sustainability

Article
The Analysis and AI Simulation of Passenger Flows in an
Airport Terminal: A Decision-Making Tool
Afroditi Anagnostopoulou * , Dimitrios Tolikas, Evangelos Spyrou, Attila Akac and Vassilios Kappatos

Centre for Research and Technology Hellas, Hellenic Institute of Transport, Thermi, 57001 Thessaloniki, Greece;
dtolikas@certh.gr (D.T.); espyrou@certh.gr (E.S.); akac.attila@certh.gr (A.A.); vkappatos@certh.gr (V.K.)
* Correspondence: a.anagnostopoulou@certh.gr; Tel.: +30-2121069828

Abstract: In this paper, a decision-making tool is proposed that can utilize different strategies to deal
with passenger flows in airport terminals. A simulation model has been developed to investigate
these strategies, which can be updated and modified based on the current requirements of an airport
terminal. The proposed tool could help airport managers and relevant decision makers proactively
mitigate potential risks and evaluate crowd management strategies. The aim is to eliminate risk
factors due to overcrowding and minimize passenger waiting times within the terminal to provide a
seamless, safe and satisfying travel experience. Overcrowding in certain areas of the terminal makes
it difficult for passengers to move freely and increases the risk of accidents (especially in the event of
an emergency), security problems and service interruptions. In addition, long queues can lead to
frustration among passengers and increase potential conflicts or stress-related incidents. Based on
the derived results, the optimized routing of passengers using modern technological solutions is the
most promising crowd management strategy for a sample airport that can handle 800 passengers
per hour.

Keywords: crowd management; risk assessment; simulation model

Citation: Anagnostopoulou, A.; 1. Introduction

Tolikas, D.; Spyrou, E.; Akac, A.;
According to the International Air Transport Association (IATA) [1], airport terminals
Kappatos, V. The Analysis and AI
should meet passenger demand and provide an adequate level of service in terms of waiting
Simulation of Passenger Flows in an
Airport Terminal: A Decision-Making
times in queues, crowdedness level and delays, considering government restrictions due
Tool. Sustainability 2024, 16, 1346.
to terrorism [2], infectious disease outbreaks (such as COVID-19) [3] and extreme weather
https://doi.org/10.3390/su16031346
events [4]. Therefore, airports may have to deal with increased demand and should
accommodate more passengers. This is not only costly but also poses a major challenge for
Academic Editors: Evangelos Bekiaris
airports, from a logistical perspective, to meet the needs of passengers and optimize their
and Maria Gkemou
movements within the terminal.
Received: 28 November 2023 Artificial Intelligence (AI) models could provide successful solutions for crowd man-
Revised: 31 January 2024 agement in the transport sector [5], offering a double benefit by not only optimizing their
Accepted: 2 February 2024 operational efficiency but also improving overall sustainability. More specifically, virtual
Published: 5 February 2024 queues for better passenger orientation when arriving at the airport [6], queue management
with buffer zones for passenger waiting areas [7] and smart queue management tools [8]
could help flight managers optimize resource allocation and airport operations. Crowd
management in airport terminals is essential to ensure safety and a high level of service
Copyright: © 2024 by the authors. for passengers, while optimizing terminal processes and resource allocation could achieve
Licensee MDPI, Basel, Switzerland.
operational resilience for a more organized response to unexpected events and a higher
This article is an open access article
level of sustainability.
distributed under the terms and
In this context, this paper aims to develop a sophisticated decision-making tool that
conditions of the Creative Commons
uses different strategies to deal with passenger flows in airport terminals based on a
Attribution (CC BY) license (https://
simulation model. The proposed tool could help airport managers and relevant decision
creativecommons.org/licenses/by/
4.0/).
makers proactively assess and mitigate potential risks by providing a data-driven approach

Sustainability 2024, 16, 1346. https://doi.org/10.3390/su16031346 https://www.mdpi.com/journal/sustainability

Sustainability 2024, 16, 1346 2 of 13

to real-time decision making. This tool will also identify and analyze potential risks, due to
queues and crowded stores, that could lead to negative consequences. Once the risks are
assessed and categorized, implementation strategies will be imposed to mitigate or control
these risks at an operational level. This will result in reduced queues and operational
vulnerabilities and prevent crowd disruption.
The remainder of this paper is organized as follows: Section 2 provides an overview
of previous research studies in the field of risk assessment in airport terminals and includes
a review of the existing literature to date. Section 3 provides a detailed analysis and
presentation of the proposed decision-making tool. Section 4 then presents the results
obtained from the practical implementation of the proposed tool. Finally, Section 5 draws
conclusions based on the results and provides key findings as valuable guidance for future
research in the dynamic field of airport terminal risk assessment and management.

2. Research Background
2.1. The Level of Service in Airport Terminals
Airport terminals are important hubs of the global transportation network, managing
internal and external flights as well as connecting flights and their corresponding passenger
flows. The continuous growth and development of the aviation industry, as well as
geopolitical crises and extreme weather conditions, make airport terminals necessary in
order to provide passengers with a safe, efficient and successful travel experience. To
cope with the complex environment of airport terminals, various crowd management
strategies have been developed to streamline processes and provide passenger-friendly
design [9]. More recent studies focus on advanced technologies, information systems and
smart applications developed to improve scheduling, optimize airport operations and
increase the quality of service for passengers [10,11].
In this context, research reports conducted both before and after the COVID-19 pan-
demic summarize the future of airport terminals and aviation in general, focusing on how
modern technology could improve security processes and increase the level of service to
passengers [12,13]. In terms of operational efficiency, Thampan et al. [14] have presented
and analyzed how analytical methods can be used to evaluate service quality, considering
the spatial, operational and technological changes in airport terminals. More recently,
Dong et al. [15] have used machine learning techniques to predict incidents leading to
flight delays. This predictive capability could improve flight scheduling for more effective
passenger service and mitigate security and economic risks in the airport environment.
In addition, several crowd management strategies have been proposed to improve
airport terminal environments that use simulation approaches, covering areas such as
optimization, passenger behavior, security, and building design. Kim and Wu [16] used an
agent-based simulation approach to optimize passenger service levels in a cost-effective
manner by considering the degree of crowding and passenger waiting times. Similarly,
Mekić et al. [17] used an agent-based model for airport terminal operations to investigate
how passenger behavior affects waiting time, expenses, missed flights and operational
costs. In terms of security, Jassen et al. [18] proposed agent-based models and Monte Carlo
simulations to comprehensively assess and manage security risks.
On the other hand, the research community has also investigated how new airport in-
frastructures and terminal expansions might affect the level of service for passengers [19,20].
Passenger characteristics have also been studied in conjunction with terminal layout, and
airport retail revenue has been investigated using an agent-based simulation approach [21].
In addition, passenger preferences for shopping and catering activities within the terminal
have been investigated [22], as well as how their arrival times should be considered when
managing the commercial area of an airport at a strategic level [23]. Recent studies also
investigated methods and strategies to improve social distancing due to COVID-19 and
minimize the crowded areas within terminals [24,25].
The high level of complexity exhibited by airport terminals can be captured using
agent-based models [26] that represent different groups of stakeholders and depict their
Sustainability 2024, 16, 1346 3 of 13

interdependencies. In addition, sustainability indicators, covering economic, environmen-

tal and social aspects, could be used to evaluate the performance of airport terminals,
using data analysis and simulation models as well as qualitative and quantitative infor-
mation [27]. Assessment tools such as LCA (Life Cycle Analysis) are increasingly being
used to help decision makers select strategies, operations and policy mechanisms [28]. It is
also worth mentioning that the CSR (Corporate Social Responsibility) of airports [29] is a
business model that could be used to measure the environmental, economic, energy and
social impacts of applied strategies and operational activities to achieve commitment from
decision makers [30].

2.2. Sustainability in Airport Terminals

Considering short-term initiatives for sustainability in airport terminals, optimizing
resource allocation and improving operational efficiency may be among the most impor-
tant goals of airport managers and relevant decision makers. Their focus on day-to-day
measures includes energy-efficient strategies and renewable energy devices, as well as
smart technologies for real-time monitoring and control [31]. Educational campaigns for
passengers could also contribute to sustainability by raising passenger awareness and
encouraging them to participate in environmentally friendly actions and programs. In
addition, a sustainable connection of the airport terminal to public transportation could
reduce the GHG (Greenhouse Gas) emissions associated with passenger and employee
commuting [32].
On the other hand, long-term sustainability includes measures that pave the way
for sustainable airports that will define their future operations, approaches and tools.
Smart infrastructures represent an indicative long-term strategy that will enable the use
of smart applications and technologies in daily operations [33]. The design of airport
terminals (either for expansion or new construction) could be developed with a focus on
sustainability, using energy-efficient materials, natural light and green terraces or roofs [34].
Finally, connecting airport managers and relevant decision makers with other stakeholders
and interested actors, such as public authorities and industry representatives, and actively
involving them can create economic opportunities and joint initiatives that benefit both the
airport and its surroundings in the long run [35].

3. The Decision-Making Tool

The proposed decision-making tool is based on a simulation model that provides a
visual representation of how passengers move within the airport terminal environment.
This enables airport managers and relevant decision makers to identify flow patterns,
potential congestion points and bottlenecks. The main goal is to reduce waiting times in the
security control area and at the entry points of stores in order to provide a higher quality of
service to passengers. The corresponding performance metrics set the basis upon which to
evaluate different usage scenarios, and Figure 1 depicts the main concept of the proposed
decision-making tool. The scenarios include the base case and the implementing of various
crowd management strategies such as technology integration, trained security personnel,
multiple entry points, access control, etc. These strategies are thoroughly designed to
optimize passenger flows and improve the efficiency of airport terminal operations.
The base case simulates passengers’ movement in a stochastic way to capture the
flows before any crowd strategy implementation. Three main strategies are examined to
simulate different technology applications and operational approaches. More specifically,
the technology integration strategy could lead to improved navigation through the use
of modern mobile applications, which consider several parameters and restrictions (i.e.,
passengers’ preferences, the maximum capacity of stores, etc.). The trained security per-
sonnel strategy is also used by the proposed tool for optimizing service time in check-in
and security control areas. Finally, the strategy of multiple entry points is also studied,
exploring higher efficiency for both economy and business/gold class passengers.
Sustainability 2024, 16, x FOR PEER REVIEW 4 of 13

Sustainability 2024, 16, 1346

security control areas. Finally, the strategy of multiple entry points is also studied, explor-
4 of 13
ing higher efficiency for both economy and business/gold class passengers.

Figure
Figure 1. Concept of
1. Concept of the
the proposed
proposed decision-making
decision-making tool.
tool.

Moreover, the decision-making tool can also consider passengers’ specific preferences
Moreover, the decision-making tool can also consider passengers’ specific prefer-
for visiting different stores within the airport terminal area. It also accounts for the number
ences for visiting different stores within the airport terminal area. It also accounts for the
of passengers waiting to visit each store and takes the maximum capacity of the stores and
number of passengers waiting to visit each store and takes the maximum capacity of the
other relevant parameters into account. The proposed tool can also facilitate continuous
stores and other relevant parameters into account. The proposed tool can also facilitate
improvements by updating and refining the model with current real-data and passengers’
continuous improvements by updating and refining the model with current real-data and
feedback in an attempt to avoid inefficiencies and prevent congestion-related issues. An
passengers’ feedback in an attempt to avoid inefficiencies and prevent congestion-related
iterative implementation process is proposed to ensure that the model remains adaptable,
issues. An iterative implementation process is proposed to ensure that the model remains
helping to proactively address inefficiencies and mitigate any potential challenges, fostering
aadaptable, helpingand
more responsive to proactively
passenger-centricaddressairport
inefficiencies and mitigate any potential chal-
environment.
lenges, fostering a more responsive and passenger-centric airport environment.
3.1. The Simulation Model
3.1. The Simulation Model
Due to overcrowding in the security control area and at the entry points of different
stores,Due to overcrowding
health in the security
hazards (especially during control
pandemics),area and at the entry
accidents pointsthreats
and safety of different
have
appeared in airport terminals. Therefore, efficient crowd management is needed to have
stores, health hazards (especially during pandemics), accidents and safety threats plan
appeared
flows duringin airport terminals.
operating times and Therefore, efficient crowd
guide passengers, management
reducing waitingistimes
needed to plan
in queues
flowsimproving
and during operating
the leveltimes and guide
of service forpassengers,
passengers.reducing waiting times
More specifically, theininput
queues of and
the
improving simulation
developed the level of model
serviceisfor passengers.
the More specifically,
list of passengers (800 in total) thethat
input of the
have developed
arrived at the
simulation
airport modeland
terminal is the
are list of passengers
waiting (800 in total)
for their departing thatEach
flight. havepassenger
arrived athas the specific
airport
terminal and are
characteristics waiting
that for their departing
are considered flight. Eachmodel
in the simulation passenger
(i.e., has specific
flight character-
number, ticket
istics thatcheck-in
category, are considered in the simulation
status, preferences for visitingmodel (i.e.,
stores in flight number,
the airport ticket category,
terminal).
check-in status,
Figure preferences
2 depicts for visiting
the sequence storesand
of actions in the airport
flows withinterminal).
the studied sample airport
Figure
terminal, as 2well
depicts the
as the sequence
decision of actions
points and flowsprocesses
in the different within the studied
that sample airport
the simulation model
terminal, asThe
considers. wellfirst
as the
stepdecision points inifthe
is to examine thedifferent
passengerprocesses that the
has checked insimulation
online or ifmodel
s/he
considers.
needs Thethe
to visit first step is desks.
check-in to examine if the passenger
The simulation modelhas checked
checks whetherin online or if s/he
the check-in is
needs(which
open to visitmeans
the check-in
that thedesks.
flight The
departssimulation model
in less than 2.5checks
h) and whether the check-in
the passenger proceeds is
open (which means that the flight departs in less than 2.5 h) and the passenger proceeds
to the desks. Otherwise, s/he needs to wait until the check-in opens. Then, the passenger
to the desks.
should Otherwise,
go to the security s/he
controlneedsareatoand
wait until
their the check-in
waiting opens. Then,
time depends on thethe passenger
queue length
should
and the go to the security
efficiency control personnel.
of the security area and their waiting
It should be time depends
mentioned onimproved
that the queueservices
length
and the
(with efficiency
reduced of thetime)
waiting security arepersonnel.
offered to It theshould
gold be mentioned
class membersthat andimproved
passengersservices
with
(with reduced
business tickets,waiting
compared time) to are
thoseoffered to the gold
in economy, in theclass members
check-in and passengers
and security with
control areas.
business tickets, compared to those in economy, in the check-in and security control areas.
 boarding. Passengers visit stores in the shopping area in a predefined order, based on
preferences, or avoid this step. In the latter case, they proceed to the catering service
extra waiting time is needed for them to be served. Last, they proceed to the gate
wait for boarding to start, or embark in case where the boarding of their flight has al
Sustainability 2024, 16, 1346
started. Note that when boarding starts, the passengers should proceed 5 ofto
13 the gate

this reason, the simulation model continuously checks if boarding has started.

Figure 2.of
Figure 2. Flow diagram Flow diagrammovements
passengers’ of passengers’ movements
in the simulatedinsample
the simulated sample airport terminal.
airport terminal.

The next 3.2. Usage Scenarios

processes involve shopping in the airport’s duty-free stores (two in total),
3.2.1. Scenarioclothing
optical stores (two in total), 0—Basestores
Case Strategy
(two in total), jewelry stores (two in total) and
perfume stores (two in total), as well as using catering and revitalization services (three
The first scenario follows a stochastic approach to simulate passengers’ movem
restaurants in total, three coffee shops in total and two WCs in total) before the passenger’s
(800 passengers per hour) [36,37] as they visit their preferred stores in the shopping
boarding. Passengers visit stores in the shopping area in a predefined order, based on their
within the airport terminal. Passengers visit the shopping area and the catering ser
preferences, or avoid this step. In the latter case, they proceed to the catering services and
in a non-deterministic manner, making their choices based on a random probab
extra waiting time is needed for them to be served. Last, they proceed to the gates and
functions. This scenario serves as the base case scenario, in which passengers ar
wait for boarding to start, or embark in case where the boarding of their flight has already
guided to visit airport stores in a predefined sequence. Instead, they independently
started. Note that when boarding starts, the passengers should proceed to the gates. For
stores based on their personal preferences in an ad hoc, non-directed fashion, contrib
this reason, the simulation model continuously checks if boarding has started.
to a realistic representation of spontaneous passenger behaviors within the shoppin
catering/revitalization areas of an airport terminal.
3.2. Usage Scenarios
3.2.1. Scenario 0—Base Case Strategy
3.2.2. Scenario 1—Technology Integration Strategy
The first scenario follows a stochastic approach to simulate passengers’ movements
(800 passengers perInhour)
this scenario,
[36,37] aspassengers are being
they visit their guided
preferred via modern
stores technology
in the shopping areaand, more
cifically,
within the airport an optimization
terminal. Passengers visit algorithm, which
the shopping areaconsiders passengers’
and the catering preferences
services in a an
stores’
non-deterministic restrictions.
manner, makingThe theiraim is to based
choices guide on
passengers
a randomtoprobabilistic
the stores with the current lo
functions.
This scenario crowdedness
serves as the levels, as described
base case scenario, ininEquation (1), via a mobile
which passengers are notapp. Therefore,
guided to the
rithm embedded in the app considers passengers’ preferences,
visit airport stores in a predefined sequence. Instead, they independently select stores the maximum allowe
based on theirpacity of each
personal store, thein
preferences number of passengers
an ad hoc, that are
non-directed aboutcontributing
fashion, to visit each store
to after
the mobile app,
a realistic representation the numberpassenger
of spontaneous of passengers in each
behaviors store the
within at that time and
shopping andthe numb
passengers
catering/revitalization thatofare
areas anwaiting
airport to be served as well as their service time.
terminal.
𝐹𝐹𝑖𝑖,𝑖𝑖𝑖𝑖
�𝐹𝐹𝑖𝑖,𝑟𝑟 + 𝐹𝐹𝑖𝑖,𝑤𝑤 + 𝐹𝐹
3.2.2. Scenario 1—Technology Integration Strategy � ∗ 𝑇𝑇𝑖𝑖 𝐹𝐹𝑖𝑖,𝑚𝑚𝑚𝑚𝑚𝑚 = 𝐹𝐹𝑖𝑖,𝑖𝑖𝑖𝑖
𝑖𝑖,𝑚𝑚𝑚𝑚𝑚𝑚
𝐹𝐹𝑖𝑖 = �
In this scenario, passengers are being guided 𝐹𝐹𝑖𝑖,𝑛𝑛𝑛𝑛 modern technology and, more
�𝐹𝐹𝑖𝑖,𝑟𝑟 + 𝐹𝐹via � ∗ 𝑇𝑇𝑖𝑖 − 𝑐𝑐 𝐹𝐹𝑖𝑖,𝑚𝑚𝑚𝑚𝑚𝑚 > 𝐹𝐹𝑖𝑖,𝑖𝑖𝑖𝑖
𝑖𝑖,𝑚𝑚𝑚𝑚𝑚𝑚
specifically, an optimization algorithm, which considers passengers’ preferences and the
stores’ restrictions. The aim is to guide passengers to the stores with the current lowest
crowdedness levels, as described in Equation (1), via a mobile app. Therefore, the algorithm
embedded in the app considers passengers’ preferences, the maximum allowed capacity of
each store, the number of passengers that are about to visit each store after using the mobile
app, the number of passengers in each store at that time and the number of passengers that
are waiting to be served as well as their service time.
 
 Fi,r + Fi,w + Fi,in ∗ Ti Fi,max = Fi,in
Fi =  Fi,max (1)
F
 Fi,r + i,nn ∗ Ti − c
F Fi,max > Fi,in
i,max
Sustainability 2024, 16, 1346 6 of 13

where
• Fi,w is the number of passengers waiting to enter store i;
• Fi,r is the number of passengers going to store i;
• Fi,in is the number of passengers inside store i;
• Fi,max is the maximum capacity in passengers of store i;
• Ti is the service time of a passenger in store i;
• c is a negative constant to capture the crowd avoidance preference.
When a passenger passes the security check, the algorithm is used to calculate store
i with the minimum crowdedness level Fi , which indicates the place with the maximum
current capacity limit. After a passenger exits this store, the algorithm re-calculates the
minimum crowdedness level Fi for the remaining stores and directs the passenger to the
next one. This process continues until the passenger has visited all the stores they like.
The same procedure is also followed for visiting the airport restaurants, coffee shops and
bathrooms.

3.2.3. Scenario 2—Trained Security Personnel Strategy

Passenger service time is reduced in this scenario as highly skilled and proficient
security personnel [38] can perform their tasks with higher efficiency. Their advanced skill
set allows them to perform tasks more competently, making substantial contributions to
expediting the passenger handling processes at key checkpoints, including the check-in and
security control areas. These well-trained personnel possess an in-depth understanding of
security protocols and are adept at streamlining and expediting the requisite procedures.
Their competence enables them to conduct security checks and verification processes
quickly, without compromising on the effectiveness of security measures. Their proficiency
in operating metal detectors, x-ray machines and other screening equipment allow them to
identify items that may pose a security risk. Therefore, a seamless and secure experience
for passengers can be achieved within the airport terminal.

3.2.4. Scenario 3—The Multiple Entry Points Strategy

This scenario is focused on minimizing passenger waiting time in an airport terminal
through the utilization of a multiple entry points strategy. This based on the ticket category,
such that passengers with “business” tickets and gold class members can use dedicated
entry points that are separate from passengers with “economy” tickets. This separation
optimizes the flow of passengers with “business” tickets and gold class members by
reducing congestion and waiting times at the check-in and security control areas. In
addition, passengers with “economy” tickets also benefit from reduced waiting times
because of the decreased congestion at their designated entry points. This grouping
approach [39] to passengers, allowing them to use multiple entry points, is expected to
result in an overall improvement in passengers’ flows and enhance their level of service
and satisfaction.

4. Experiments and Results

4.1. Experimental Setup
We conducted 500 simulation runs for each scenario and the experimental results for
efficient crowd management strategies in an airport terminal are ranked according to their
average performance metrics, i.e., waiting time and level of crowdedness. The primary
objective is to minimize the waiting times for the different processes (i.e., check-in, security
check, shopping/catering/revitalization areas, boarding area, boarding queue) and the
secondary objective is to keep a low crowdedness level for all areas of the airport terminal.
However, these two objectives can be either conflicting or complementary, because the
low crowdedness level may either reduce or increase the waiting time of the different
processes. Thus, for the evaluation of the different proposed strategies, several experiments
are performed for each of the established scenarios (Table 1).
Sustainability 2024, 16, 1346 7 of 13

Table 1. Examined strategies.

Strategy Scenario
Base Case Scenario 0: Stochastic
Technology Integration Scenario 1: Optimized Guidance
Trained Security Personnel Scenario 2: Minimizing Service Time
Multiple Entry Points Scenario 3: Minimizing Waiting Time

4.1.1. Scenario 0—Base Case Strategy

In Scenario 0, we consider 800 passengers that arrive at the airport terminal; 15% of
them either have purchased “business” tickets or are gold class members. A total of 60% of
the passengers have already checked-in online and, thus, the service time in the check-in
area is 4 min for each passenger. Additionally, the service time at security control is 3 min
for each passenger.

4.1.2. Scenario 1—Technology Integration Strategy

In Scenario 1, we keep the same values for all the parameters, as we need to determine
how a mobile app guides the passengers more efficiently within the airport terminal
environment.

4.1.3. Scenario 2—Trained Security Personnel Strategy

In Scenario 2, we consider 800 passengers that arrive at the airport terminal; 15% of
them either have purchased “business” tickets or are gold class members. In addition,
60% of them have already checked-in online. As we are examining the higher efficiency
of trained security personnel, the service time in the check-in area is now 2 min for each
passenger and the service time at security control is 1.5 min for each passenger.

4.1.4. Scenario 3—The Multiple Entry Points Strategy

Scenario 3 examines the effects of the utilization of a multiple entry points strategy, and
of the 800 passengers that arrive at the airport terminal, we considered 40% of passengers
to have either purchased “business” tickets or been gold class members. The service time is
4 min in the check-in area and 3 min at security control, while 60% of all passengers have
already checked-in online (the same as the base case).

4.2. Computational Results

Figure 3 summarizes the results obtained in the different scenarios in terms of mini-
mizing the waiting time during the different processes within the airport terminal. Based
on the results reported, we realized that Scenario 3 presents the minimum waiting time
overall, i.e., 47 min on average per passenger. However, each of the different scenarios
manages to achieve a waiting time reduction within the terminal processes, i.e., the check-in
area, security check area, shopping, catering and revitalization areas, boarding area and
the boarding queue during embarkment. More specifically, Scenario 1 minimizes the total
waiting time in the shopping, catering and revitalization areas, i.e., 1 min on average per
passenger. Scenario 2 minimizes the total waiting times in the check-in and security areas,
i.e., 1 min on average per passenger and 17 min on average per passenger, respectively.
Correspondingly, Scenario 3 minimizes the total waiting time in the boarding area, i.e.,
20 min on average per passenger, and in the boarding queue during embarkment, i.e.,
3 min on average per passenger.
Sustainability 2024, 16, x FOR PEER REVIEW 8 of 13
Sustainability 2024, 16, 1346 8 of 13

Figure
Figure 3.
3. Waiting
Waiting times
times within
within the
the airport terminal for
airport terminal for each
each scenario.
scenario.

Regarding
Regarding the thelevel
levelofofcrowdedness,
crowdedness, Figure
Figure 4 summarizes
4 summarizes the the results
results obtained
obtained for
for each
each part
part of theofshopping,
the shopping, catering
catering and and revitalization
revitalization areas.
areas. We We cancan observe
observe thatthat
thethe level
level of
of crowdedness
crowdedness is minimized
is minimized by implementing
by implementing Scenario
Scenario 1 in 1allincases
all cases
except except
for theformaximum
the max-
imum
number number of passengers
of passengers waitingwaiting
for WC2, for which
WC2, which is 14 passengers
is 14 passengers in total.inThe total. The mini-
minimum is
mum
13 is 13 passengers
passengers (same as(same the baseas the base
case) and case)
this and thisachieved
is also is also achieved
by Scenarios by Scenarios
2 and 3. 2Moreand
3. More specifically,
specifically, Scenario Scenario
1 achieves1 aachieves
reduction a reduction
of up to 25% of up to 25%
in the totalinmaximum
the total number
maximum of
number
passengers of passengers
in different in different
stores storestocompared
compared the base case to the base case
(Scenario 0).(Scenario
It is worth 0).mentioning
It is worth
mentioning here that
here that Scenario Scenario
2 results in an2 results
increased in an increased
level level of crowdedness
of crowdedness for the shopping, for the shop-
catering
ping, catering andareas
and revitalization revitalization
of 9%. This areas of 9%. ThisasisScenario
is reasonable, reasonable, as Scenario
2 improves 2 improves
the efficiency of
the efficiency of the check-in and security control areas, which belong to the first part of
the check-in and security control areas, which belong to the first part of processes in an
processes in an airport
airport terminal. Hence,terminal.
passengers Hence,
gain passengers
some time to gain
spendsome timeshopping,
in the to spend catering
in the shop-and
ping, catering and revitalization areas before their embarkment.
revitalization areas before their embarkment.
In
In more detail, Figure
more detail, Figure 5 5 presents
presents how how each each scenario
scenario affects
affects the the crowdedness
crowdedness level level
within the
within theterminal.
terminal.The Thehighest
highestreductions
reductions were
were achieved
achieved usingusing Scenario
Scenario 1. The
1. The maxi-
maximum
mum
number number of passengers
of passengers is minimized
is minimized in Restaurant
in Restaurant 3 (a 57%3 reduction)
(a 57% reduction)compared compared
to the baseto
the base case scenario. It is worth mentioning that Scenario 2 increases the crowdedness
case scenario. It is worth mentioning that Scenario 2 increases the crowdedness level within
level within the
the terminal, andterminal,
the highestandincrease
the highestin theincrease
maximum in thenumber
maximum number ofappeared
of passengers passengers in
appeared in Duty-Free Shop 2. Reductions are also achieved using Scenario 3, and its
Duty-Free Shop 2. Reductions are also achieved using Scenario 3, and its highest decrease
in the crowdedness
highest decrease in the level (13%) is achieved
crowdedness in Coffee
level (13%) Shop 3. in Coffee Shop 3.
is achieved
Overall, itit isisapparent
Overall, apparentthat thatthe
theoptimal
optimalstrategy
strategyisis toto enable
enable technology
technology integration
integration in
in the processes, which will guide passengers more efficiently.
the processes, which will guide passengers more efficiently. This strategy keeps This strategy keeps thethe
crowdedness level
crowdedness level low,
low, with
with aa maximum
maximum number number of of 14
14 passengers
passengers waiting waiting for for service
service at at
WC1 and WC2. Moreover, the crowdedness level is remarkably
WC1 and WC2. Moreover, the crowdedness level is remarkably reduced (about 25% less) reduced (about 25% less)
for all
for all processes
processes in in the
the shopping, catering and
shopping, catering and revitalization
revitalization areas areas (i.e.,
(i.e., 144
144 passengers
passengers in in
total for all processes). Additionally, the cumulative total waiting time for all the processes
total for all processes). Additionally, the cumulative total waiting time for all the processes in
in the airport terminal is 48 min on average per passenger, only 1 min more than the
the airport terminal is 48 min on average per passenger, only 1 min more than the optimum
optimum waiting time achieved by Scenario 3. On the other hand, Scenario 3 results in
waiting time achieved by Scenario 3. On the other hand, Scenario 3 results in a higher crowd-
a higher crowdedness level of up to 34% compared to Scenario 1 (i.e., 193 passengers in
edness level of up to 34% compared to Scenario 1 (i.e., 193 passengers in total for all processes
total for all processes in the shopping, catering and revitalization areas). This underscores
in the shopping, catering and revitalization areas). This underscores that the efficacy of tech-
that the efficacy of technology integration is not only minimizing waiting times but also
nology integration is not only minimizing waiting times but also optimizing overall passenger
optimizing overall passenger flow and, as such, enhancing passenger satisfaction within
flow and, as such, enhancing passenger satisfaction within the airport terminal.
the airport terminal.
Sustainability 2024, 16, x FOR PEER REVIEW 9 of 13

Sustainability 2024, 16, 1346 9 of 13

Figure
Figure 4.
4. Crowdedness
Crowdedness level
level within
within the
the airport
airport terminal
terminal for
for each
each scenario.
scenario.
Sustainability 2024, 16, x FOR PEER REVIEW 10 of 13
Sustainability 2024, 16, 1346 10 of 13

Figure 5. Percentage of reduction/increase in the crowdedness level within the airport terminal for each scenario compared to the base case (Scenario 0).
Figure 5. Percentage of reduction/increase in the crowdedness level within the airport terminal for each scenario compared to the base case (Scenario 0).
Sustainability 2024, 16, 1346 11 of 13

The experiments illustrate the relevance of the simulation model for the representation
of passengers’ flows in applying different management strategies and underlining the
role of the proposed decision-making tool. Based on computational experience, Scenario
3 excels in total waiting time (i.e., 4 min less compared to the base case) and Scenario 1
excels in the crowdedness level (i.e., 49 passengers less compared to the base case). To
that end, we observed that, in this case, the best overall improvement is achieved by
Scenario 1 as it also reduces the waiting time by 4% compared to the base case. On the
other hand, Scenario 3 manages to reduce the crowdedness level by only 3%. Clearly, a
more efficient guidance of passengers using a mobile app could offer improved service
quality to passengers and reduce risks due to a high level of crowdedness in the airport
terminal. Therefore, Scenario 3 could offer enhanced service quality and mitigate the risks
associated with high levels of crowdedness within an airport terminal.
To sum up, Scenario 1, the “Technology Integration” strategy, provides a more efficient
and pleasant experience to passengers and reflects a commitment to adopting eco-friendly
and innovative solutions aligned with modern sustainability practices. Satisfied passengers
are more likely to be engaged in sustainable behaviors and foster support for subsequent
initiatives within the airport terminal environment. The maintenance of a low crowdedness
level allows for the more efficient identification and mitigation of potential risks, contribut-
ing to improved operational resilience. This is crucial for managing unexpected events,
reducing disruptions, and maintaining a sustainable and consistent level of service for
passengers.

5. Conclusions
This paper presents a simulation model developed for a decision-making tool in order
to analyze different crowd management strategies in the context of the risks posed by pas-
senger flows in airport terminals. Different usage scenarios were developed to investigate
the different management strategies, to optimize passenger routing and minimize service
and waiting times. The scenarios of the simulation model can be updated according to
current requirements, which enriches the academic discourse on dynamic modeling in
airport operations. This flexibility of the simulation model is also important for manage-
ment, as it can be adapted to the specific needs of an airport terminal. Managers can tailor
crowd management strategies to the unique characteristics and current challenges of their
terminals, ensuring a customized and effective approach to managing passenger flows that
promotes long-term sustainability.
One research direction worth pursuing is the integration of real-time data into the
simulation model to improve its accuracy and responsiveness. Research into human-
centered terminal design should investigate how physical layouts and design elements
influence crowd dynamics and passenger satisfaction. As research moves towards more
realistic and richer problems, the development of real-world models that capture the use
of strategies that improve sustainability and resilience is of great interest, as they could
mitigate the potential risks in airport terminals more efficiently.

Author Contributions: Conceptualization, V.K.; methodology, A.A. (Afroditi Anagnostopoulou);

Software, D.T. and E.S.; Validation, E.S. and V.K.; Formal analysis, D.T. and A.A. (Afroditi Anagnos-
topoulou); Resources, A.A. (Attila Akac); Data curation, E.S.; Writing—original draft, A.A. (Afroditi
Anagnostopoulou) and A.A. (Attila Akac); Visualization, D.T. and A.A. (Attila Akac); Supervision,
V.K.; Project administration, A.A. (Afroditi Anagnostopoulou) and V.K.; Funding acquisition, V.K. All
authors have read and agreed to the published version of the manuscript.
Funding: This research was carried out as part of the HAIKU project. This project has received
funding from the European Union’s Horizon Europe research and innovation programme HORIZON-
CL5-2021-D6-01-13, under grant agreement no 101075332, but this document does not necessarily
reflect the views of the European Commission.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Sustainability 2024, 16, 1346 12 of 13

Data Availability Statement: No new data were created or analyzed in this study. Data sharing is
not applicable to this article.
Conflicts of Interest: The authors declare no conflict of interest. The funders had no role in the design
of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or
in the decision to publish the results.

References
1. IATA. Worldwide Slot Guidelines; International Air Transport Association: Montreal, QC, Canada, 2019; p. 58.
2. Boger, C.A., Jr.; Varghese, N.; Rittapirom, S.D. The impact of the September 11 attacks on airline arrivals and conventions in nine
major US cities. J. Conv. Event Tour. 2005, 7, 21–41. [CrossRef]
3. Alomar, I.; Belitskaya, M.; Belitskaya, A. Comparative Statistical Analysis of Airport Flight Delays for the Period 2019–2020.
Almaty International Air-port Case Study. In Proceedings of the International Conference on Reliability and Statistics in
Transportation and Communication, Riga, Latvia, 14–15 October 2021; pp. 110–124.
4. Zhou, L.; Chen, Z. Measuring the performance of airport resilience to severe weather events. Transp. Res. D Transp. Environ. 2020,
83, 102362. [CrossRef]
5. Tubbs, J.; Meacham, B. Egress Design Solutions: A Guide to Evacuation and Crowd Management Planning; John Wiley & Sons: Hoboken,
NJ, USA, 2007; pp. 365–400.
6. Stojić, S.; Dvořáková, T.; Had, P.; Vittek, P. Advanced Airport Virtual Queueing Utilizing Smart City’s Infrastructure. In
Proceedings of the Smart City Symposium Prague (SCSP), Prague, Czech Republic, 25–26 May 2023; pp. 1–5.
7. Dvořáková, T.; Svítek, M.; Voráčová, Š.; Řehoř, V.; Vittek, P. Smart Airports–Balancing Queue Management and Anti-epidemic
Measures. In Proceedings of the 2022 Smart City Symposium Prague (SCSP), Prague, Czech Republic, 26–27 May 2022; pp. 1–6.
8. Rodríguez-Sanz, Á.; de Marcos, A.F.; Pérez-Castán, J.A.; Comendador, F.G.; Valdés, R.A.; Loreiro, Á.P. Queue behavioural patterns
for passengers at airport terminals: A machine learning approach. J. Air Transp. Manag. 2021, 90, 101940. [CrossRef]
9. Schultz, M.; Schulz, C.; Fricke, H. Passenger Dynamics at Airport Terminal Environment. In Pedestrian and Evacuation Dynamics
2008; Klingsch, W., Rogsch, C., Schadschneider, A., Schreckenberg, M., Eds.; Springer: Berlin/Heidelberg, Germany, 2010;
pp. 381–396.
10. Bouyakoub, S.; Belkhir, A.; Bouyakoub, F.M.H.; Guebli, W. Smart airport: An IoT-based airport management system. In
Proceedings of the International Conference on Future Networks and Distributed Systems, Cambridge, UK, 19–20 July 2017;
pp. 1–7.
11. Dou, X. Big data and smart aviation information management system. Cogent Bus. Manag. 2020, 7, 1766736. [CrossRef]
12. Rajapaksha, A.; Jayasuriya, N. Smart airport: A review on future of the airport operation. Glob. J. Manag. Bus. Res. 2020, 20, 25–34.
[CrossRef]
13. Bakır, M.; Özdemir, E.; Akan, Ş.; Atalık, Ö. A bibliometric analysis of airport service quality. J. Air Transp. Manag. 2022, 104,
102273. [CrossRef]
14. Thampan, A.; Sinha, K.; Gurjar, B.R.; Rajasekar, E. Functional efficiency in airport terminals: A review on Overall and Stratified
Service Quality. J. Air Transp. Manag. 2020, 87, 101837. [CrossRef]
15. Dong, X.; Zhu, X.; Zhang, J. Departure flight delay prediction due to ground delay program using Multilayer Perceptron with
improved sparrow search algorithm. Aeronaut. J. 2023, 127. [CrossRef]
16. Kim, T.H.; Wu, C.L. Methodology for defining the new optimum level of service in airport passenger terminals. Transp. Plan.
Technol. 2021, 44, 378–399. [CrossRef]
17. Mekić, A.; Mohammadi Ziabari, S.S.; Sharpanskykh, A. Systemic agent-based modeling and analysis of passenger discretionary
activities in airport terminals. Aerospace 2021, 8, 162. [CrossRef]
18. Janssen, S.; Sharpanskykh, A.; Curran, R. Agent-based modelling and analysis of security and efficiency in airport terminals.
Transp. Res. Part C Emerg. 2019, 100, 142–160. [CrossRef]
19. Boc, K.; Štimac, I.; Pivac, J.; Bračić, M. An Empirical Investigation: Does New Airport Terminal Infrastructure Improve the
Customer Experience? Sustainability 2023, 15, 13188. [CrossRef]
20. Di Mascio, P.; Moretti, L.; Piacitelli, M. Airport landside sustainable capacity and level of service of terminal functional subsystems.
Sustainability 2020, 12, 8784. [CrossRef]
21. Wu, C.L.; Chen, Y. Effects of passenger characteristics and terminal layout on airport retail revenue: An agent-based simulation
approach. Transp. Plan. Technol. 2019, 42, 167–186. [CrossRef]
22. Kalakou, S.; Moura, F. Analyzing passenger behavior in airport terminals based on activity preferences. J. Air Transp. Manag.
2021, 96, 102110. [CrossRef]
23. da Silva, L.M.; Borille, G.M.R.; da Silva Pinto, M.C.G. The effect of arrival time of travelers at the airport on consumption in
commercial establishments. J. Retail. Consum. Serv. 2022, 68, 103064. [CrossRef]
24. Gajewicz, Ł.; Walaszczyk, E.; Nadolny, M.; Nowosielski, K. Criteria of quality assessment of regional airport services-A very last
picture before the COVID-19 pandemic. J. Air Transp. Manag. 2022, 103, 102231. [CrossRef] [PubMed]
Sustainability 2024, 16, 1346 13 of 13

25. Bakır, M.; Akan, Ş.; Özdemir, E.; Nguyen, P.H.; Tsai, J.F.; Pham, H.A. How to achieve passenger satisfaction in the airport?
Findings from regression analysis and necessary condition analysis approaches through online airport reviews. Sustainability
2022, 14, 2151. [CrossRef]
26. Abbasi, K.M.; Khan, T.A.; Haq, I.U. Framework for Integrated Use of Agent-Based and Ambient-Oriented Modeling. Mathematics
2022, 10, 4157. [CrossRef]
27. Karagiannis, I.; Vouros, P.; Skouloudis, A.; Evangelinos, K. Sustainability reporting, materiality, and accountability assessment in
the airport industry. Bus. Strategy Environ. 2019, 28, 1370–1405. [CrossRef]
28. Raimundo, R.J.; Baltazar, M.E.; Cruz, S.P. Sustainability in the Airports Ecosystem: A Literature Review. Sustainability 2023,
15, 12325. [CrossRef]
29. Lee, S.; Park, J.-W.; Chung, S. The Effects of Corporate Social Responsibility on Corporate Reputation: The Case of Incheon
International Airport. Sustainability 2022, 14, 10930. [CrossRef]
30. Anagnostopoulou, A. Corporate sustainability in freight transport based on European Commission strategy. In Proceedings of
the 5th QUAESTI Scientific Virtual Conference-Multidisciplinary Studies and Approaches, Virtual, 9–16 December 2017.
31. Greer, F.; Rakas, J.; Horvath, A. Airports and environmental sustainability: A comprehensive review. Environ. Res. Lett. 2020,
15, 103007. [CrossRef]
32. Costabile, F.; Angelini, F.; Barnaba, F.; Gobbi, G.P. Partitioning of black carbon between ultrafine and fine particle modes in an
urban airport vs. urban background environment. Atmos. Environ. 2015, 102, 136–144. [CrossRef]
33. Mohamed, M.; Gomaa, H.; El-Sherif, N. Exploring the Potentiality of Applying Smart Airport Technologies in Egyptian
International Airports. Int. J. Herit. Tour. Hosp. 2018, 12, 122–129. [CrossRef]
34. Sreenath, S.; Sudhakar, K.; Yusop, A.F. Sustainability at airports: Technologies and best practices from ASEAN countries. J.
Environ. Manag. 2021, 299, 113639. [CrossRef] [PubMed]
35. Eid, A.; Salah, M.; Barakat, M.; Obrecht, M. Airport Sustainability Awareness: A Theoretical Framework. Sustainability 2022,
14, 11921. [CrossRef]
36. Wayne County Airport Authority. Official Statement 2023. Available online: https://www.metroairport.com/sites/default/
files/business_documents/financial/Wayne%20County%20Airport%20Authority%20Official%20Statement%20for%20Bond%
20Series%202023A-E.pdf (accessed on 10 October 2023).
37. Aeropuerto Internacional Reina Beatrix. Annual Report 2022. Available online: https://www.airportaruba.com/storage/app/
media/PDF/AAA-AnnualReport2022.pdf (accessed on 27 August 2023).
38. ICAO. Safety Oversight Manual, 2nd ed.; ICAO Doc 9734; ICAO: Montreal, QC, Canada, 2006; pp. 31–32.
39. Wong, J.T.; Liu, T.C. Development and application of an airport terminal simulation model—A case study of CKS airport. Transp.
Plan. Technol. 1998, 22, 73–86. [CrossRef]

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.