CALLE 36 CALLE 37-CALLE 37A
8 CALLE 36-CALLE 35 CALLE 37A-CALLE 37B
9 CALLE 35-RESTAURAN CALLE 37B-CALLE 37D
10 RESTAURAN-ARQUITECT CALLE 37D-CALLE 39D
11 ARQUITECTURA-PANTANOVARGAS CALLE 39D-CALLE 40
12 APANTANOVARGAR-CALLE33B CALLE 40-CALLE 40A
13 CALLE 33B-CALLE 32 CALLE 40A-CALLE 41
14 CALLE 41-GRAMA
2.2. Plate License Method
The method was used to obtain travel time at two established points. This procedure consists of
taking data of the hour of passage, in which the vehicles that travel a segment of the path pass
through two or more points of the segments identifying them by their license plate. Then the
travel time between the points of the road is calculated knowing the distances that separate them.
It only measures travel times; however, It does not measure delays. Therefore, its efficiency
depends mainly on the number of vehicles that pass through the considered segment; if the
traffic is heavy, an acceptable sample can be obtained to study some aspects of mobility at the
location of interest.
2.3. Data Acquisition
A group of practitioners was arranged for fieldwork for the data collection by hand [9]. Two
experimenters were located at the corner of the segment under study, called point controls.
First, using a format, the last letter of the plate was written with their respective number and
the vehicle number. Finally, with the help of the cell phone timer, the exact time the vehicle
passed in front of the practitioner was recorded until 40 data were obtained. The data were
collected for a peak hour determined by the time of day with a significant number of people and
vehicles on the streets; there is greater use of public service. On the other hand, the off-peak
261
Carlos Alberto Díaz Riveros et al. CEUR Workshop Proceedings 259–270
or valley hour corresponds to the additional time of the day with fewer people and vehicles
circulating the city. Data for peak and valley hours were acquired as follows.
• Valley Hour (Off-Peak) Daytime: Wednesday from 9:10 a.m to 11:35 a.m
• Peak Hour Daytime: Thursday from 7:35 am to 8:12 am, Friday from 7:50 am to 8:11 am,
and Tuesday from 7:35 am to 8:15 am
• Valley Hour Night: Wednesday from 3:06 pm to 4:45 pm
• Peak Hour Night: Thursday from 5:32 pm to 6:37 pm and Friday from 5:50 pm to 6:46 pm
2.4. Data Processing
Travel time is determined as the time required to traverse a specific route. For this study, the
distance of each segment was taken with an odometer. On the other hand, the travel when the
vehicles crossed the zone was obtained with a cellphone stopwatch. As a result, the average
speed of each vehicle was estimated with a simple physical relation of variables. However, it is
crucial to take into mind some considerations of the variables involved as it is mentioned as
follows:
• Travel time: It is the total elapsed time, including stops and waiting, required for a vehicle
to travel from one point to another on a specified route under existing traffic conditions.
• Travel speed: It is the average speed in the studied segments. It is obtained by dividing
the total distance by the total time elapsed between terminal points (including all delays,
i.e., travel time)
The results shown in the following section are the product of the tabulation of the data by
the plate method, where the travel times are acquired in each Path of Alfonso López avenue.
Furthermore, the data taken for each segment was compared by the practitioners. Finally, the
delay times were determined for each registered vehicle to move from the initial point to the
final point of the sample. For analysis, the waiting time at traffic lights and the valley and peak
hours were considered. The study aimed to acquire data and semi-quantitative traffic behavior
analysis by estimating travel and speed time indicators. Therefore, an statistical analysis was
not presented for this study.
3. Results
3.1. Experimental Zone Characterization
The characterization of the analysis for the two Paths under study was developed using Google
Maps as an informatic tool to collect information related to the distance and relevant conditions
of the zone shown in Figure 1. However, the in situ recognition was also made to verify the
information and optimize the experimental design before fieldwork. Figure 1 (a) shows the
GRAMMA-VORAGINE path traveled in a vehicle at pick hour. It comprises the GRAMA avenue
and the headquarters of the University of Meta called Pantano de Vargas, with a route on Carrera
33 from the Pantano de Vargas building between the Carrera 33 with Calle 34 to Calle 32 with
262
Carlos Alberto Díaz Riveros et al. CEUR Workshop Proceedings 259–270
Figure 1: Google Earth App Images for the two Paths. (a) GRAMMA-VORAGINE with a distance of
1643 meters. (b) VORAGINE-GRAMMA with a distance of 1517 m. (c) The thirteenth segment for
GRAMMA-VORAGINE (d) The fourteenth segment for VORAGINE-GRAMMA
Carrera 33, which is a sector known as La Vorágine Service Station. The total distance was 1643
m, according to Google Maps App data; Path Two called VORAGINE-GRAMMA has presented
in Figure 1 (b), and it was traveled between Calle 32 and Carrera 33 to Glorieta de la GRAMA,
with a distance of 1517 m. Each path was analyzed by segments, where thirteen segments
were selected for GRAMMA-VORAGINE and Fourteen for VORAGINE-GRAMMA, as shown in
Table. 1. Figure 1 (c) shows the thirteenth segment analyzed for GRAMMA-VORAGIME, which
corresponds to CALLE 33B with CALLE 32, and Figure 1 (d) shows the fourteenth segment at
CALLE 41 -GRAMMA for VORAGINE-GRAMMA.
3.2. Average Speed Estimation
The License Plate Method described in Section 2 was used to collect information about travel
time in Valley and Peak hours for the two Paths surveyed during two workdays.
Figure 2 presents the average vehicular speed for GRAMMA-VORAGINE in the valley hour
for daytime and afternoon-time on Wednesday from 9:10 am to 11:35 am and from 3:06 pm
to 4:45 pm. For the day, the average speed was 15.21 km/h, and the afternoon (called night
for this case) was 14.05km/h. The segment with the lowest speeds was ARQUITECTURA-
PANTANO DE VARGAS, while the one with the highest was CALLE 40 - CALLE 38. On the
other hand, it is graphically observed that the average speeds were similar for each segment in
GRAMMA-VORAGINE at Valley hour.
Figure 3 shows the average speeds by segments for GRAMMA-VORAGINE at Peak hour on
263
Carlos Alberto Díaz Riveros et al. CEUR Workshop Proceedings 259–270
Figure 2: Vehicular Average Speed in each segment for GRAMMA-VORAGINE path at Valley Hour.
Figure 3: Vehicular Average Speed in each segment for GRAMMA-VORAGINE path at Peak Hour.
Thursday from 7:30 am to 8:06 am, on Friday from 7:46 am to 8:15 am, and Tuesday from 11:46 am
to 12:16 pm, with an average speed of 11.02 km/h. They were taken on Thursday from 5:25 pm
to 6:35 pm to collect information at night peak hour with; as a result, the average path speed was
264
Carlos Alberto Díaz Riveros et al. CEUR Workshop Proceedings 259–270
Figure 4: Vehicular Average Speed in each segment for VORAGINE-GRAMMA path at Valley Hour.
9.45 km/h. The segment with the lowest speeds was ARQUITECTURA-PANTANO DE VARGAS,
while the one with the highest was GLORIETA LA GRAMMA-CALLE41. Additionally, it is
appreciated graphically that the speeds are similar for each segment for GRAMMA-VORAGINE
at peak hours.
Figure 4 exhibits the average vehicular speeds for VORAGINE-GRAMMA at valley hour, from
Calle 32 and Glorieta de la Grama on Wednesday from 8:30 am to 10:58 am, with an average
speed of 12.26 km/ h. The late valley hour was also taken on Wednesday from 2:45 pm to 4:49
pm with an average speed of 11.27 km/h. The segment with the lowest speeds was CALLE 35A
– CALLE 36, while the one with the highest speeds was PALACIO DE JUSTICIA – CALLE 35.
Furthermore, it is possible to observe a similar situation to path one, where the speeds are near
for all segments.
Figure 5 presents the average speed for VORAGINE-GRAMMA at peak hour, from Calle 32 to
Glorieta de la Grama, during the daytime on Thursday from 7:35 am to 8:12 am, on Friday from
7:50 am to 8:11 am, and Tuesday from 7:35 am to 8:15 am, with an average speed of 10.36 km/h.
For the night peak hour, the data was taken on Thursday from 5:32 pm to 6:37 pm and Friday
from 5:50 pm to 6:46 pm, with an average speed of 9.16 km/h. The segment with the lowest
speeds was CALLE 35A – CALLE 36, while the one with the highest speeds was PALACIO
DE JUSTICIA – CALLE 35. The results exhibit similar behavior to Path one; it is graphically
observed that the speeds are also near for each segment.
265
Carlos Alberto Díaz Riveros et al. CEUR Workshop Proceedings 259–270
Figure 5: Vehicular Average Speed in each segment for VORAGINE-GRAMMA path at Peak Hour.
3.3. Speed of Type of Vehicle
The estimation of the average speed by type of vehicle, namely Bus, taxi, and private car, is
presented in Figure 6 for the two paths under study, daytime and night and valley-peak hour.
Figure 6(a) Shows the average speed for GRAMMA-VORAGINE in the valley and peak daytime
hours. It is easily observed that the Bus shows the lowest speeds, and all vehicles transit slower
at peak hours than at valley hours. The same situation is present for the night, as shown in
Figure 6(b), where the average speed for GRAMMA-VORAGINE at the valley and peak hours is
illustrated. It is also possible to observe that there is no substantial difference between daytime
and night for average speed per type of vehicle.
On the other hand, Figure 6(c) shows the average speed for the path two VORAGINE-
GRAMMA at the valley and peak hours during the daytime, and Figure 6(d) for the valley
and peak hours at night. It is also observed that the bus is the slowest of the vehicles. Likewise,
it could be inferred that circulation is more complicated at night, and there are no crucial
changes between daytime and night for average speed per type of vehicle.
4. Related work
Contemporary urban planning strategies are varied in their efforts to design efficient and
balanced transportation systems [10, 11]. In particular, simulation-based approaches offer the
ability to proactively and quickly diagnose traffic safety problems and to evaluate appropriate
initiatives [12]. Many of these consist of microscopic traffic flow modeling, i.e., models for the
266
Carlos Alberto Díaz Riveros et al. CEUR Workshop Proceedings 259–270
Figure 6: Average Speed by Type of Vehicle. (a) Vehicular average speed in GRAMMA-VORAGINE
path for a valley-peak hour at daytime. (b) Vehicular average speed in VORAGINE-GRAMMA path
for a valley-peak hour at daytime. (c) Vehicular average speed in GRAMMA-VORAGINE path for a
valley-peak hour at night. (d) Vehicular average speed in VORAGINE-GRAMMA path for a valley-peak
hour at night.
study of vehicle traffic dynamics where single vehicle-driver units are simulated with their
relevant properties, such as position and speed, being represented as if these units were single
points [13, 14]. The advantages of microscopic modeling include the assessment of the future
design, the analysis of several aspects of transport infrastructure, etc. Monitoring systems,
such as those using aerial footage, can be used to automatically detect vehicle trajectories and
subsequently assess the effectiveness of public policy measures to regulate these [15]. Examples
of this include the use of microscopic traffic modeling in combination with data obtained from
this type of footage in order to quantify the effect of traffic-calming approaches, e.g., speed
cushions, with recent results showing that these measures are effective in reducing the influx
of vehicles on roads [16]. Real-time traffic data captured by these monitoring systems have
also been used in the development and continuous update of machine learning models for the
prediction of driver behavior. Examples of this include the use of these monitoring data for the
continuous update of the hyperparameters of deep learning prediction models allowing the
digital twin modeling of real-time traffic [17]. However, in spite of these and other significant
advances, the current literature reportedly presents a void in the use of this type of modeling
for the evaluation of traffic safety in environments that are commonly observed in emergent
countries [12].
Other approaches focus on macroscopic traffic modeling, rather than microscopic modeling,
with the difference between the two is that the latter considers the interaction between individual
vehicles whereas the former considers the aggregate behavior of traffic flow [18, 19]. This
macroscopic modeling paradigm is especially used when studying vehicle behavior at junctions,
which reportedly constitute the most critical point on the road network [20]. Automated
planning platforms using the macroscopic paradigm have been proposed for the implementation
267
Carlos Alberto Díaz Riveros et al. CEUR Workshop Proceedings 259–270
of efficient urban traffic control measures in various metropolitan areas. Domain-dependent
model update approaches have been proposed using learning techniques that monitor traffic
junctions and automatically generate actions that match the dynamics of the junction, thus
allowing automated planning platforms to quickly adapt to changing city traffic behavior,
generating better planning domains that require less model engineering work [21]. Similar
models in both paradigms have been used in the study of junctions controlled by traffic lights
resulting in methods for the better understanding and improvement of traffic at these critical
points of the road [20]. Examples of macroscopic modeling include the work of [22], where
traffic flow modeling is used in order to determine whether a roundabout in Hlohovec, Slovakia,
has sufficient spatial capacity for managing the traffic load of the town. Similarly, [23] use flow
modeling in order to assess the impact of buses on the traffic indicators in Melbourne, Australia,
showing that the congestion created by this type of public transportation is greatly outweighed
by their reduction of traffic in the city.
Many other contemporary models also focus on traffic conflicts, defined as situations involving
two or more drivers where at least one of these is forced to take evasive action in order to
prevent an accident with the others, e.g., where one driver has to swerve abruptly in order to
avoid a collision [24, 25, 26]. [27] propose a methodology for the assessment of human driver
behavior in urban scenarios where automated vehicles also use the roads, thus interacting with
their traditional, human-driven counterparts, and where traffic conflicts might arise.
Many research initiatives on traffic modeling require the acquisition of transit data, e.g., a
number of vehicles moving through a street or intersection within a time window and the
speed of each, a process that traditionally, and still frequently, is carried out mostly by manual
surveys, such as in the research conducted by [22], [23] and [27]. This, by itself, poses challenges
related to the reliability of the data and the associated costs of the manual process which in
recent years have increasingly been addressed through the growing use of computer vision
techniques, as these become more sophisticated, for the automated acquisition of traffic data
[28, 29, 30, 31, 32]. Additional challenges are introduced when studying the traffic of bicycles,
for instance, given that these are able to use routes other than the regular roads, including
cycle paths and passageways for pedestrians [33]. [34] propose a model for the study of bicycle
behavior in Warsaw, Poland, showing that, during rush hours, cyclists generally choose to use
main roads.
5. Conclusions
The results presented in this article evidence the feasibility of the proposed approach, namely
the license plate method, for the characterization of vehicle speed in the area under study, which
is a focal point in the city of Villavicencio and an attractor of traffic. This characterization,
discriminated by vehicle type, namely public transport vehicles and private ones, is expected to
contribute to the further understanding of the underlying causes of congestion at this point of
the city, especially how this varies by vehicle type between peak and valley (off-peak) times.
Future work contemplates the extension of the proposed approach and its validation in other
areas of Villavicencio and larger urban areas in the country.
268
Carlos Alberto Díaz Riveros et al. CEUR Workshop Proceedings 259–270
References
[1] Comisión de Transporte y Turismo, 2015, Informe sobre movilidad urbana sostenible, URL:
https://www.europarl.europa.eu/doceo/document/A-8-2015-0319_ES.html.
[2] G. Messori, E. Morello, E. Perotto, F. Infussi, G. Mondini, E. Faroldi, S. Tolentino, M. Ugolini,
Mobility management at politecnico di milano: new infrastructures and behavioural
change, IOP Conference Series: Earth and Environmental Science 1 (2019) 1–2.
[3] M. C. Sánchez, M. Davila, J. D. López, L. Gonzales, L. Cobo, C. Diaz, Towards the construc-
tion of a smart city model in bogotá, CEUR Workshop Proceedings (2020).
[4] Naciones Unidas, 2016, Agenda 2030 y los objetivos de desarrollo sostenible una oportu-
nidad para América Latina y el Caribe, URL: www.un.org/sustainabledevelopment/es.
[5] N. Ochoa, C. Diaz, M. Dávila, M. Martínez, O. Acosta, J. Rios, A. Rodriguez, G. Acuña,
A. Garcia, Towards the design and implementation of a Smart City in Bogotá, Colombia,
volume 93, Revista Facultad de Ingeniería Universidad de Antioquía REDIN, 2019.
[6] Secretaría de Movilidad Bogotá, 2016, Plan Maestro de Movilidad Bogotá, URL: https:
//www.movilidadbogota.gov.co/web/plan-maestro-movilidad.
[7] C. Díaz, K. Beltrán, C. Diaz, A. Baena, Traffic flow indicators analysis to determine causes
of vehicular congestion, ParadigmPlus 2 (2021) 1–16.
[8] C. A. Diaz Riveros, K. A. Beltran Rodriguez, C. O. Diaz, A. J. Baena Vasquez, Mobility
in smart cities: Spatiality of the travel time indicator according to uses and modes of
transportation, in: International Conference on Applied Informatics, Springer, 2021, pp.
433–448.
[9] R. Reyes Spindola, M. Cal, G. Cardenas, Ingeniería de tránsito, Fundamentos y Aplicaciones,
México:Alfaomega, 2007.
[10] A. Pell, A. Meingast, O. Schauer, Trends in real-time traffic simulation, Transportation
Research Procedia 25 (2017) 1477–1484.
[11] Y. Li, W. Tu, Traffic modelling for IoT networks: A survey, in: Proceedings of the 2020
10th International Conference on Information Communication and Management, 2020, pp.
4–9.
[12] S. Mahmud, L. Ferreira, M. Hoque, A. Tavassoli, Micro-simulation modelling for traffic
safety: A review and potential application to heterogeneous traffic environment, IATSS
Research 43 (2019) 27–36.
[13] R. K. C. Billones, A. A. Bandala, L. A. G. Lim, E. Sybingco, A. M. Fillone, E. P. Dadios,
Microscopic road traffic scene analysis using computer vision and traffic flow modelling,
Journal of Advanced Computational Intelligence and Intelligent Informatics 22 (2018)
704–710.
[14] D. Song, R. Tharmarasa, M. C. Florea, N. Duclos-Hindie, X. N. Fernando, T. Kirubara-
jan, Multi-vehicle tracking with microscopic traffic flow model-based particle filtering,
Automatica 105 (2019) 28–35.
[15] C. Yinka-Banjo, O. Daniel, S. Misra, O. Jonathan, H. Florez, Comparative analysis of three
obstacle detection and avoidance algorithms for a compact differential drive robot in v-rep,
in: International Conference on Applied Informatics, Springer, 2019, pp. 357–369.
[16] J. Paszkowski, M. Herrmann, M. Richter, A. Szarata, Modelling the effects of traffic-calming
introduction to volume–delay functions and traffic assignment, Energies 14 (2021) 3726.
269
Carlos Alberto Díaz Riveros et al. CEUR Workshop Proceedings 259–270
[17] H. Naing, W. Cai, N. Hu, T. Wu, L. Yu, Data-driven microscopic traffic modelling and
simulation using dynamic LSTM, in: Proceedings of the 2021 ACM SIGSIM Conference on
Principles of Advanced Discrete Simulation, 2021, pp. 1–12.
[18] Z. H. Khan, T. A. Gulliver, A macroscopic traffic model for traffic flow harmonization,
European Transport Research Review 10 (2018) 1–12.
[19] L. Adacher, M. Tiriolo, A macroscopic model with the advantages of microscopic model:
A review of Cell Transmission Model’s extensions for urban traffic networks, Simulation
Modelling Practice and Theory 86 (2018) 102–119.
[20] K. Čulík, V. Harantová, A. Kalašová, Traffic modelling of the circular junction in the city
of Žilina, Advances in Science and Technology Research Journal 13 (2019).
[21] A. Pozanco, S. Fernández, D. Borrajo, On-line modelling and planning for urban traffic
control, Expert Systems 38 (2021) e12693.
[22] J. Palúch, A. Kalašová, Traffic modelling on the roundabout in the town of Hlohovec and
using the information from the traffic counter, in: International Conference on Transport
Systems Telematics, Springer, 2018, pp. 129–141.
[23] D. Q. Nguyen-Phuoc, G. Currie, C. De Gruyter, I. Kim, W. Young, Modelling the net traffic
congestion impact of bus operations in Melbourne, Transportation Research Part A: Policy
and Practice 117 (2018) 1–12.
[24] C. Dong, R. Ma, Y. Yin, B. Shi, W. Zhang, Y. Zhang, Traffic conflict analysis of motor
vehicles and nonmotor vehicles based on improved cellular automata, Mathematical
Problems in Engineering 2020 (2020).
[25] M. Essa, T. Sayed, Traffic conflict models to evaluate the safety of signalized intersections at
the cycle level, Transportation Research Part C: Emerging Technologies 89 (2018) 289–302.
[26] L. Wanumen, J. Moreno, H. Florez, Mobile based approach for accident reporting, in:
International Conference on Technology Trends, Springer, 2018, pp. 302–311.
[27] A. Sharma, Y. Ali, M. Saifuzzaman, Z. Zheng, M. D. Haque, Others, Human factors in
modelling mixed traffic of traditional, connected, and automated vehicles, in: International
Conference on Applied Human Factors and Ergonomics, Springer, 2017, pp. 262–273.
[28] N. Saunier, T. Sayed, Automated analysis of road safety with video data, Transportation
Research Record 2019 (2007) 57–64.
[29] W. Hu, X. Xiao, D. Xie, T. Tan, S. Maybank, Traffic accident prediction using 3-D model-
based vehicle tracking, IEEE Transactions on Vehicular Technology 53 (2004) 677–694.
[30] S. Messelodi, C. M. Modena, A computer vision system for traffic accident risk measurement:
A case study, 2005.
[31] S. Atev, H. Arumugam, O. Masoud, R. Janardan, N. P. Papanikolopoulos, A vision-based
approach to collision prediction at traffic intersections, IEEE Transactions on Intelligent
Transportation Systems 6 (2005) 416–423.
[32] S. Kamijo, Y. Matsushita, K. Ikeuchi, M. Sakauchi, Traffic monitoring and accident detection
at intersections, IEEE transactions on Intelligent Transportation Systems 1 (2000) 108–118.
[33] J. E. Cantor, G. A. Montoya, C. Lozano-Garzon, Establishing best pedestrian paths con-
sidering sars-cov-2 contagions: Mathematical optimization model and mobile application
approach, ParadigmPlus 2 (2021) 14–36.
[34] M. Jacyna, M. Wasiak, M. Klodawski, P. Golkebiowski, Modelling of bicycle traffic in the
cities using VISUM, Procedia Engineering 187 (2017) 435–441.
270