Content uploaded by Ammar Bouchair
Author content
All content in this area was uploaded by Ammar Bouchair on Jan 12, 2023
Content may be subject to copyright.
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=tasr20
Architectural Science Review
ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/tasr20
Contribution of numerical simulation to the study
of pedestrian mobility in the context of COVID-19:
case of a university campus in Algeria
Mustapha Blibli & Ammar Bouchair
To cite this article: Mustapha Blibli & Ammar Bouchair (2023): Contribution of numerical simulation
to the study of pedestrian mobility in the context of COVID-19: case of a university campus in
Algeria, Architectural Science Review, DOI: 10.1080/00038628.2022.2160694
To link to this article: https://doi.org/10.1080/00038628.2022.2160694
Published online: 12 Jan 2023.
Submit your article to this journal
View related articles
View Crossmark data
ARCHITECTURAL SCIENCE REVIEW
https://doi.org/10.1080/00038628.2022.2160694
Contribution of numerical simulation to the study of pedestrian mobility in the
context of COVID-19: case of a university campus in Algeria
Mustapha Blibli and Ammar Bouchair
Research Laboratory: “Cadre Bâti et Environnement”, University Mohamed Seddik Benyahia, Jijel, Algeria
ABSTRACT
Corona Virus (COVID-19) is forcing us to re-examine our current travel patterns in order to adapt to social
distancing. For this purpose, simulation of pedestrian mobility remains relevant and can facilitate the
design of buildings and urban spaces in this newly emerging context. Under COVID-19, a university campus
in Algeria has been selected to demonstrate digital simulation for the study of pedestrian mobility. Infra-
works software founded on multi-agent simulation is used. Based on the establishment and comparison of
various scenarios, we are able to confirm the effects of the anti-pandemic measures on pedestrian behavior
in the studied area, such as social distance and the decline in student population, and the implementation
of new traffic plans to enhance working conditions on the university campus. The results show that this
tool enables the readjustment of space and people’s behavior so that the university activities carry on with
a minimum health risk.
ARTICLE HISTORY
Received 26 June 2021
Accepted 12 December 2022
KEYWORDS
Numerical simulation;
pedestrian mobility; social
distance; COVID-19;
University campus;
infraworks
Introduction
In January 2020, the COVID-19 epidemic started in China, and
quickly spread to more than 200 countries. This pandemic is
caused by a new coronavirus called Severe Acute Respiratory
Syndrome Coronavirus 2 (SARS-CoV-2), which has a very high
transmissibility. Restricting public mobility is a crucial public
health tool to control and prevent infectious respiratory dis-
eases (Dietz et al. 2020;Zhouetal.2020). After several months
of restriction and confinement of communities, either by reduc-
ing public transport and gatherings (Abdullah et al. 2021), clos-
ing schools and universities, or possibly working from home.
Many countries have begun the process of decontamination or
have scheduled the gradual reopening of public facilities. While
waiting for the discovery of remedies and vaccines against this
scourge, the relaunch of socio-economic life is conditioned by
prudence, respect for health protection measures and the social
and physical distancing prescribed by official national and inter-
national health authorities (Organisation mondiale de la Santé
2020).
Researchers and professionals from different disciplines have
been mobilized to meet the challenges of COVID-19 (Campisi
et al. 2022; Campisi et al. 2022). Architects, planners and builders
are called upon more than ever to participate in this struggle and
work together to find solutions and remedies that can alleviate
the situation. In the short term, they should redesign the exist-
ing infrastructures. In the medium and long term, they should
review the design methods and approaches, as well as the con-
struction and operation of new structures in order to cope with
possible future crises and plagues.
For this purpose, we must not only continue to design cost-
effective, safe, sustainable and energy-efficient buildings and
infrastructure (Bouchair 2014; Tebbouche, Bouchair, and Grimes
CONTACT Mustapha Blibli mblibli@univ-jijel.dz BP 98 Ouled Aissa, 18000 Jijel, Algeria
2017; Kaoula and Bouchair 2018; Kaoula and Bouchair 2020), but
also to take into account the important additional criterion of
ensuring the recommended social distance between users in
common spaces (Rosenfeld 2020). Hence the need to re-examine
the way people move around and use urban and architectural
space. The risk of infection is an additional parameter to the
complexity of building and managing space that should be well-
thought-out (Figure 1).
To resume work in the context of COVID-19, and minimize the
spread of the virus, it will be necessary to ensure compliance
with the instructions for preserving people’s health by wear-
ing masks, periodically washing hands and disinfecting furniture
and any object likely to be touched, avoid proximity and respect
the required distance (Dietz et al. 2020). This last parameter is the
one that this study is interested in. It varies from country to coun-
try, but many of them have agreed on a range that lies between
1.5 and 2.0 m (Jones et al. 2020).
Maintaining the distance between people in institutions and
public spaces will generate additional needs in terms of surface
area. This means that the usual ratios (number of people per
square metre) in the programming of buildings must be revised
downwards, as well as the lengthening of distances and areas
of circulation and the adaptation of users’ behaviour. For this
purpose, the study of pedestrian mobility and its digital simu-
lation can be a good solution and a good support to facilitate
mastering the design of buildings and urban spaces.
The crisis of the COVID-19 is pushing towards the review of
our existing travel patterns and to find new ways of moving
around in order to best adapt to the new constraint of social dis-
tancing. In other word, how can this be achieved? Some scenar-
ios may be necessary to answer this question and to be prepared
for all eventualities.
© 2023 Informa UK Limited, trading as Taylor & Francis Group
2M. BLIBLI AND A. BOUCHAIR
Figure 1. The World Health Organisation (WHO) has prescribed that an interpersonal distance of 1.5–2m (about 6 feet) be maintained to minimize the risk of infection.
This distance is an upper limit for saliva droplets from a human cough that cannot travel more than 2m through space at approximately zero wind speed – Drawing from
(Ciric 2020). Adapted by the author.
The novelty of this work consists of the introduction of the
recent problem of COVID-19 in the architectural design of pub-
lic building environment such as the respect of social distance
capacity as a principal prescription facing this plague. COVID-
19 is a recent sanitary problem which appeared in January 2020.
Fewer studies were found in the literature and they are not well
developed. With the sudden appearance of such virus, the social
space practice changes and needs new requirements. As archi-
tects, we believe that the interaction between space and public
should be reviewed and new guidelines should be considered.
The purpose of this study is to show to what extent digital simu-
lation of pedestrian mobility can be a suitable solution and aide
to support the design of buildings and urban spaces.
State of the art
There was no research on the relationship between pandemic
and public space design until 2020. The only concerns were
about biomedical subjects (Zhou et al. 2020) and microbiol-
ogy. However, fewer efforts are being made to reduce disease
transmission to occupants through the built environment (Dietz
et al. 2020). Some studies recommend more rigorous behaviour
as a prevention strategy during the de-confinement process
(Meloche-Holubowski 2020). Ciric (2020) has defined a social dis-
tance that allows one to avoid contamination and its variability
in several countries (Ciric 2020). The impact of confinement on
air pollution was studied by (Aloi et al. 2020; Patrick Hertzke, Neu,
and Weaver 2020) and shows its advantages in the protection of
the environment. Moreover, the impact of human behaviour on
energy consumption due to confinement was given an in-depth
reflexion by (Ateek 2020).
Some have even gone so far as to say that the pandemic
could bring the final change in the way technology is used
in smart cities (Sophie Reynolds: Peter Baeck 2020). It will
challenge universities around the world to improve the quality
and relevance of their teaching in all forms (Witze 2020). It will
bring about changes in the world of work (Astroza et al. 2020;
Kramer and Kramer 2020). Finally, there are many reflections on
the current crisis and its influence on the spatial organization of
countries at different spatial scales, international, national and
local (Decoville 2020).
The architectural design studio ‘Shift’ in Rotterdam, has cre-
ated a concept that can be implemented during confinement to
ensure food distribution with minimal risk. By limiting travel in
the city and physical contact, open-air fresh food markets can be
set up to reduce the pressure on supermarkets (Harrouk 2020).
But this was not validated by a research study.
Recently, a work presented how the number of simulated
clients (occupancy) affects the social distance in an ideal super-
market. The study defines a social distance coefficient that
informs how many events suffer each agent in the system
(Parisi et al. 2020). Most of the publications concern operational
research for professional purposes as in the case of the French
group Engineering Systems International (ESI), who simulated
different work configurations to determine the best organiza-
tions to limit the contamination of employees. Or the appli-
cations of mobility simulation software providers (Stifter 2020)
who are innovating by offering tools that help cities to plan traf-
fic with these new constraints. For example, Planung Transport
Verkehr (PTV)-Group’s Vissim, Bentley’s Legion and Autodesk’s
Infraworks (Rosenfeld 2020).
Based on observations and knowledge gained about people’s
movements, behaviours, and interactions with one another and
with the environment, know-how could be developed for the
design of the crowd and pedestrian mobility simulators aimed
at reproducing the movement of real human beings with max-
imum fidelity and anticipating interaction behaviour between
individuals (Moussaïd et al. 2010). However, other researchers
also study the interaction of the crowd with the physical envi-
ronment and traffic (Olivier et al. 2018).
These crowd simulators can be divided into two main cate-
gories of algorithms:
- The macroscopic approaches model the crowd as a compress-
ible active substance (Hughes 2003; Treuille, Cooper, and
Popović 2006). The calculation of the variations of this den-
sity in space and time according to the continuum principle
allows an estimation of a collective movement. The macro-
scopic algorithms are fast and tend to take into account
only uniform crowds.
- Microscopic approaches allow a more accurate simulation of
individual behaviour. They follow the paradigm of complex
ARCHITECTURAL SCIENCE REVIEW 3
Figure 2. Presentation of the Jijel campus (a) location within central university campus (b) 3d model. Source: authors.
systems, where the behaviour of the group is the result of
the combination of interactions between individuals. This
approach is mainly based on the principle of the local inter-
action model, which can be done in several ways. The first
is the cellular automaton, which consists of discretizing the
individual movement in space and time, and simulating
the agent’s movement as a pawn on a chessboard (Blue
and Adler 2001; Schadschneider 2002). The second is the
social forces model (Helbing and Molnar 1995) which can
be used to calculate the continuous trajectories of agents.
Inspired by Newton’s 2nd law to simulate the movement of
agents attracted by forces towards their goals, and repelled
by other forces to avoid collisions between them.
In this study, we try to find out how to readjust the space
and people’s behaviour so that the university activities can
continue to be performed with a minimum health risk. Multi-
Agent Simulation is used which, allows the combination of
microscopic and macroscopic dynamics of pedestrian move-
ments. It is well adapted to the type of movement concerned
by this research. Hence the choice of the simulation tool ‘Mobil-
ity Simulation’ from Autodesk Infraworks among many others
on the market that use this paradigm. The university campus
of Jijel represents a typical situation to verify our hypothe-
ses by numerical simulation and comparison of the circulation
of students and staff in normal conditions and in the context
of COVID-19.
Methods and materials
Presentation of the university campus
The University of Mohamed Seddik Ben Yahia (MSB) is located in
the city of Jijel, Algeria. The university is spread over two sites;
the central university pole of Jijel and the one of Tassoust. The
first site concerns our work (Figure 2(a)). Our study area includes
the central part (see Figure 2(b)) with the rectorate building, the
central library, the two teaching blocks, the four amphitheatres
and the research laboratory block, which will be referred to in
the following as the Jijel Campus. The latter is shared between
the students of the three faculties of Science and Technology,
Natural and Life Sciences and Exact and Computer Sciences
(Table 1).
The Campus represents the heart of the central pole of the
University of Jijel and is used by the three faculties in addition
to the infrastructure of each one, where we estimate that in
ordinary conditions, 40% of the total number of students of the
pole of Jijel use it at the same time. That is to say an approximate
number of 3060 students (Figure 3).
Mobility simulation tool
The pedestrian mobility was simulated using the ‘Autodesk Infra-
works Mobility Simulation’ software version 6.0. It is an inte-
grated multimodal mobility simulation engine for Infraworks.
This latter is an Autodesk platform for modelling infrastructure
from real-world data in real time. It enables better creation, visu-
alization, analysis, sharing and management of information and
decision-making in this context. ‘Mobility Simulation’ is a multi-
agent simulation or agent-based model (Hermellin, Michel, and
Ferber 2015).Insimpleterms,anagentisapersonoravehi-
cle. A numerical simulation generates a sample of representa-
tive scenarios for a model in which a complete enumeration of
all possible states would be prohibitive or impossible. Graphi-
cal computer simulations have the advantage of visualizing the
system as it evolves over time.
While mobility simulation is a highly specialized field in
vehicle design and multi-modal transport, depending on the
number of transport modes, passengers and stops in the jour-
ney, modelling mobility for different design proposals can be
an extremely complex or very simple task (such as simulating
pedestrians moving through indoor or outdoor spaces). This
task becomes much more fun with Autodesk Infraworks with-
out the need for users to have extensive transport engineer-
ing skills (Rosenfeld 2020). In fact, once we have learned the
basics of pedestrian mobility simulation with this program, then
it will be easier to refocus our simulation objectives in the con-
text of COVID-19. The analysis by the Mobility Simulation soft-
ware systematically goes through the following stages (Duncan
2017):
(1) Build the network – buildings, roads, pavements and
crossings.
(2) Adding zones for the origin/destination of people.
(3) Define origin-destination demand matrices (directed) or
volumes (undirected).
(4) Generate trips on demand.
(5) Run a simulation and collect results.
(6) Results analysis (Figure 4).
4M. BLIBLI AND A. BOUCHAIR
Tab le 1 . Statistics of the central university campus for Jijel between 2019 and 2020.
Faculties Abbreviation Number of students Bachelor’s degree Number of students Master’s degree Total faculty
01 Science and Technology S. T 1705 799 2 504
02 Natural and Life Sciences N.S. V 2445 625 3070
03 Exact Sciences and Computer E.S.C 1529 541 2070
Total 5679 1965 7644
Source: Statistics and Forecasting Service of Jijel University.
Figure 3. Master plan of the Jijel Campus (Source: Architect’s Office CHOUIKI Seddik).
The main point for this process is that the Trips must be gen-
erated from the Demand before the simulation begins. Without
trips generation, we will not be able to run the simulation. Each
generation of Trips is called a Trips Page. All the randomness
required to effectively simulate people and traffic is contained
within the Trips page.
A Mobility Simulation model is saved as a single file with
extension ‘.aza’. This file contains all the input components,
including:
- Demand: one or more tables of combined travel demand, defin-
ing how many people want to travel between origin and
destination, the time profile of this demand and a distribu-
tion by person-type.
- Trips: a disaggregated list of trips, defining the Agent charac-
ter, origin, destination and departure time of every trip for
a particular scenario.
- Parameters: defining how people behave in the model, for
example on mode choice, route choice and speed choice.
- Network: paths and roads representation used by the people in
the model and the geographical areas defining their origins
and destinations.
- Validation: the tables of observations that are used to validate
the performance of the model. These may take many forms,
but typical examples include turn counts and travel times.
The menus in Mobility Simulation are laid out according
to these Components and Tables – the interface has been
Figure 4. Software simulation process « Infraworks Mobility ». Source: authors.
ARCHITECTURAL SCIENCE REVIEW 5
Figure 5. Levels of service of pedestrian crossings (Fruin 1971).
structured to provide a guide to the best order of defining the
data for the model.
This application allows us to create vehicles, people, parking
and transit mode simulations. Key performance indicators, such
as people/hours spent travelling, people/km travelled, level of
service (LOS), which we will develop in the next paragraph, as
well as economic and environmental evaluations, are available
in Excel and PDF reports.
Level of service
The LOS is a standard qualitative indicator used to describe the
flow characteristics in a pedestrian environment. In the Autodesk
simulator, there are several ways to define the LOS of pedestrian
crossings and pavements; we can define user levels ourselves
which requires a lot of work and time; choose the principles of
the HCM (Highway Capacity Manual, which is an American trans-
port institution) which are much more suitable for pavements
and crossings or take the levels of service defined by Fruin (Fruin
1971) as a reference, which are more suitable for our model of
pedestrian mobility isolated from mechanical lanes. These levels
of service characterize the quality of traffic on a navigable surface
according to its level of congestion (Georgiou et al. 2021).
Based on empirical observations, it defines different levels of
service on a scale from A to F (Table 2and Figure 5). Any sur-
face can be assigned a level according to its area and density.
The value of the level can, for example, be used to predict the
speed of pedestrians on that surface (Nag et al. 2020). The level of
service for a walkway is measured either by the number of peo-
ple per unit area on the walkway (the density) or by the average
speed of walking. LOS S is the level of service based on space.
LOS V is the level of service based on speed.
Study design
After six months of restrictions, it is necessary to resume a normal
socio-economic life while putting forward measures of adap-
tation to the requirements of the fight against the COVID-19
Tab le 2 . Corresponding measures at each level of service pedestrian crossings
(Fruin 1971).
LOS Den sit y p/m2Espace m2/p Flow rate p/min Speed m/s
LOS A ≤0.27 >3.24 ≤23 ≥1.3
LOS B 0.31–0.43 2.32–3.24 23–33 1.27
LOS C 0.43–0.72 1.39–2.32 33–49 1.22
LOS D 0.72–1.08 0.93–1.39 49–66 1.14
LOS E 1.08–2.17 0.46–0.93 66–82 0.76
LOS F >2.17 ≤0.46 Variable ≤0.76
pandemic . At the level of the university establishments, the
resumption must not be done at the expense of the health
of the students, teachers and workers. For this reason, the fol-
lowing measures ruled by the authorities and taken within the
universities and national schools are essential:
- Maintaining and strengthening distance learning (EAD)
- Organizing face-to-face teaching by waves of students and by
blocked periods e.g. reserving 2 weeks of teaching for stu-
dents in the 1st year of the bachelor’s degree, then 2 weeks
for those in the 2nd year and finally 2 weeks for the 3rd
year and those in the master cycle. This proposal has the
advantage of regulating the flow of students and allowing
compliance with health standards!
- Cleaning and disinfection of premises and equipment
- The application of barrier measures
- Training, information and communication.
- Limiting the mixing of students
- Maintaining physical distance
This crisis is pushing us to review our existing travel patterns
and find new ways of moving in order to best adapt to social
distancing. The study of mobility in our campus according to
different states (before COVID-19, during COVID-19 . .. etc.) and
their comparison thanks to the numerical simulation tool would
help us a lot to deal with these new constraints. In this regard, in
this study we will model the existing travel pattern with a pre-
pandemic scenario, on the one hand, and imagine two other
scenarios of the same pattern in the context of COVID-19, the
first one with the application of the social distance (1.50 m),
the second one with the application of the wave system (the
reduction of the student population to one third) and on the
other hand, model an exemplary travel pattern (with one-way
direction, no crossing of flows and control points), as follows:
(1) Establish a model according to the traffic pattern in the cam-
pus before the pandemic, i.e. under normal conditions (Jijel
Campus Model) and generate two scenarios according to
the needs of this study which we will call; before COVID-19
and COVID-19:
(A) Pre-COVID-19 scenario: this represents travel as it was
before the pandemic, based on the existing spatial con-
figuration (free, unsignalled traffic with no application
of social distancing) and the maximum number of users,
i.e. with ordinary capacities (normal person-to-square-
metre ratio) (Figure 6).
(B) Scenario COVID-19 with social distance: With the same
traffic scheme, we will test a second scenario by chang-
ing only the spacing parameter between the agents to
6M. BLIBLI AND A. BOUCHAIR
Figure 6. Views showing the use of the courtyard and the esplanade of the central campus before COVID-19. Source: authors.
1.5 m (social distance as a parameter discussed in this
study), with the aim of seeing the impact of distancing
on students’ travel behaviour, observing the interac-
tions between them (agents) and with the environment
(campus traffic areas)
(C) Scenario COVID-19 with social distancing and wave sys-
tem: once again, with the same traffic pattern, we will
test a third scenario by applying social distancing and
the wave system prescribed in the local protocol of the
university, which reduces the student population of the
campus to one third (1/3). This scenario also aims to ver-
ify the effect of the reduction of the number of users on
their trips with social distancing in comparison with the
previous scenario (Figure 7).
The comparison of the scenarios; (a) before COVID-19 and
(b) COVID-19 with social distancing will allow us to see in the
first place if the social distance (1.5 m) has an impact on the
movements of the users of the campus, that is to say by the
fact that if people want to keep a distance between them, will
slow them down and create congestion and delays, or show
us the incapacity of respecting this distance which is a sani-
tary safety measure hence the risk of contamination. The com-
parison of scenario (C) COVID-19 with social distancing and
the wave system with (b) will tell us if the application of the
wave system by reducing the number of people would reduce
congestion and delays and allow the social distance to be
respected.
(2) Reflect on improving the traffic pattern by establishing one
or more movement models according to the prescriptions
of the different health protocols (Ministry framework, local
university or another one disseminated by an international
organization such as the WHO), as in the framework of this
study, we will opt for a traffic model scenario with uni-
directional flows, without crossing paths and flows, with
the possibility of control points (stops), the application of
the protocol’s wave system (reduction of the number of
students to one third) and also the respect of the social
distance.
The simulation of this scenario (COVID-19 protocol) will allow
us to verify the impact of the movement scheme (arrange-
ments and traffic plans) on the fluidity of traffic and mobility in
the campus by reducing congestion and delays, while allowing
users to respect the social distance imposed by the pandemic
(Figure 8).
Modelling and simulation
In this section we will describe the application of the main phases
of the modelling and simulation process seen previously on our
case study ‘University Campus of JIJEL’
Network of buildings and roads
As a result of our experience as teacher-researchers in the cam-
pus we have been able to establish a traffic diagram representing
the ordinary situation on the basis of which our network was
built (Figure 9), the most important flows emanate from three
(03) places which are the origin zones that we have designated
as:
(1) The West entrance and personal parking
(2) The East entrance (Coach) and residences
(3) The Esplanade and central library.
These flows converge on the buildings that represent the
destination zones of the model
(1) Rectorate.
(2) Pedagogical blocks 4 and 5.
(3) Research laboratories
(4) Amphitheatres I, J, K, L.
The connections between these zones are the pedestrian
crossings (courtyard, pavements and paths)
Denition of origin-destination demand matrices
Parameters
In the parameter menu, we define the duration of the simula-
tion (departure and arrival times, days of the week, etc.). The
types of people who make up the agents of the model, or from
Table 1, we can parameterize 04 types of people; the students
of the faculty of science and technology (S.T) in blue, those of
the sciences of nature and life (S.N.V) in green, the exact sciences
and Computer Science (S.C.I.) in red, and lastly, the personnel in
yellow.
There are several parameters that we can modify to change
the way pedestrians move and interact with others in the model
(Figure 10):
- Parameters / People / Space: The ‘personal space’ parameter
by the type of person only defines the minimum distance
ARCHITECTURAL SCIENCE REVIEW 7
Figure 7. Preparation for the resumption of work at the Jijel Campus by applying the directives of the Protocols of the Ministry and the MSB University (a) postingof
informative posters, (b) disinfection of the amphitheatres, (c) reduction of the number of pedagogical seats, (d) and (e) printing of signs indicating the directions of
circulation on the ground. Source: authors.
Figure 8. Proposal of a traffic plan in the Jijel Campus according to the local protocol of the university. Source: Architect’s office CHOUIKI Seddik. Adapted by the authors.
8M. BLIBLI AND A. BOUCHAIR
Figure 9. Interface of the ‘mobility simulation’ software workspace and layers menu to manage the network modelling with studied pedestrian walkways (2, 2b, 3,3b
and 5). Source: authors.
Figure 10. The physical distance setting in the Autodesk simulator in different ways; person >Size >Space and Walkway >Adjust >Space factor or >Capacity.
Source: authors.
that this person will seek to create between him or herself
and another person in front or to the side.
- Walkway / Adjust / Space factor: This multiplicative factor
(default 1.0) modifies the space parameter for this walkway.
- Walkway / Adjust / Speed Factor: This multiplicative factor
(default 1.0) modifies the speed at which people walk on a
walkway.
- Walkway / Adjust / Capacity: This function controls the num-
ber of people allowed to walk on a walkway.
O-D directed demand matrices
The directed demand is a set of trips, where each trip has an ori-
gin destination. This demand is an input to the model so-called
Origin-Destination matrix (OD matrix).
In our simulator, a matrix can be for people, vehicles or
freight. The people matrix allows each person to use all the
modes of transport available to them, so there is the possibility
to enter several people matrices in one request (Figure 11).
The models and scenarios
For the modelling we refer to the previous section of the
study design to carry out the simulation models and scenarios
(Table 3).
Model 1:
(A) Scenario before COVID-19.
(B) Scenario COVID-19 with social distance.
(C) Scenario COVID-19 with social distance and wave system.
ARCHITECTURAL SCIENCE REVIEW 9
Figure 11. Origin and destination Request Matrix settings window. Source: authors.
Tab le 3 . Summary of possible models and simulation scenarios.
Modelling Simulation
Models Description Scenarios Setting
Model 1 According to the circulation pattern in
the pre-pandemic campus, i.e. under
normal conditions (Jijel Campus
Model)
A-Pre-COVID-19 scenario - Ordinary parameters
B-COVID-19 scenario with social
distance
- Parameter person space
- Walkway space factor
C-COVID-19 scenario with social
distance and wave system
- Parameter person space
- Walkway space factor
- Request: divide the number of
people/3
Model 2 Establish an ideal or exemplary
model with the application of the
maximum requirements of the =
protocols, examples:
-Traffic plan with unidirectional flows
-avoid crossings
A- Scenario with local protocol
prescription (University of Jijel)
- Set space person
- Walkway space factor
- Request: divide the number of
people /3
- Temps d’arrêt aux points de con-
trôle
B- Scenario with framework protocol
prescription (Ministry)
Parameters according to the study needs . . .
C-Scenario with framework protocol
prescription (WHO)
Parameters according to the study needs
Model n . .. A. Scenario Parameters according to the study needs . . .
B. Scenario Parameters according to the study needs
C. Scenario Parameters according to the study needs .. .
Model 2:
(A) Scenario with framework protocol prescription (University
of Jijel).
Results
Real-time numerical simulation generates a sample of represen-
tative scenarios for a model in which a complete enumeration
of all possible states would be prohibitive or impossible. Graph-
ical simulations have the advantage of visualizing the system as
it evolves over time (Figure 12(a)).
For this purpose, the execution of the simulations of scenar-
ios A, B and C of model 01 and that of scenario A of model 2,
allowed us to appreciate, analyse and understand the behaviour
of pedestrians on the campus for the different cases estab-
lished according to the objectives of this study. The simulator
offers us graphic and alphanumeric results in real time, either
global (averages) or detailed by walkway or by agent (Table 4).
The scenarios can be viewed continuously (real time) or in a
step-by-step mode (sequences), which is better suited to the
presentation of articles and publications.
In the simulator, the flattened view display of the minimum
distance an agent will seek to create between him or herself and
another in front or to the side is illustrated by a semi-circle in
front of the person. The radius of the semicircle is the size of the
agent added to the distance of the personal space. If a person is
forced to accept a smaller distance, by another person moving
into his or her space, the semicircle changes colour to yellow.
For this purpose, for model 1 and scenario A the distance
parameter is set as default, i.e. the height of the person (r=0.3),
in scenario B the social distance is added (R=0.3+1. 5/2 m) and
this scenario showed a lot of areas where this distance could
not be respected (yellow semicircles, Figure 13), in scenario
C the reduction of the number of staff mitigated this finding
and allowed more social distance. Finally, the proposal of the
10 M. BLIBLI AND A. BOUCHAIR
Figure 12. Mobility simulation presentation window (a) 3d solid view and (b) 2D flattened view with coloured LOS representations of the pedestrian walkways. Source:
authors.
Tab le 4 . Report example for some pedestrian routes of the scenario before COVID-19.
Origine Destination Number (u) Time (h, m, s) Distance (m) speed (m/s)
West entrance and staff car parks BLOCK 04 342 0:01:03 75.6 4.28
Esplanade and central library BLOCK 05 287 0:01:16 94.1 4.4
East entrance (Bus) and residences Rectorate 8 0:01:29 110 4.39
West entrance and staff car parks Amphitheater I 36 0:01:19 94.4 4.28
West entrance and staff car parks Rectorate 31 0:00:35 43.6 4.39
Esplanade and central library Amphitheater K 30 0:01:04 80 4.49
East entrance (Bus) and residences BLOCK 04 577 0:01:47 132 4.44
West entrance and staff car parks BLOCK 05 324 0:02:20 161 4.12
West entrance and staff car parks Laboratories 57 0:01:42 115 4.04
Esplanade and central library Amphitheater L 2 0:01:11 85.2 4.3
Figure 13. Screenshot of the COVID-19 scenario with physical distance at an instant t°, in circular areas the semicircles of the personal space indicate that the social
distance could not be respected. Source: authors.
new traffic model in the COVID-19 protocol scenario, the traffic
became more fluid and the distance between agents was well
respected (yellow semicircles have almost disappeared).
The Level of service (LOS) is a measure of performance,
usually by dividing all possible measurements into bands, and
assigning a single letter to each band. From the highest letter A
to the lowest letter F this classification is graphically supported
by a colour gradient from green A to orange F. During the simu-
lation this tool (LOS) displays the results in real time for the whole
duration. By checking the ‘average’ box in the menu one will
have average results for each route of the duration of the sim-
ulation. It is also possible to visualize the LOS in the graphical
window by checking the ‘Pedestrian route colours’ box, which
will highlight the observed pedestrian crossings with the colour
corresponding to their LOS (Figure 12(a)).
The results obtained from the average LOS V (velocity) of
model 01 show relatively low values for scenario A (Figure 14(b)).
Out of a total of 23 walkways, we get 5 in level E (Table 5
in appendix). For scenario B, the values have decreased fur-
ther (Figure 15(b)). We get 13 walkways in level E and the
appearance of 2 walkways in level F (Table 6 in appendix).
For scenario C an improvement is observed (Figure 16(b)). we
get 6 walkways in level E (Table 7 in appendix). For the LOS
S (space), the values are clearly better as shown in Figures
14–16(a), almost walkways are in green (see Tables 5–7 in the
appendix).
For model 02 COVID-19 protocol scenario, we have average
LOS V levels of service (velocity) as shown in Figure 17(b). Out of
a total of 49 walkways, we get 4 walkways in level E (Table 8 in
appendix). For LOS space we get all in green (Figure 17(a)).
In order to have precise results, we chose five pedestrian
walkways that had the most unfavourable LOS V and that coin-
cided with the central courtyard (Figure 9). We asked for sim-
ulation results every 5 min over the duration of the simulation
(45 min) and exported in excel tables the values of two deter-
mining parameters of these levels of service, namely density and
ARCHITECTURAL SCIENCE REVIEW 11
Figure 14. Scenario before COVID-19 of model 1, plan diagrams of levels of service: (a) LOS S, (b) LOS V. Source: authors.
Figure 15. Scenario COVID-19 Social distance of model 1, plan diagrams of levels of service: (a) LOS S, (b) LOS V. Source: authors.
Figure 16. Scenario COVID-19 Social distance and waves of model 1, plan diagrams of levels of service: (a) LOS S, (b) LOS V. Source: authors.
Figure 17. Scenario COVID-19 Model 2 Protocol, plan diagrams of levels of service: (a) LOS S, (b) LOS V. Source: authors.
12 M. BLIBLI AND A. BOUCHAIR
Figure 18. Evolution of densities and speeds of the walkway 2 over time and Comparison of the Model 1 Scenarios. Source: authors.
Figure 19. Evolution of densities and speeds of walkway 2 over the duration of the COVID-19 Scenario protocol. Source: authors.
speed for the three scenarios of the model 01. Then, a parame-
ter graph was generated for the three scenarios for comparison
purposes. The same approach was made for the COVID-19 proto-
col scenario of the proposed traffic model. The numerical results
are shown in Tables 9–14 (see appendix). If the average results
of the LOS of model 01 showed clear differences between the
scenarios, the values of densities and speeds do not keep these
differences during the whole duration of the simulation. Accord-
ingly, most of the curves intersect in several times because of the
random distribution of the agents and their non-homogeneous
distribution in time and space as shown in Figure 18 and Figures
20–23 in appendix. For most of the walkways, the density curve
for the scenario before COVID-19 is in the middle that for sce-
nario B (with social distance) is at the top (high densities) and for
scenario C (social distance and wave system) the curve is at the
bottom ( <1 m/s), for the displacement speed curves which are
inversely proportional to the densities (congestion), the curve for
scenario B is at the bottom, those for A and C are almost merged
at the top of the graph ( ∼1 m/s) with a slight upper shift of the
speeds for scenario C.
For the COVID-9 protocol scenario we have density curves
quite close to the previous scenarios and this is explained by the
fact that the surface of the walkways has been divided in two
to have unidirectional paths and acceptable speeds that allow
for traffic flow (1 <V<1.5 m/s). Figure 19 and Figures 24–27 in
appendix show the trends of the density evolution and walkway
velocities over the duration of the COVID-19 Scenario of proto-
col. Tables 14–18 in appendix present numerical values of these
figures.
Discussion of the results
The results of the simulations of the different scenarios gave a
clear idea of the pedestrian mobility on this campus. Scenario A
of the model 1 represents the state of the campus before COVID-
19, where the maximum number of students is more than 3000.
These students use the space simultaneously where overcrowd-
ing is felt at the central courtyard due to the coming flows from
all sides (Figure 14). Despite of low level of services, this situation
was considered as normal before COVID-19. Scenario B, which
represents the state where the same layout, the same number
of students and applied the social distance (1.5 m) the simula-
tion showed that it would be very difficult to respect the social
distance with the same data. This was confirmed by viewing the
plane view where the distance of each agent is represented by
a white semicircle that changes colour to yellow when this gap
is reduced, on the one hand, the levels of service of this scenario
were even lower (Figure 15), Scenario C which stipulates reduc-
ing the number of students to one third (Wave System) showed
a clear improvement where the application of social distancing
was better (less yellow half circles), on the other hand, the levels
of service were higher (Figure 16).
In order to improve as much as possible the function-
ing and the mobility in the campus by decongesting it even
more while respecting the maximum of constraints imposed
by the pandemic, we proposed a slightly more elaborate traf-
fic plan (model 02) respecting the maximum of the prescrip-
tions of the local protocol of the university with unidirec-
tional flows, by avoiding the crossing of routes and flows, and
ARCHITECTURAL SCIENCE REVIEW 13
also by providing access control points (with adjustment of
the stopping time), while reducing the number of students
to a third following the wave system of the protocol in order
to have a respected social distance. The simulation of this
scenario gave very good results as shown in Figure 17.We
noted a good fluidity of traffic, the spacing between agents is
well respected and also the levels of service were practically
good.
The results showed, on the one hand, that by keeping the
same model of movement and changing the variable of dis-
tance the traffic becomes more and more difficult (congestion,
delay and transgression of this distance). On the other hand,
the variable of the number of staff or its reduction allowed a
clear improvement of the traffic, but which remains incomplete.
The new proposed mobility plan, which takes into account the
imposed prescriptions of the protocol, gave better results and
better traffic flow.
In the light of the results, the simulation of pedestrian mobil-
ity has shown that in order to have a fluid mobility with the
capacity to respect the social distance, it will be necessary to
work on all these parameters; the reduction of the number of
people, the revision of the movement plans by favouring the
one-way flows and the lengthening of the distances and there-
fore the times.
To integrate the social distance in our designs, the numer-
ical simulation of pedestrian mobility represents an effective
tool to verify this criterion. Architects and building engineers
should consider health risks due to COVID-19 and similar pan-
demics in their building and town planning design. Public health
professionals and space designers should come together to
find better solutions and healthier cities during any sanitary
crisis.
The case of conversations or silently resting in the passages is
not considered here. We have simulated the students in move-
ment where the speed variable is the major factor. When the
speed is reduced to zero, the software will notify that LOS is low
and vice-versa.
The study of pedestrian mobility does not require advanced
skills in mobility and traffic for architects and builders. It remains
affordable especially with the simulator’s availability. Due to
such tools, we will be able to anticipate the movements in the
design.
The Covid-19 crisis has revealed the unpredictable vulnerabil-
ities of the built environment design and planning. Traditional
vision for the design and planning of space (indoors and out-
doors) should change to meet the requirement of the time.
Space should be designed to ensure public health especially
for the unexpected periods of pandemics. Designers, builders,
decision-makers and planers have to work together and create
a common strategy for these situations in the future especially,
perception, design, physical distancing, accessibility, mainte-
nance, connectivity and distribution in the urban areas and
cities. Built environment needs to be cleaned regularly espe-
cially high-touch surfaces like handles and furniture. Tools such
as Simulation Mobility of Infraworks can be developed and gen-
eralized to play key roles for various pandemic circumstances
within the city.
Conclusion
This work has shown that the problem of COVID-19 is not only
a question of human health but also it concerns building space
and how it should be readjusted in the future, where architects
and builders play a key role. The choice of the study of pedes-
trian mobility and its digital simulation has shown that it can be a
worthwhile solution and also a good support for facilitating and
controlling the design of buildings and urban spaces. It is pos-
sible to review existing travel patterns and to find new ways of
moving around in order to best adapt to the new constraint of
physical distance.
The results show that with the scenario before Covid-19
which stands in normal health conditions is no longer valid after
the sudden appearance of the new pandemic. This is confirmed
by the simulated results of the second scenario which applied
the social distance principle. However, applying the social dis-
tance with the same space layout without reducing the number
of students will cause a lower level of service and bad function-
ing of the campus. The second scenario showed a clear improve-
ment when reducing the number of students to one third (wave
system). Accordingly, the levels of service were higher. A better
scenario was proposed and showed a good fluidity of traffic, the
spacing between agents is well respected and the levels of ser-
vice were generally better. This was by applying unidirectional
flows and avoiding their crossings.
Due to the digital simulation tool for pedestrian mobility,
architects will be able to anticipate travel in their designs and
be able to revise certain layouts to meet the new constraints
imposed by COVID-19. The digital simulation has shown that in
order to have a good fluid circulation with the capacity to respect
the required social distance, it will be necessary to work on sev-
eral parameters such as: the reduction of the number of people
(ratio =person/m⊃2), the revision of the circulation plans by
favouring one-way traffic and by avoiding the crossing of flows
by the use of signalling or the construction of physical markers
in order to allow in fine the lengthening of the distances and the
time ranges.
This work was carried out in the difficult context of COVID-19.
Most of the exchanges and contributions by the authors were
made remotely, which lengthened the time of its realization.
We simplified the circulation networks by limiting them to the
external and horizontal displacements in order to facilitate and
accelerate the numerical simulations. We regret the impossibil-
ity to perform the validation of the simulations (scenarios) and
the calibration of the data against the real data of the campus
where the observation of these scenarios was almost impossible
during the pandemic.
These new working methods can be introduced into the train-
ing of architects and professional practice. They can be inte-
grated in the first phase of design at the sketch and preliminary
design level due to the simulation of user behaviour, in relation
to fire safety, crowd evacuation in the event of disasters and
social distancing as a new parameter to be taken into account.
These can be extended to school facilities, which require good
management of staff during this plague, and also to all public
facilities and outdoor urban spaces.
14 M. BLIBLI AND A. BOUCHAIR
In the new pandemic time, designers, builders and decision-
makers for the public space need to change their future vision,
tools and strategies to make the built environment (indoors and
outdoors) more secure and healthier. The relationship between
space and the pandemic challenges should be readapted to
ensure the well-being and public health. Numerical tool devel-
opment, rational design and planning, maintenance of public
space are key elements in fighting the spread of the Coronavirus
and similar viruses.
The major limitation was that only one case study for data
gathering was considered. The use of only one case study lim-
its the findings’ generalizability. The study could have been
stronger if more cases outside the one studied in our work were
added, so we could get a balanced view of the subject.
Another limitation is the lack of previous research studies on
the topic of COVID-19 related to architectural design. Such a lim-
itation prevents us from understanding thoroughly the environ-
ment of study because we would not have the scope of studies
to guide our current research.
We recommend those aspects of the methodology left out as
a suggestion for further study.
Disclosure statement
No potential conflict of interest was reported by the author(s).
ORCID
Mustapha Blibli http://orcid.org/0000-0002-6466-6085
Ammar Bouchair http://orcid.org/0000-0002-4193-1210
References
Abdullah, M., N. Ali, C. Dias, T. Campisi, and M. A. J. A. S. Javid. 2021. “Explor-
ing the Traveler’s Intentions to Use Public Transport During the COVID-
19 Pandemic While Complying with Precautionary Measures.” Applied
Sciences 11 (8): 3630. doi:10.3390/app11083630.
Aloi, A., B. Alonso, J. Benavente, R. Cordera, E. Echániz, F. González, C. Ladisa,
R. Lezama-Romanelli, Á López-Parra, and V. Mazzei. 2020. “Effect of the
COVID-19 Lockdown on Urban Mobility. Emperical Evidence from the City
of Santander (Spain).” Sustainability 12: 3870. doi:10.3390/su12093870.
Astroza, S., A. Tirachini, R. Hurtubia, J. A. Carrasco, A. Guevara, M. Munizaga,
M. Figueroa, and V. J. F. Torres. 2020. “Mobility Changes, Teleworking, and
Remote Communication During the COVID-19 Pandemic in Chile.” 13489.
Ateek, G. 2020. “Future of Sustainable Architecture: Rethinking COVID-19 a
Pandemic or turning point?”
Blue, V. J., and J. L. Adler. 2001. “Cellular Automata Microsimulation for Mod-
eling Bi-Directional Pedestrian Walkways.” Transportation Research Part B:
Methodological 35 (3): 293–312. doi:10.1016/S0191-2615(99)00052-1.
Bouchair, A. 2014. “Sustainability Features of Vernacular Architecture in
Southern Algeria.” In Vernacular Architecture: Towards a Sustainable
Future. C. Mileto, F. Vegas, L. García and V. Cristini, 183–188. London: CRC
Press.
Campisi, T., S. Basbas, M. Trouva, T. Papas, N. Akgün-Tanbay, N. Stamatiadis,
and G. Tesoriere. 2022. “Anxiety, Fear and Stress Feelings of Road Users
During Daily Walking in COVID-19 Pandemic: Sicilian Cities.” Transporta-
tion Research Procedia 62: 107–114. doi:10.1016/j.trpro.2022.02.014.
Campisi, T., G. Tesoriere, A. Skoufas, D. Zeglis, C. Andronis, and S.
Basbas. 2022. “Perceived Pedestrian Level of Service: The Case of
Thessaloniki, Greece.” Transportation Research Procedia 60: 124– 131.
doi:10.1016/j.trpro.2021.12.017.
Ciric, L. 2020, June 23. “Un mètre ou deux? La distanciation physique, selon
la science.” Accessed 15 August 2020. https://theconversation.com/one-
metre-or-two-the-science-behind-social-distancing-139929.
Decoville, A. 2020. “Peut-on apprendre quelque chose de la crise actuelle
pour mieux organiser notre territoire? Recueil de contributions COVID-
19.” Fondation Idea Asbl, 62–67.
Dietz, L., P. F. Horve, D. A. Coil, M. Fretz, J. A. Eisen, and K. Van Den
Wymelenberg. 2020. “2019 Novel Coronavirus (COVID-19) Pandemic: Built
Environment Considerations to Reduce Transmission.” mSystems 5 (2):
e00245–e00220. doi:10.1128/mSystems.00245-20.
Duncan, G. 2017.InfraWorks Mobility Simulation, User Manual, UK English.
Fruin, J. J. 1971. “Designing for Pedestrians: A Level-of-Service Concept.”
Georgiou, A., A. Skoufas, S. J. C. S. o, and T. P. Basbas. 2021.“Perceived
Pedestrian Level of Service in an Urban Central Network: The Case of a
Medium Size Greek City.” Case Studies on Transport Policy 9 (2): 889–905.
doi:10.1016/j.cstp.2021.04.009.
Harrouk, C. 2020. “Shift Architecture Urbanism Creates Hyperlocal Micro Mar-
kets that Operate During COVID-19 Shutdowns.” Accessed 4 September
2020. https://www.archdaily.com/936856/shift-architecture-urbanism-
creates-hyperlocal-micro-markets-that-operate-during-covid-19-shut
downs#.
Helbing, D., and P. Molnar. 1995. “Social Force Model for Pedestrian Dynam-
ics." Physical Review E 51 (5): 4282. doi:10.1103/PhysRevE.51.4282.
Hermellin, E., F. Michel, and J. Ferber. 2015. “État de l’art sur les simulations
multi-agents et le GPGPU Évolution et perspectives de recherches.” Revue
des Sciences et Technologies de l’Information-Série RIA: Revue d’Intelligence
Artificielle 29 (3-4): 425–451. doi:10.3166/ria.29.425-451.
Hughes,R.L.2003. “The Flow of Human Crowds.” Annual Review of Fluid
Mechanics 35 (1): 169–182. doi:10.1146/annurev.fluid.35.101101.161136.
Jones, N. R., Z. U. Qureshi, R. J. Temple, J. P. Larwood, T. Greenhalgh, and L.
Bourouiba. 2020. “Two Metres or One: What is the Evidence for Physical
Distancing in Covid-19?” BMJ 370 (m3223). doi:10.1136/bmj.m3223.
Kaoula, D., and A. Bouchair. 2018. “Evaluation of Environmental Impacts
of Hotel Buildings Having Different Envelopes Using a Life Cycle
Analysis Approach.” Indoor and Built Environment 27 (4): 561–580.
doi:10.1177/1420326X16683235.
Kaoula, D., and A. Bouchair. 2020. “Identification of the Best Material-Energy-
Climate Compatibility for Five Ecological Houses and the Contribution of
Their Impact Sources to the Overall Balance.” Sustainable Cities and Society
52: 101781. doi:10.1016/j.scs.2019.101781.
Kramer, A., and K. Z. Kramer. 2020. “The potential impact of the Covid-19
pandemic on occupational status, work from home, and occupational
mobility.” Journal of Vocational Behavior 119 (103442.
Meloche-Holubowski, M. 2020. “Respect des règles sanitaires : « Montre-
moi comment mes sacrifices portent fruit ».” Accessed 10 August 2020.
https://ici.radio-canada.ca/nouvelle/1717643/comportement-pandemie-
deconfinement.
Moussaïd, M., N. Perozo, S. Garnier, D. Helbing, and G. Theraulaz. 2010.“The
Walking Behaviour of Pedestrian Social Groups and its Impact on Crowd
Dynamics.” PLoS One 5 (4): e10047. doi:10.1371/journal.pone.0010047.
Nag, D., A. K. Goswami, A. Gupta, and J. J. T. R. Sen. 2020. “Assessing
Urban Sidewalk Networks Based on Three Constructs: A Synthesis of
Pedestrian Level of Service Literature.” Transport Reviews 40 (2): 204–240.
doi:10.1080/01441647.2019.1703841.
Olivier, A.-H., J. M. Auberlet, A. Dommes, M.-A. Granie, L. Hoyet, and J. Pet-
tré. 2018. “Etude du piéton en réalité virtuelle: état de l’art, enjeux et
perspectives.”
Organisation Mondiale de la Santé. 2020.Éléments à prendre en considération
concernant les mesures de santé publique et les mesures sociales sur le lieu
de travail dans le cadre de l’épidémie de COVID-19: annexe du document élé-
ments à prendre en considération lors de l’ajustement des mesures de santé
publique et des mesures sociales dans le cadre de l’épidémie de COVID-19 10,
mai 2020. Organisation mondiale de la Santé.
Parisi, D. R., G. A. Patterson, L. Pagni, A. Osimani, T. Bacigalupo, J. Godfrid, F.
M. Bergagna, M. R. Brizi, P. Momesso, and F. L. Gomez. 2020. “Social Dis-
tance Characterization by means of Pedestrian Simulation.” arXiv preprint
arXiv:2009.04019.
Patrick Hertzke, S. M., Guillaume Neu, and Henry Weaver. 2020. “Moving
Forward: How COVID-19 Will Affect Mobility in the United Kingdom.”
Rosenfeld, E. 2020. “Using InfraWorks to Rethink How People Move Through
Shared Spaces.” Accessed 22 August 2020. https://blogs.autodesk.com/
infrastructure-reimagined/using-infraworks-to-rethink-how-people-
move-through-shared-spaces/#.
ARCHITECTURAL SCIENCE REVIEW 15
Schadschneider, A. 2002. “Traffic Flow: A Statistical Physics Point of View.”
Physica A: Statistical Mechanics and its Applications 313 (1-2): 153–187.
doi:10.1016/S0378-4371(02)01036-1.
Sophie Reynolds: Peter Baeck. 2020. “Smart Cities During COVID-19: How
Cities are Turning to Collective Intelligence to Enable Smarter Approaches
to COVID-19.” https://digileaders.com/smart-cities-during-covid-19-how-
cities-are-turning-to-collective-intelligence-to-enable-smarter-appro
aches-to-covid-19/.
Stifter, K. 2020. “Covid-19: La simulation de piétons peut aider les villes
à s’adapter à la distanciation sociale.” Accessed 22 August 2020.
https://blog.ptvgroup.com/fr/ville-et-mobilite/covid-19-simulation-pie
tonne-distanciation-sociale/.
Tebbouche, H., A. Bouchair, and S. Grimes. 2017. “Towards an Environmental
Approach for the Sustainability of Buildings in Algeria.” Energy Procedia
119: 98–110. doi:10.1016/j.egypro.2017.07.053.
Treuille, A., S. Cooper, and Z. Popović. 2006. “Continuum Crowds.” ACM Trans-
actions on Graphics (TOG) 25 (3): 1160–1168. doi:10.1145/1141911.114
2008.
Witze, A. 2020. “Universities Will Never be the Same After the Coronavirus
Crisis.” Nature 582 (7811): 162–165.
Zhou, Y., R. Xu, D. Hu, Y. Yue, Q. Li, and J. Xia. 2020. “Effects of Human
Mobility Restrictions on the Spread of COVID-19 in Shenzhen, China: A
Modelling Study Using Mobile Phone Data.” The Lancet Digital Health
2 (8): e417–e424. doi:10.1016/S2589-7500(20)30165-5.
Appendix
Tab le 5 . Scenario A, before COVID-19 of model 1, Average values of levels of service: LOS S and LOS V.
Walkway Persons Length (m) Width (m) Area (m2) Density (p/m2) Space (m2/p) Speed (m/s) Flow P/h LOS S LOS V
Walkway2 4 5.35 5.4 28.91 0.25 4.87 1.1 1.46 A E
Walkway3-B 7 14.99 5.4 80.94 0.16 7.73 1.06 0.89 A E
Walkway2-B 23 15.86 5.4 85.64 0.15 7.6 1.1 0.89 A E
Walkway3 24 8.03 5.4 43.38 0.28 4.33 1.08 1.61 A E
Walkway5 0 5.65 2 11.29 0.1 8.37 1.18 0.23 A D
Walkway4-B 0 4.51 5.4 24.35 0 24.35 1.07 0.01 A E
Walkway4-C 24 4.51 5.4 24.35 0.53 2.17 1.16 3.31 C D
Walkway14 7 25.23 3 75.7 0.1 11.73 1.16 0.35 A D
Walkway13 0 18.16 2 36.32 0.02 28.59 1.21 0.06 A D
Walkway1 7 3.78 5.4 20.42 0.2 6.46 1.19 1.27 A D
Walkway2-D 4 24.95 5.4 134.71 0.04 30.18 1.17 0.27 A D
Walkway8-B 0 3.8 3 11.41 0.08 9.01 1.22 0.29 A C
Walkway6 14 20.8 8 166.41 0.07 16.87 1.23 0.7 A C
Walkway5-B 11 20.93 8 167.47 0.07 16.18 1.19 0.7 A D
Walkway4 1 13.02 5.4 70.29 0 65.5 1.19 0.03 A D
Walkway19 20 24.62 5 123.09 0.13 8.28 1.28 0.83 A B
Walkway2-C 3 7.73 5.4 41.75 0.12 11.14 1.28 0.8 A B
Walkway9 1 12.22 2 24.44 0.03 20.55 1.3 0.08 A B
Walkway12 2 9.54 2 19.08 0.04 15.31 1.29 0.11 A B
Walkway11 0 11.69 2 23.38 0.04 18.08 1.29 0.11 A B
Walkway10 0 10.11 2 20.22 0.05 15.99 1.3 0.12 A B
Walkway15 0 4.47 2 8.94 0 8.94 1.26 0 A C
Walkway8 0 5.91 2 11.83 0 11.83 1.3 0 A B
Tab le 6 . Scenario B, COVID-19 social distance of model 1, average values of levels of service: LOS S and LOS V.
Walkway Persons Length (m) Width (m) Area (m2) Density (p/m2) Space (m2/p) Speed (m/s) Flow P/h LOS S LOS V
Walkway2 2 5.35 5.4 28.91 0.1 13.38 1.08 0.54 A E
Walkway3-B 10 14.99 5.4 80.94 0.07 20.89 1.03 0.35 A E
Walkway2-B 5 15.86 5.4 85.64 0.07 19.92 0.96 0.34 A E
Walkway3 1 8.03 5.4 43.38 0.11 12 1.08 0.63 A E
Walkway5 1 5.65 2 11.29 0.03 10.55 1.16 0.08 A D
Walkway4-B 0 4.51 5.4 24.35 0 24.35 0.96 0.01 A E
Walkway4-C 6 4.51 5.4 24.35 0.2 6.62 1.17 1.23 A D
Walkway14 2 25.23 3 75.7 0.04 33.45 1.14 0.13 A E
Walkway13 1 18.16 2 36.32 0.02 28.6 1.21 0.06 A D
Walkway1 1 3.78 5.4 20.42 0.08 13.44 1.17 0.48 A D
Walkway2-D 2 24.95 5.4 134.7 0.02 78.24 1.16 0.09 A D
Walkway8-B 0 3.8 3 11.41 0.03 10.21 1.16 0.1 A D
Walkway6 4 20.8 8 166.4 0.02 50.58 1.23 0.24 A C
Walkway5-B 6 20.93 8 167.5 0.02 50.77 1.21 0.23 A D
Walkway4 1 13.02 5.4 70.29 0 64.77 1.22 0.03 A C
Walkway19 3 24.62 5 123.1 0.05 26.91 1.28 0.29 A B
Walkway2-C 0 7.73 5.4 41.75 0.04 25.9 1.29 0.3 A B
Walkway9 0 12.22 2 24.44 0.01 23.43 1.3 0.03 A B
Walkway12 1 9.54 2 19.08 0.01 17.33 1.3 0.03 A B
Walkway11 0 11.69 2 23.38 0.01 21.63 1.3 0.04 A B
Walkway10 0 10.11 2 20.22 0.01 19.19 1.3 0.04 A B
Walkway15 0 4.47 2 8.94 0 8.94 1.3 0 A B
Walkway8 0 5.91 2 11.83 0 11.83 1.3 0 A B
16 M. BLIBLI AND A. BOUCHAIR
Tab le 7 . Scenario C, COVID-19 social distance and waves of model 1, average values of levels of service: LOS S and LOS V.
Walkway Persons Length (m) Width (m) Area (m2) Density (p/m2) Space (m2/p) Speed (m/s) Flow P/h LOS S LOS V
Walkway2 4 5.35 5.4 28.91 0.25 4.87 1.1 1.46 A E
Walkway3-B 7 14.99 5.4 80.94 0.16 7.73 1.06 0.89 A E
Walkway2-B 23 15.86 5.4 85.64 0.15 7.6 1.1 0.89 A E
Walkway3 24 8.03 5.4 43.38 0.28 4.33 1.08 1.61 A E
Walkway5 0 5.65 2 11.29 0.1 8.37 1.18 0.23 A D
Walkway4-B 0 4.51 5.4 24.35 0 24.35 1.07 0.01 A E
Walkway4-C 24 4.51 5.4 24.35 0.53 2.17 1.16 3.31 C D
Walkway14 7 25.23 3 75.7 0.1 11.73 1.16 0.35 A D
Walkway13 0 18.16 2 36.32 0.02 28.59 1.21 0.06 A D
Walkway1 7 3.78 5.4 20.42 0.2 6.46 1.19 1.27 A D
Walkway2-D 4 24.95 5.4 134.7 0.04 30.18 1.17 0.27 A D
Walkway8-B 0 3.8 3 11.41 0.08 9.01 1.22 0.29 A C
Walkway6 14 20.8 8 166.4 0.07 16.87 1.23 0.7 A C
Walkway5-B 11 20.93 8 167.5 0.07 16.18 1.19 0.7 A D
Walkway4 1 13.02 5.4 70.29 0 65.5 1.19 0.03 A D
Walkway19 20 24.62 5 123.1 0.13 8.28 1.28 0.83 A B
Walkway2-C 3 7.73 5.4 41.75 0.12 11.14 1.28 0.8 A B
Walkway9 1 12.22 2 24.44 0.03 20.55 1.3 0.08 A B
Walkway12 2 9.54 2 19.08 0.04 15.31 1.29 0.11 A B
Walkway11 0 11.69 2 23.38 0.04 18.08 1.29 0.11 A B
Walkway10 0 10.11 2 20.22 0.05 15.99 1.3 0.12 A B
Walkway15 0 4.47 2 8.94 0 8.94 1.26 0 A C
Walkway8 0 5.91 2 11.83 0 11.83 1.3 0 A B
Tab le 8 . Scenario COVID-19 protocol of Model 2, average values of levels of service: LOS S and LOS V.
Walkway Persons Length (m) Width (m) Area (m2) Density (p/m2) Space (m2/p) Speed (m/s) Flow P/h LOS S LOS V
Walkway6 0 20.8 2 41.6 0 37.02 1.13 0.01 A E
Walkway7-B27 0 7.76 3 23.27 0.01 21.75 0.88 0.03 A E
Walkway7-B30 4 23.86 3.5 83.5 0.04 31.3 1.14 0.17 A E
Walkway7-B6 0 6.73 3.5 23.54 0.02 20.16 0.83 0.07 A E
Walkway7-B8 7 13.12 5 65.62 0.12 10.31 1.2 0.68 A D
Walkway2 1 62.76 2 125.5 0 111.4 1.25 0.01 A C
Walkway3 1 2.93 2 5.87 0.03 5.68 1.26 0.08 A C
Walkway4 0 5.22 2 10.44 0.01 10.14 1.27 0.04 A C
Walkway4-B 0 23.51 2 47.01 0.02 37.87 1.26 0.04 A C
Walkway7-B10 3 9.88 2.5 24.71 0.15 7.87 1.24 0.47 A C
Walkway7-B20 5 5.07 2.5 12.67 0.27 4.53 1.25 0.82 A C
Walkway7-B3 4 64.59 3.5 226.1 0.02 73.38 1.24 0.08 A C
Walkway7-B7 1 4.36 3.5 15.25 0.06 11.85 1.26 0.25 A C
Walkway7-B9 1 15.03 2.5 37.58 0.03 27.06 1.24 0.1 A C
Walkway9 0 14.14 2 28.28 0 27.3 1.25 0.01 A C
Walkway1 0 23.42 2 46.84 0.02 38.28 1.29 0.04 A B
Walkway2-B 0 36.72 2 73.44 0 72.6 1.3 0.01 A B
Walkway3-B 1 25.85 2 51.71 0.01 42.77 1.3 0.04 A B
Walkway3-B2 0 56.59 2 113.2 0 100.1 1.3 0.01 A B
Walkway5 0 24.71 4 98.83 0.01 75.15 1.3 0.05 A B
Walkway6-B 1 17.42 2 34.84 0.01 31.01 1.3 0.02 A B
Walkway7 0 39.28 2 78.57 0.01 65.96 1.3 0.01 A B
Walkway7-B 0 8.26 2 16.53 0 16.53 1.3 0.01 A B
Walkway7-B11 0 11.4 2.5 28.51 0.04 21.58 1.3 0.12 A B
Walkway7-B12 1 22.25 2 44.51 0.01 41.48 1.3 0.02 A B
Walkway7-B13 0 6.08 2 12.15 0 12.15 1.3 0 A B
Walkway7-B14 1 19.52 2 39.04 0.01 34.9 1.3 0.03 A B
ARCHITECTURAL SCIENCE REVIEW 17
Tab le 9 . Evolution of densities and speeds of the walkway 2 over time and comparison of the Model 1 scenarios.
Walkway2 Scenario A before Covid 19 Scenario B Covid 19 Soc Dist Scenario C Covid 19 waves
Time (m,s) Persons Length (m) Width (m) Area (m2) Density1 (p/m2) Speed1 (m/s) LOS S LOS V Density 2 (p/m2) Speed 2 (m/s) LOS S LOS V Density 3 (p/m2) Speed 3 (m/s) LOS S LOS V
5 9 5.35 5.4 28.91 0.31 1.08 BE 0.07 1.29 AB 0.1 0.57 A F
10 12 5.35 5.4 28.91 0.42 1.16 BD 0.42 0.92 BE 0.14 1.28 A B
15 6 5.35 5.4 28.91 0.21 0.74 AF 0.45 1.01 CE 0.07 0.52 A F
20 4 5.35 5.4 28.91 0.14 1.2 AD 0.31 1.14 BE 0.03 1.3 A B
25 10 5.35 5.4 28.91 0.35 1.29 BB 0.42 0.71 BF 0.28 1.11 A E
30 3 5.35 5.4 28.91 0.1 1.29 AB 0.62 0.6 CF 0.31 0.76 B F
35 4 5.35 5.4 28.91 0.14 0.68 AF 0.62 0.49 CF 0.17 1.09 A E
40 6 5.35 5.4 28.91 0.21 1.25 AC 0.42 0.7 BF 0.03 1.3 A B
42 35 8 5.35 5.4 28.91 0.28 1.08 AE 0.28 0.69 AF 0.07 0.74 A F
Table 10. Evolution of densities and speeds of walkway 2b over time and Comparison of Model 1 Scenarios.
Walkway2-b Scenario A before Covid 19 Scenario B Covid 19 Soc Dist Scenario C Covid 19 waves
Time (m,s) Persons Length (m) Width (m) Area (m2) Density1 (p/m2) Speed1 (m/s) LOS S LOS V Density 2 (p/m2) Speed 2 (m/s) LOS S LOS V Density 3 (p/m2) Speed 3 (m/s) LOS S LOS V
5 17 15.86 5.4 85.64 0.2 1.14 AE 0.12 0.91 AE 0.06 0.73 A F
10 19 15.86 5.4 85.64 0.22 1.19 AD0.3 0.8BE 0.07 1.26 A C
15 7 15.86 5.4 85.64 0.08 1.29 AB 0.34 0.55 BF 0.11 0.84 A E
20 8 15.86 5.4 85.64 0.09 1.15 AE 0.18 0.7 AF 0.04 1.3 A B
25 21 15.86 5.4 85.64 0.25 0.79 AE 0.34 0.68 BF 0.06 1.18 A D
30 11 15.86 5.4 85.64 0.13 1.1 AE 0.41 0.54 BF 0.08 0.63 A F
35 16 15.86 5.4 85.64 0.19 1.01 AE 0.3 0.61 BF 0.08 0.73 A F
40 7 15.86 5.4 85.64 0.08 1.23 AC 0.27 0.73 AF 0.12 0.87 A E
42 35 15 15.86 5.4 85.64 0.18 1.21 AD 0.22 0.69 AF 0.06 1.3 A B
18 M. BLIBLI AND A. BOUCHAIR
Table 11. Evolution of densities and speeds of walkway 3 over time and comparison of Model 1 scenarios.
Walkway3 Scenario A before Covid 19 Scenario B Covid 19 Soc Dist Scenario C Covid 19 waves
Time (m,s) Persons Length (m) Width (m) Area (m2) Density1 (p/m2) Speed1 (m/s) LOS S LOS V Density 2 (p/m2) Speed 2 (m/s) LOS S LOS V Density 3 (p/m2) Speed 3 (m/s) LOS S LOS V
5 8 8.03 5.4 43.38 0.18 1.21 AD 0.53 0.81 CE 0.14 1.12 A E
10 15 8.03 5.4 43.38 0.35 1.19 BD 0.6 0.56 CF 0.18 0.69 A F
15 5 8.03 5.4 43.38 0.12 0.75 AF 0.35 0.76 BF 0.07 1.3 A B
20 12 8.03 5.4 43.38 0.28 1.21 AD 0.23 0.86 AE 0.07 1.3 A B
25 16 8.03 5.4 43.38 0.37 0.99 BE 0.3 0.94 AE 0.16 1.23 A C
30 25 8.03 5.4 43.38 0.58 0.89 CE 0.25 0.98 AE 0.23 0.8 A E
35 6 8.03 5.4 43.38 0.14 1.22 AD 0.35 0.79 BE 0.07 1.27 A B
40 8 8.03 5.4 43.38 0.18 1.25 AC 0.23 0.86 AE 0.09 1.28 A B
42 35 22 8.03 5.4 43.38 0.51 0.97 CE 0.51 1.04 CE 0.02 1.3 A B
Table 12. Evolution of densities and speeds of walkway 3b over time and comparison of Model 1 scenarios.
Walkway3-b Scenario A before Covid 19 Scenario B Covid 19 Soc Dist Scenario C Covid 19 waves
Time (m,s) Persons Length (m) Width (m) Area (m2) Density1 (p/m2) Speed1 (m/s) LOS S LOS V Density 2 (p/m2) Speed 2 (m/s) LOS S LOS V Density 3 (p/m2) Speed 3 (m/s) LOS S LOS V
5 11 14.99 5.4 80.94 0.14 0.95 AE 0.15 0.73 AF 0.04 1.27 A C
10 11 14.99 5.4 80.94 0.14 0.87 AE 0.27 0.68 AF 0.07 0.7 A F
15 19 14.99 5.4 80.94 0.23 1.07 AE 0.27 0.72 AF 0.04 1.3 A B
20 15 14.99 5.4 80.94 0.19 1.14 AE 0.35 0.65 BF 0.05 1.3 A B
25 11 14.99 5.4 80.94 0.14 1.2 AD 0.35 0.84 BE 0.11 0.66 A F
30 16 14.99 5.4 80.94 0.2 1.19 AD 0.19 0.69 AF 0.09 1.09 A E
35 8 14.99 5.4 80.94 0.1 1.2 AD 0.26 0.81 AE 0.06 0.96 A E
40 8 14.99 5.4 80.94 0.1 1.29 AB 0.16 1.06 AE 0.01 1.3 A B
42 35 7 14.99 5.4 80.94 0.09 1.18 AD 0.38 0.64 BF 0.12 0.38 A F
ARCHITECTURAL SCIENCE REVIEW 19
Table 13. Evolution of densities and speeds of walkway 5 over time and comparison of Model 1 scenarios.
Walkway5 Scenario A before Covid 19 Scenario B Covid 19 Soc Dist Scenario C Covid 19 waves
Time (m,s) Persons Length (m) Width (m) Area (m2) Density1 (p/m2) Speed1 (m/s) LOS S LOS V Density 2 (p/m2) Speed 2 (m/s) LOS S LOS V Density 3 (p/m2) Speed 3 (m/s) LOS S LOS V
5 0 5.65 2 11.29 0 1.5 AA 0.09 1.18 AD 0.09 1.25 A C
10 1 5.65 2 11.29 0.09 1.3 AB 0.09 0.96 AE 0.09 1.3 A B
15 1 5.65 2 11.29 0.09 1.15 AE 0.09 1.18 AD0 1.5A A
20 1 5.65 2 11.29 0.09 1.3 AB 0.09 1.28 AB 0.09 1.3 A B
25 3 5.65 2 11.29 0.27 1.29 AB0.09 0 AF 0.09 1.3 A B
30 0 5.65 2 11.29 0 1.5 AA 0.18 0.65 AF 0.09 1.3 A B
35 3 5.65 2 11.29 0.27 1.26 AC 0.09 1.2 AD 0.09 1.3 A B
40 0 5.65 2 11.29 0 1.5 AA 0.09 1.29 AB0 1.5A A
42 35 2 5.65 2 11.29 0.18 1.14 AE 0.18 1.25 AC 0.09 1.3 A B
20 M. BLIBLI AND A. BOUCHAIR
Table 14. Evolution of densities and speeds of walkway 2 over the duration of the COVID-19 Scenario protocol.
Time (m,s) Persons Length (m) Width (m) Area (m2) Density1 (p/m2) Speed1 (m/s) LOS S LOS V
5 9 5.35 5.4 28.91 0.31 1.08 B E
10 12 5.35 5.4 28.91 0.42 1.16 B D
15 6 5.35 5.4 28.91 0.21 0.74 A F
20 4 5.35 5.4 28.91 0.14 1.2 A D
25 10 5.35 5.4 28.91 0.35 1.29 B B
30 3 5.35 5.4 28.91 0.1 1.29 A B
35 4 5.35 5.4 28.91 0.14 0.68 A F
40 6 5.35 5.4 28.91 0.21 1.25 A C
45 8 5.35 5.4 28.91 0.28 1.08 A E
Table 15. Evolution of densities and speeds of walkway 2b over the duration of the COVID-19 Scenario protocol.
Time (m,s) Persons Length (m) Width (m) Area (m2) Density1 (p/m2) Speed1 (m/s) LOS S LOS V
5 17 15.86 5.4 85.64 0.2 1.14 A E
10 19 15.86 5.4 85.64 0.22 1.19 A D
15 7 15.86 5.4 85.64 0.08 1.29 A B
20 8 15.86 5.4 85.64 0.09 1.15 A E
25 21 15.86 5.4 85.64 0.25 0.79 A E
30 11 15.86 5.4 85.64 0.13 1.1 A E
35 16 15.86 5.4 85.64 0.19 1.01 A E
40 7 15.86 5.4 85.64 0.08 1.23 A C
45 15 15.86 5.4 85.64 0.18 1.21 A D
Table 16. Evolution of densities and speeds of walkway 3 over the duration of the COVID-19 Scenario protocol.
Time (m,s) Persons Length (m) Width (m) Area (m2) Density1 (p/m2) Speed1 (m/s) LOS S LOS V
5 11 14.99 5.4 80.94 0.14 0.95 A E
10 11 14.99 5.4 80.94 0.14 0.87 A E
15 19 14.99 5.4 80.94 0.23 1.07 A E
20 15 14.99 5.4 80.94 0.19 1.14 A E
25 11 14.99 5.4 80.94 0.14 1.2 A D
30 16 14.99 5.4 80.94 0.2 1.19 A D
35 8 14.99 5.4 80.94 0.1 1.2 A D
40 8 14.99 5.4 80.94 0.1 1.29 A B
45 7 14.99 5.4 80.94 0.09 1.18 A D
Table 17. Evolution of densities and speeds of walkway 3b over the duration of the COVID-19 Scenario protocol.
Time (m,s) Persons Length (m) Width (m) Area (m2) Density1 (p/m2) Speed1 (m/s) LOS S LOS V
5 8 8.03 5.4 43.38 0.18 1.21 A D
10 15 8.03 5.4 43.38 0.35 1.19 B D
15 5 8.03 5.4 43.38 0.12 0.75 A F
20 12 8.03 5.4 43.38 0.28 1.21 A D
25 16 8.03 5.4 43.38 0.37 0.99 B E
30 25 8.03 5.4 43.38 0.58 0.89 C E
35 6 8.03 5.4 43.38 0.14 1.22 A D
40 8 8.03 5.4 43.38 0.18 1.25 A C
45 22 8.03 5.4 43.38 0.51 0.97 C E
Table 18. Evolution of densities and speeds of walkway 5 over the duration of the COVID-19 Scenario protocol.
Time (m,s) Persons Length (m) Width (m) Area (m2) Density1 (p/m2) Speed1 (m/s) LOS S LOS V
5 0 5.65 2 11.29 0 1.5 A A
10 1 5.65 2 11.29 0.09 1.3 A B
15 1 5.65 2 11.29 0.09 1.15 A E
20 1 5.65 2 11.29 0.09 1.3 A B
25 3 5.65 2 11.29 0.27 1.29 A B
30 0 5.65 2 11.29 0 1.5 A A
35 3 5.65 2 11.29 0.27 1.26 A C
40 0 5.65 2 11.29 0 1.5 A A
45 2 5.65 2 11.29 0.18 1.14 A E
ARCHITECTURAL SCIENCE REVIEW 21
Figure 20. Evolution of densities and speeds of walkway 2b over time and comparison of Model 1 scenarios. Source: authors.
Figure 21. Evolution of densities and speeds of walkway 3 over time and comparison of Model 1 scenarios. Source: authors.
Figure 22. Evolution of densities and speeds of walkway 3b over time and comparison of Model 1 scenarios. Source: authors.
Figure 23. Evolution of densities and speeds of walkway 5 over time and comparison of Model 1 scenarios. Source: authors.
22 M. BLIBLI AND A. BOUCHAIR
Figure 24. Evolution of densities and speeds of walkway 2b over the duration of the COVID-19 scenario protocol. Source: authors.
Figure 25. Evolution of densities and speeds of walkway 3 over the duration of the COVID-19 scenario protocol. Source: authors.
Figure 26. Evolution of densities and speeds of walkway 3b over the duration of the COVID-19 scenario protocol. Source: authors.
Figure 27. Evolution of densities and speeds of walkway 5 over the duration of the COVID-19 scenario protocol. Source: authors.
