Lin - Cheng - Thesis 1 PDF
Lin - Cheng - Thesis 1 PDF
Lin - Cheng - Thesis 1 PDF
ODELLING
IRPORT
P
ASSENGER G
ROUP
D
YNAMICS
U
SING AN
A
GENT
- B
ASED
M
ETHOD
Lin Cheng
Submitted in fulfilment of the requirements for the degree of
Master of Engineering (Research)
Science and Engineering Faculty
Queensland University of Technology
June 2014
Keywords
Agent-based model Airport Airport design Evacuation plan Group dynamics Passenger flow
Pedestrian model Simulation
Modelling Airport Passenger Group Dynamics Using an Agent-Based Method i
ii Modelling Airport Passenger Group Dynamics Using an Agent-Based Method
Abstract
Passenger traffic in an airport reflects the level of economic development,
business activity and tourism of a city. A good passenger experience is likely to
result in repeat visits, which not only generate airport’s financial profit, but also
satisfy the need of other stakeholders such as operating airlines, retailers, passengers
and visitors. Hence, passenger experience has become a major factor that influences
the success of an airport. In this context, passenger flow simulation has become a
significant approach in designing and managing airports.
The literature review in this thesis revealed that grouping is a common
phenomenon among pedestrians. However, most research failed to consider the
group dynamics when developing pedestrian flow models. In order to reflect more
realistic passenger flow conditions, the group dynamics must be included in the
model.
An agent-based model is a feasible and effective approach to model passenger
movements in airports. Unlike many models that treat passengers as individual
agents, the proposed model in this thesis incorporates group behaviour attributes as
well and evaluates the simulation performance of passenger movement within
airports. Results from experiments show that incorporating group behaviour,
particularly the interactions with fellow travellers and wavers can have significant
influences on the performance and utilisation of services in airport terminals. The
impacts can be seen in terms of dwell time at each processing unit, discretionary
activity preference, and the level of service (LOS) at processing areas.
Based on the airport passenger flow model that includes group dynamics, a
case study of an airport evacuation event has been conducted. The simulation results
show that the evacuation time can be influenced by passenger group dynamics. The
model also provides a convenient way to design airport evacuation strategy and
examine its efficiency.
For airport designers and operators, the model also provides a convenient way
to investigate the effectiveness of space design and service allocations, which may
contribute to the enhancement of passenger airport experiences. The model was
Modelling Airport Passenger Group Dynamics Using an Agent-Based Method iii
created using AnyLogic software and it was initialised using the data obtained
through previous research existing in the literature.
The main contributions of this thesis are embodied in the following four
aspects: (1) improve the understanding of group dynamics among pedestrians; (2)
provide a more realistic agent-based passenger flow model by incorporating group
dynamics; (3) demonstrate the influence of group dynamics on passenger flow in an
airport departure terminal; and (4) introduce the potential application of the agent-
based pedestrian flow model in design and management of pedestrian facilities.
iv Modelling Airport Passenger Group Dynamics Using an Agent-Based Method
Table of Contents
Keywords .................................................................................................................................................i
Abstract ................................................................................................................................................. iii
Table of Contents .................................................................................................................................... v
List of Figures ...................................................................................................................................... vii
List of Tables .........................................................................................................................................ix
List of Abbreviations ............................................................................................................................... x
Statement of Original Authorship ..........................................................................................................xi
Acknowledgements .............................................................................................................................. xii
Publications ......................................................................................................................................... xiii
CHAPTER 1: INTRODUCTION ....................................................................................................... 1
1.1 Background .................................................................................................................................. 1
1.2 Knowledge Gap ........................................................................................................................... 2
1.3 Research Aims and Scope ............................................................................................................ 3
1.4 Methodology ................................................................................................................................ 3
1.5 Thesis Outline .............................................................................................................................. 4
CHAPTER 2: LITERATURE REVIEW ........................................................................................... 5
2.1 The Need for Modelling Pedestrians............................................................................................ 5
2.2 Pedestrian Modelling Methods .................................................................................................... 6 2.2.1 Model
Classification ......................................................................................................... 6 2.2.2 Social Force Models
......................................................................................................... 7 2.2.3 Cellular Automata Models
................................................................................................ 9 2.2.4 Agent-Based Models
...................................................................................................... 11 2.2.5 Combination of Modelling Approaches
.......................................................................... 14 2.2.6 Validation of Pedestrian Models
..................................................................................... 16
2.3 Pedestrian Group Dynamics....................................................................................................... 18 2.3.1 Ubiquitous
Social Groups ............................................................................................... 18 2.3.2 Group Size in Statistic Models
....................................................................................... 20 2.3.3 Pedestrian Speed and Group Size
................................................................................... 22 2.3.4 Walking Behaviour of Groups
........................................................................................ 24
2.4 Airport Passenger Flow Modelling ............................................................................................ 26
2.5 Evacuation Models..................................................................................................................... 28
2.6 Pedestrian Modelling Software .................................................................................................. 30
2.7 Chapter Summary ...................................................................................................................... 32
CHAPTER 3: DEVELOPING AN AGENT-BASED PASSENGER FLOW MODEL WITH GROUP
DYNAMICS IN AN AIRPORT TERMINAL ................................................................... 35
3.1 Introduction ................................................................................................................................ 35
3.2 Airport Passengers ..................................................................................................................... 35 3.2.1 Arrival in
the Airport ...................................................................................................... 35 3.2.2 Pedestrian Group Assemble
............................................................................................ 39 3.2.3 Pedestrian Characteristics
............................................................................................... 40
3.3 Pedestrian Interactions ............................................................................................................... 43
Modelling Airport Passenger Group Dynamics Using an Agent-Based Method v
3.4 Airport Environment and Passenger Interaction With the Airport Services .............................. 45 3.4.1 Pedestrian
Interaction at Check-in Area ......................................................................... 47 3.4.2 Passenger Interaction at Security
Control ....................................................................... 48 3.4.3 Passenger Interaction at Customs
................................................................................... 50 3.4.4 Passenger Discretionary Activities
................................................................................. 51
3.5 Model Structure and Model Validation ..................................................................................... 53
3.6 Chapter Summary ...................................................................................................................... 65
CHAPTER 4: THE IMPACT OF GROUP DYNAMICS ON AIRPORT PASSENGER ACTIVITIES 67
4.1 Introduction................................................................................................................................ 67
4.2 Model Configuration.................................................................................................................. 67
4.3 Pedestrian Behaviour at Check-in Process ................................................................................. 68
4.4 Passenger Behaviour at Security and Customs .......................................................................... 72
4.5 Discretionary Activities and Retail Choices .............................................................................. 76
4.6 Result Analysis and Discussion ................................................................................................. 78
4.7 Chpater Summary ...................................................................................................................... 79
CHAPTER 5: CASE STUDY – IMPACT OF PASSENGER GROUP DYNAMICS ON AIRPORT
EVACUATION PROCESS ............................................................................................. 81
5.1 Introduction................................................................................................................................ 81
5.2 Configuration and Procedure During Evacuation ...................................................................... 82
5.3 Behaviour Responses to Emergency Evacuation ....................................................................... 84
5.4 Results and Analysis .................................................................................................................. 85 5.4.1
Distribution of Passengers in the Airport Terminal ........................................................ 85 5.4.2 Evacuation Time
............................................................................................................. 88
5.5 Discussion .................................................................................................................................. 91
5.6 Chapter Summary ...................................................................................................................... 92
CHAPTER 6: CONCLUSIONS ........................................................................................................ 93
6.1 Introduction................................................................................................................................ 93
6.2 Thesis Summary ........................................................................................................................ 93
6.3 Research Contribution ............................................................................................................... 94
6.4 Research Limitations ................................................................................................................. 96
6.5 Recommandtions ....................................................................................................................... 96
BIBLIOGRAPHY ............................................................................................................................... 99
APPENDICES ................................................................................................................................... 111
A. Flight Schedule ........................................................................................................................ 111
B. IATA Level of Service (LOS) Framework .............................................................................. 112
C. Pedestrian Density Map of the Airport Departure Terminal .................................................... 113
vi Modelling Airport Passenger Group Dynamics Using an Agent-Based Method
List of Figures
Figure 2-1 Illustration of pedestrian movement in 4 consecutive time steps (Blue, et al., 1997). ......... 10
Figure 2-2 Pedestrian moving on hexagonal cells (Köster, et al., 2011). .............................................. 11
Figure 2-3 General elements of an agent-based model (Macal & North, 2011).................................... 13
Figure 2-4 Representative simulation result of two intersecting pedestrian streams using the
social force model (Helbing, et al., 2005). ........................................................................... 17
Figure 2-5 Illustration of the strip formation in two intersecting pedestrian streams (Ando, et
al., 1988)............................................................................................................................... 17
Figure 2-6 The sizes and proportions of subgroups within a crowd (Singh, et al., 2009). .................... 20
Figure 2-7 Observed group size distribution and zero-truncated Poisson fit (gray curve)
(Moussaïd, et al., 2010). ....................................................................................................... 21
Figure 2-8 Effects of group size on pedestrian walking speed (Moussaïd, et al., 2010). ...................... 22
Figure 2-9 Group size interdependencies regarding to speed (Schultz, et al., 2010). ........................... 23
Figure 2-10 The avoidance action taken by people walking straight towards another (Singh, et
al., 2009)............................................................................................................................... 25
Figure 2-11 Illustration of the measurement method (Moussaïd, et al., 2010). ..................................... 26
Figure 2-12 Group formations according to Moussaïd, et al. (2010) (Karamouzas & Overmars,
2010). ................................................................................................................................... 26
Figure 2-13 Agent-based modelling software [Macal and North as cited in (Ma, 2013)]. .................... 31
Figure 3-1 Pedestrian classification according to role and travel purpose in the simulation
model. ................................................................................................................................... 36
Figure 3-2 Example of relationship of arrival time for enplaning passengers and type of flight
(Ashford, et al., 2011b). ....................................................................................................... 36
Figure 3-3 Accumulative passenger arrival pattern in an airport at three different time periods
of a day. ................................................................................................................................ 38
Figure 3-4 Flow chart of passenger generating process. ....................................................................... 39
Figure 3-5 Pedestrian assemble process. ............................................................................................... 39
Figure 3-6 Overview of the terminal (a) landside of the terminal; (b) airside of the terminal. ............. 46
Figure 3-7 Airport departure processes. ................................................................................................ 47
Figure 3-8 Illustration of airport check-in area and check-in process. .................................................. 48
Figure 3-9 Illustration of airport security control area and detailed processing sequence. ................... 49
Figure 3-10 Illustration of airport customs area and detailed processing sequence. ............................. 50
Figure 3-11 Passenger activity and decision making process in an international departure
terminal. ............................................................................................................................... 54
Figure 3-12 Overview of airport departure terminal simulation environment (landside of the
terminal). .............................................................................................................................. 56
Figure 3-13 Overview of airport departure terminal simulation environment (airside of the
terminal). .............................................................................................................................. 57
Figure 3-14 Illustration of pedestrian dynamics at check-in area. ......................................................... 58
Figure 3-15 Illustration of pedestrian dynamics at security area. .......................................................... 59
Figure 3-16 Illustration of pedestrian dynamics at customs area. ......................................................... 60
Modelling Airport Passenger Group Dynamics Using an Agent-Based Method vii
Figure 3-17 Illustration of pedestrian dynamics at discretionary area. ................................................. 61
Figure 3-18 Simulation data of passenger dwell time distribution in the airport departure
terminal. ............................................................................................................................... 62
Figure 3-19 Simulation data of passenger dwell time at airport processing activities. ......................... 63
Figure 3-20 Simulation data of airport discretionary activities and auxiliary airport operation
status. ................................................................................................................................... 64
Figure 4-1 Facilitation and overall congestion at check-in for three different scenarios.
Passenger travelling: (a) alone; (b) in groups; (c) in groups with wavers. ........................... 69
Figure 4-2 Pedestrian density map of check-in area for three different scenarios. Passenger
travelling: (a) alone; (b) in groups; (c) in groups with wavers. ............................................ 70
Figure 4-3 Pedestrian density at check-in area during check-in opening hours. ................................... 71
Figure 4-4 Regroup, queue and dwell times at the check-in process for the three different
scenarios. .............................................................................................................................. 72
Figure 4-5 Passenger’s behaviour at security process. Passenger travelling: (a) alone; (b) in
groups. .................................................................................................................................. 73
Figure 4-6 Regroup, queue and dwell times at (a) security and (b) customs for the two different
scenarios. .............................................................................................................................. 74
Figure 4-7 Density maps at security and customs area for the two different scenarios. Passenger
travelling (a) alone; (b) in groups. ........................................................................................ 75
Figure 4-8 Pedestrian density in the waiting area between security and customs. ................................ 76
Figure 4-9 Passenger discretionary time in airport for three different scenarios. Passenger
travelling: (a) alone; (b) in groups; (c) in groups with wavers. ............................................ 77
Figure 4-10 Retail visits in the airport for three different scenarios. Passenger travelling: (a)
alone; (b) in groups; (c) in groups with wavers. ................................................................... 77
Figure 5-1 Airport environment defined in our simulation. The exits are marked as red circles.
(a) check in area and retail (landside); (b) Security, Customs, Boarding and retail (airside).
............................................................................................................................... 83
Figure 5-2 Typical response followed by passengers during evacuation. ............................................. 85
Figure 5-3 Overview of the landside of the terminal during evacuation. .............................................. 86
Figure 5-4 Overview of the airside of the terminal during evacuation. ................................................ 87
Figure 5-5 Average evacuation time of passengers on: (a) level 4, landside; and (b) level 3,
airside for the two different settings. .................................................................................... 89
Figure 5-6 Pedestrian density map during the evacuation process. (a) level 4; (b) level 3. .................. 90
Figure C-1 Pedestrian density map of the airport departure terminal after whole day simulation.
(a) level 4; (b) level 3. ........................................................................................................ 113
viii Modelling Airport Passenger Group Dynamics Using an Agent-Based Method
List of Tables
Table 2-1 Frequency distributions of 18 observations (James, 1953). .................................................. 18
Table 2-2 Mean walking speed (m/s) of pedestrians in different group sizes. ...................................... 24
Table 2-3 Comparison of agent-based modelling toolkits (Ma, 2013). ................................................. 32
Table 3-1 Example of passenger arrival rate at check-in counters in three periods of the day
(IATA, 2004). ....................................................................................................................... 37
Table 3-2 Adjustment of IATA passenger arrival pattern in international airport departure
terminals. .............................................................................................................................. 38
Table 3-3 Passenger arrival time distribution and arrival intervals for the first flight on
timetable. .............................................................................................................................. 39
Table 3-4 Age distribution of global airport passengers (IATA, 2013b). ............................................. 41
Table 3-5 Gender distribution of global airport passengers (IATA, 2013b). ........................................ 41
Table 3-6 Country of residence of airport departure passengers in 2012-2013 financial year in
Australia (ABS, 2013). ......................................................................................................... 41
Table 3-7 Percentage of departure passengers travelling for business and leisure purpose in
2012-2013 financial year in Australia (ABS, 2013). ............................................................ 41
Table 3-8 Influence of age, gender, travel purpose and group size on passenger walking speed
in airport terminals. .............................................................................................................. 42
Table 3-9 Passenger activity preference in airport. ............................................................................... 43
Table 3-10 Time passengers spend interacting with check-in personnel (Kirk, 2013).......................... 48
Table 3-11 The percentage of passengers failing certain mandatory activities at security (Kirk,
et al., submitted). .................................................................................................................. 49
Table 3-12 Time passengers spend in each activity at security (Kirk, 2013). ....................................... 49
Table 3-13 Time passengers spend in each activity at customs (Kirk, 2013). ...................................... 51
Table 3-14 Probabilities for passengers to use public services in airport departure terminal................ 51
Table 3-15 Dwell time distribution for airport discretionary activities (Ma, 2013). ............................. 52
Table 3-16 Discretionary activity occurrence time, location and criteria. ............................................. 52
Table 3-17 Default parameter setting in the simulation. ....................................................................... 53
Table 3-18 Comparisons of queue and dwell times at check-in, security and customs between
the actual time and the simulation. ....................................................................................... 55
Table 4-1 Input parameters of the model. ............................................................................................. 68
Table 5-1 Exits assigned for passengers of different security levels. .................................................... 84
Table 5-2 Pedestrian response time to terminal evacuation signal. ....................................................... 85
Table 5-3 The distribution of agents in the airport terminal under the setting of passengers
travelling: (1) alone; and (2) in groups. ................................................................................ 88
Table A-1 Flight timetable in the model. ............................................................................................ 111
Table B-2 IATA LOS Framework (IATA, 2004). .............................................................................. 112
Table B-3 IATA LOS Congestion Standards (Ashford, Mumayiz, & Wright, 2011a). ...................... 112
Modelling Airport Passenger Group Dynamics Using an Agent-Based Method ix
List of Abbreviations
AAA Australian Airport Association
ABM Agent-based model/modelling
ABMS Agent-based modelling and simulation
ABS Australian Bureau of Statistics
ACCC Australian Competition and Consumer Commission
CA Cellular Automata
CAFE Cellular automata with force essentials
HiDAC High-Density Autonomous Crowds
IATA International Air Transport Association
LOS Level of Service
OPC Outgoing Passenger Card
STD Scheduled Time of Departure
x Modelling Airport Passenger Group Dynamics Using an Agent-Based Method
Modelling Airport Passenger Group Dynamics Using an Agent-Based Method xi
= F⃑
α
0 + ∑ β
F⃑
αβ + ∑ B
F⃑
αB + ∑ i
F⃑
αi .
= F⃑
α
+ fluctuations.
Using the social force model, several observed collective phenomena in
pedestrian crowds have been successfully reproduced. This includes the lane forming
behaviour in crowds and the oscillatory walking pattern at a narrow exit (Helbing &
Molnár, 1995) as well as the mechanisms in escape panic situations (Helbing, Farkas,
& Vicsek, 2000). Helbing, et al. (2005) conclude that the simplicity and small
number of parameters are the main advantages of the social-force-based simulation.
Moreover, it is suggested that those parameters in the model do not need to be
calibrated for each new situation, which makes social force models adaptive when
applying to different simulation environments.
However, some researchers suggested that it is not easy to model heterogeneity
and complex behaviours using social force model since the goals, characteristics and
interactions of pedestrians must be represented through equations (Manenti,
Manzoni, Vizzari, Ohtsuka, & Shimura, 2012). Moreover, simulation update of
Helbing’s model is O(n2) because each agent is influenced by all the other agents.
This may limit the computational ability of the social force model to simulate many
agents (Henein & White, 2005). Another disadvantage of the social force model is
that in high density environments, agents will ‘shake’ or ‘vibrate’ unnaturally
(Pelechano, Allbeck, & Badler, 2007). In spite of these drawbacks, the social force
model is still very popular and has become the foundation of many other subsequent
pedestrian models.
Derived from the social force model, a HiDAC (High-Density Autonomous
Crowds) model had been used to address the problem of simulating high-density
crowds in dynamically changing virtual environments (Pelechano, et al., 2007).
Through analysing the crowd scenes, Mehran, Oyama, and Shah (2009) evaluated
the interaction forces among pedestrians and successfully detected and localised
abnormal behaviours in crowds. Parisi, Gilman, and Moldovan (2009) proposed a
8 Chapter 2: Literature Review
modification of the social force model that overcame the limitations of Helbing’s
model in describing the experimental data of pedestrian flow in normal conditions.
Moussaïd, et al. (2010) had extended Helbing’s model to simulate the walking
behaviour of pedestrian social groups by adding an extra social interaction term into
Helbing’s social force formula.
2.2.3 Cellular Automata Models
A relatively novel model called Cellular Automata (CA) uses intuitive rules
that make the model easy to understand without complex mathematical equations and
thus demand less computation than social force models. In cellular automata models,
space is represented by a uniform grid of cells. At each discrete time step, the values
of variables in each cell are updated according to a set of local rules and the values of
variables in the cells at its neighbourhood (Zheng, Zhong, & Liu, 2009).
The idea of cellular automata was originally introduced by Von Neumann and
Burks (1966) with the purpose of modelling biology self-reproduction. It was then
developed and applied to a wide variety of purposes because physical systems that
containing many discrete elements and local interactions are convenient to convert
into cellular automata (Wolfram, 1983). Cellular automata has been extended to
model pedestrian movement by Blue, Embrechts, and Adler (1997) who analysed
multidirectional microscopic behaviour in a crowded open space. Blue, et al. (1997)
introduced a ‘bump’ rule to avoid pedestrian being locked in one position at high-
density hypothetical floor area. At each time step, each entity in the matrix moves
one step towards its destination according to the sequence of the entity number. If the
forward movement is blocked by the other entity, the pedestrian will sidestep into the
next cell relative to its movement. If the next cell is still occupied, the pedestrian will
‘bump’ the occupant to make sidestep, and so on, until an empty cell is found or one
entity is bumped off the floor matrix (Blue, et al., 1997). Figure 2-1 illustrates the
bumping rule between entities. At time step 2, entity 11 and 12 both desire to move
into cell (2, 2). According to the bumping rule, entity 11 bumps entity 12 into the
adjacent cell of its desired cell at time step 3. The entities move toward their
destinations afterwards. Though Blue’s model is based on simple local rules, it is
instructive that cellular automata can be used to model pedestrian movement and
more realistic pedestrian interactions can be demonstrated by optimising the rule set.
Chapter 2: Literature Review 9
Figure 2-1 Illustration of pedestrian movement in 4 consecutive time steps (Blue, et al., 1997).
Vizzari, Manenti, Ohtsuka, and Shimura (2012) investigated the pedestrian and
group dynamics in crowds. They proposed a simple experiment to test the validity of
the model. It was observed in the experiment that pedestrian pairs can easily form a
line to avoid facing crowds. However, pedestrian groups with bigger numbers of
members had difficulties to form such a line shape, thus they tend to form a
triangular shape which is similar to the ‘V’ shape observed by Moussaïd, et al.
(2010). Then, a simulation model was applied to the experimental scenario and
showed consistent pedestrian dynamics with the previous experiment. After the
validation, the model is further adopted to simulate real-world scenarios.
Chapter 2: Literature Review 17
2.3 PEDESTRIAN GROUP DYNAMICS
2.3.1 Ubiquitous Social Groups
In real life, it is a common phenomenon that many of the pedestrians are
walking in groups. It is easy to identify a group of people through the interactions
and characteristics of the members such as appearance, gender and age. However, the
existing crowd modelling approaches tend to neglect the existence of pedestrian
groups, which leads to less convincing results.
Researchers have been studying the behaviour of pedestrian social groups for
more than half-century. In order to investigate the size determinant in small human
group interaction, James (1953) observed 22,625 pedestrian samples in 18 public
situations in Eugene and Portland, Oregon. The observations were taken in the winter
and spring, 1950. Several different places including public markets, playgrounds,
schools, swimming pools, public beaches were chosen as observation sites. The
diversity of observation times and locations ensures that the observation results
reflect the common behaviour of pedestrians of different roles. The observation
shows that crowds are split into ‘free-forming’ small groups with varying sizes. The
group relationship was identified through the face-to-face interactions such as
gesture, laughter, smiles, talk, play, or work. A total of 15,486 small groups were
recorded in the observation. The observation results are consolidated into Table 2-1.
Group size 1 indicated that individuals are also considered as small groups that only
have one group member. From Table 2-1, it can be calculated that more than half of
the observed population are in groups with at least two members.
Table 2-1 Frequency distributions of 18 observations (James, 1953).
18 Chapter 2: Literature Review
A similar field study was conducted by Aveni (1977), who interviewed 204
celebrating fans in a football event. The findings of the study showed that three
quarters of the crowd were with one or more friends. This result shows higher group
proportion in crowds than James’ observation. The reason for the this difference may
be that the data in this research was collected in a special sports event, which may
not reflect the group behaviour of mundane crowds in a variety of different locations.
In spite of this, this study still suggests that crowds consist of both individual
pedestrians and persons in groups.
It has been decades since the research carried out by James and Aveni. The
social background has greatly changed, so are the ways people communicate and
interact. Therefore, it is necessary to examine whether the crowd preserves group
behaviour in more recent studies. Singh et al. (2009) investigated the behaviour of
pedestrian subgroups by means of filming and observation. The behaviour of crowds
was recorded every 10 seconds for half an hour in the following four locations:
Nottingham train station, Broadmarsh shopping centre, Clumber Street and
Nottingham University campus. Pedestrians in these four locations represent crowds
in three environments: travelling, shopping and working. As can be seen from Figure
2-6, a large percentage of people in crowds are in subgroups of two or more
members. In travelling environment (train station), the percentages of people in
groups are about 55%. In shopping environments (Broadmarsh shopping centre and
Clumber Street), the percentage is about 65%. On university campus where people
study or work, the figure is about 47%. The varying numbers in different observation
locations indicate that the proportion of people in groups can be influenced by the
surrounding environment. One limitation of the observation approach is that the
observation period is not long enough. Therefore it is possible that observers only
captured a small section of the big picture which may cause deviation to the ‘ground
truth’.
Chapter 2: Literature Review 19
Figure 2-6 The sizes and proportions of subgroups within a crowd (Singh, et al., 2009).
Similar findings were discussed in the research of Schultz, et al. (2010), who
recorded and analysed the walking behaviour of passengers in Dresden International
Airport. Figure 2-9 compares the differences in speed between groups with one and
three members. As the author had expected, groups with three members are clearly
22 Chapter 2: Literature Review
slower than groups that only have one member. Since it was mentioned in many
studies that the environment has great influence on pedestrians’ waking speed (Finnis
& Walton, 2008; Harney, 2002; Rastogi, et al., 2011), above research results can
only be applied in airport environment and cannot represent the pedestrian speed at
any other situations wilfully.
Figure 2-9 Group size interdependencies regarding to speed (Schultz, et al., 2010).
For comparison, Table 2-2 summarises the mean walking speeds for pedestrian
group with varying sizes at different locations in previous research. Tarawneh (2001)
investigate the speed of pedestrians when at 27 crosswalks in the Great Amman
Area, Jordan. Results show that pedestrian group size with three or more people walk
much slower than smaller groups when crossing the street. Tarawneh argued that the
reasons could be: (1) pedestrians in small groups (single or couple) feel less secure in
the crosswalk; and (2) larger groups of pedestrians are more likely to engage in
conversations during the walk. However, the author overlooked the particularity of
the environment. In crosswalks, the speed of pedestrians is often constrained by
traffic signals. Moreover, dwelling in the crosswalk may cause potential safety
hazards. Thus, pedestrians in crosswalks tend to finish crossing as fast as possible
and the effect of group interactions is relatively weak during this time. This can be
seen from that individuals or couples were faster than larger groups by only 0.02 m/s,
which is hardly noticeable.
Chapter 2: Literature Review 23
Mean walking speed (m/s)
Source Locations
Group sizes 1 2 3 4 5 >5 Tarawneh (2001) Crosswalk 1.35 1.35 1.33 - - -
Klüpfel (2007)
World Exhibition
1.38 1.28 1.24 1.24 1.22 1.10
Schultz, et al. (2010)
International Airport
1.36 1.06 0.96 - - -
Rastogi, et al. (2011)
Sidewalks - 1.19 1.06 0.91 1.01 0.99 Wide Sidewalks
- 1.13 1.01 0.98 0.90 -
Precincts - 1.09 1.00 1.00 0.89 0.83
Table 2-2 Mean walking speed (m/s) of pedestrians in different group sizes.
Rastogi, et al. (2011) reported that pedestrians travelling in groups walk at an
average speed and almost 73% of the pedestrians who fall behind will catch up with
other group members by increasing their speed. Rastogi also observed an interesting
phenomenon: on sidewalks, pedestrians in large groups (have 5 or more people)
often split into smaller sub-groups in order to avoid incoming pedestrian flow. This
splitting behaviour decreases the group sizes, but increases the speed of pedestrian
sub-groups. Therefore, it can be seen from Table 2-2 that the mean speed of five-
people groups is faster than that of four-people groups on the sidewalks. This
phenomenon is absent on wide sidewalks and precincts because there is no restriction
in space and large groups are not necessary to split into small sub-groups.
2.3.4 Walking Behaviour of Groups
From the filmed evidence, Singh, et al. (2009) discovered the avoidance
behaviour of pedestrians that walk in groups. Figure 2-10 shows the percentage of
avoidance action taken when facing incoming pedestrians. It shows that in most
cases, a person or a group of people will move to the right (34%) or left (44%) to
avoid colliding with others (the ratio of people moving to the left is higher than that
of moving to the right, a possible explanation of this phenomenon is that the
experiment was conducted in UK, where left-hand traffic rule is applied). Only 22%
of the groups will split in order to avoid colliding. This finding indicates a group of
people are trying to remain together if possible. The social connection between group
members creates an invisible bond that forces them to maintain a group structure, as
is described in Helbing’s ‘social force’ theory (Helbing & Molnár, 1995). Singh also
24 Chapter 2: Literature Review
noticed in their research that if a group is split to avoid more than one obstacle, the
group will remain apart and regroup once all the obstacles have been avoided.
Figure 2-10 The avoidance action taken by people walking straight towards another (Singh, et al., 2009).
AnyLogic is the only simulation tool nowadays that supports all common
simulation methodologies which include system dynamics, discrete event, and agent-
based modelling (AnyLogic, 2013). The Pedestrian Library of AnyLogic software is
dedicated to simulate pedestrian flows in ‘physical’ environments. In the software,
pedestrian agents move in continuous space and interact with other pedestrians and
different kind of obstacles. The pedestrian movement is governed by the customised
Chapter 2: Literature Review 31
social force algorithm, while the pedestrian dynamics are defined in the style of
flowcharts.
Table 2-3 Comparison of agent-based modelling toolkits (Ma, 2013).
The overall flow rate of arriving passengers for a single flight has shown some
general arrival pattern. Ashford, Mumayiz, and Wright (2011b) provided an example
of accumulative passenger arrivals before Scheduled Time of Departure (STD) for a
British airport is shown in Figure 3-2. As can be seen in Figure 3-2, almost all
passengers of an international flight had arrived 60 minutes before scheduled flight
departure time. While this figure for domestic flights are 20 minutes.
Figure 3-2 Example of relationship of arrival time for enplaning passengers and type of flight (Ashford, et al.,
2011b).
Pedestrian
Travel Role
Purpose
Pedestrians
Passengers Leisure
Wavers
Business
36 Chapter 3: Developing an Agent-Based Passenger Flow Model with Group Dynamics in an Airport Terminal
The arrival pattern can be influenced by many factors such as airport
accessibility, public transportation, security requirement and traffic situation. Even
for one airport, the arrival pattern can vary at different periods of the day
(Ahyudanari, 2003; Stefanik, Kandera, & Badanik, 2012). In spite of these, airports
share some universal arrival behaviour for international travellers: (1) almost all
passengers had arrived 60 minutes before scheduled flight departure time; (2) leisure
passengers arrive earlier than business passengers; (3) for each flight, its check-in
peak hours usually happen 100 – 120 minutes prior to scheduled departure time; and
(4) peak hours in the morning are relatively shorter but busier compared with that in
the afternoon and in the evening (Ahyudanari & Vandebona, 2005; Ashford, et al.,
2011b; Stefanik, et al., 2012).
International Airport Transport Association (IATA, 2004) showed an example
of the passenger arrival rate at check-in counters at three different periods of a day in
an airport in Table 3-1. For most departure passengers, it is noted that they tend to
complete their processing, travel related activities before other discretionary
activities (Popovic, et al., 2010). Because check-in is the first processing activity in
the departure terminal, it is reasonable to assume that the passenger arrival rate at the
airport equals to the rate they arrive at check-in counters. Assuming the arrival rate
of 00:00 to 06:00 is the average of the time periods: 18:00-24:00 and 06:00-10:00,
the passenger arrival rate at an airport can be transformed to an accumulative
passenger arrival pattern (Figure 3-3).
Period of day
Percentage of passengers per flight arriving at the Check-in counters by 10 minute periods prior to flight departure 120- 110
110- 100
100- 90
90- 80
80- 70
70- 60
60- 50
50- 40
40- 30
30- 20
20- 10
10- 0 06:00- 10:00
0 0 1 2 6 10 20 26 20 12 3 0
10:00- 18:00
0 1 3 8 11 15 17 18 15 10 2 0
18:00- 24:00
3 4 6 9 11 14 15 15 15 7 1 0
Table 3-1 Example of passenger arrival rate at check-in counters in three periods of the day (IATA,
2004).
Compare Figure 3-3 with the passenger arrival pattern presented in Figure 3-2
by Ashford, et al. (2011b), the characteristics of the IATA arrival pattern is more
likely to be a domestic flight (the latest passengers arrive in the airport about 20
minutes prior to flight departure) rather than an international flight. Thus, to obtain
Chapter 3: Developing an Agent-Based Passenger Flow Model with Group Dynamics in an Airport Terminal 37
the distribution of passenger arrival for international flights, the time axis in Table
3-1 need to be shifted forward for 60 minutes, as is shown in Table 3-2.
100 90 ) % ( y
120 110 100 90 80 70 60 50 40 30 20 10
Figure 3-3 Accumulative passenger arrival pattern in an airport at three different time periods of a day.
Period of day
80
c n e
70
u q e r
60
f e v i
50 40
00:00-06:00
06:00-10:00 t a l u m u
30
10:00-18:00
C
20
18:00-24:00
10 0
Arrival time in airport prior to flight departure (min)
Passengers per flight arriving at the airport by 10 minute periods prior to flight departure 180- 170
170- 160
160- 150
150- 140
140- 130
130- 120
120- 110
110- 100
100- 90
90- 80
80- 70
70- 60 00:00- 06:00
1.5% 2% 3.5% 5.5% 8.5% 12% 17.5% 20.5% 17.5% 9.5% 2% 0%
06:00- 10:00
0% 0% 1% 2% 6% 10% 20% 26% 20% 12% 3% 0%
10:00- 18:00
0% 1% 3% 8% 11% 15% 17% 18% 15% 10% 2% 0%
18:00- 24:00
3% 4% 6% 9% 11% 14% 15% 15% 15% 7% 1% 0%
Table 3-2 Adjustment of IATA passenger arrival pattern in international airport departure terminals.
Given the airport timetable (Appendix A) and the passenger arrival pattern
(Table 3-2), it is able to calculate the arrival intervals between passengers and
introduce passengers into the airport successively. A flow chart of the passenger
generating procedure in the model is presented in Figure 3-4. Take the first flight on
the timetable as an example, it departs at 3:30 a.m. and has 222 passengers on board.
According to the passenger arrival pattern in Table 3-2, the pattern in the period of
00:00 to 06:00 should be adopted in this case. Arriving passenger number at each 10
minute period before flight departure can be calculated and the arrival intervals can
be obtained (see Table 3-3). The program can therefore ‘inject’ agents into the
simulation environment at the predetermined rate.
38 Chapter 3: Developing an Agent-Based Passenger Flow Model with Group Dynamics in an Airport Terminal
Figure 3-4 Flow chart of passenger generating process.
Time prior to departure (min)
180- 170
170- 160
160- 150
150- 140
140- 130
130- 120
120- 110
110- 100
100- 90
90- 80
80- 70
70- 60
Passenger number
3 4 8 12 19 27 39 46 39 21 4 0
Arrival rate (passenger/min)
0.3 0.4 0.8 1.2 1.9 2.7 3.9 4.6 3.9 2.1 0.4 0
Table 3-3 Passenger arrival time distribution and arrival intervals for the first flight on timetable.
,
where N i
is the probability for an agent to have the group size of i(i = 1,2,3...).
While an agent is generated, the program calculates its group size i based on
the group size probability N i
in the above equation, and assembles this agent with
subsequent (i − 1) agents as a group. Once the predetermined group size is reached,
and a passenger group is generated, wavers are added into the group. The distribution
of waver number of a passenger group is assumed to be a zero-truncated Poisson
distribution as well, with an expected value: λ = 1. Therefore, a complete pedestrian
group in the model is composed of passengers and wavers. The group assembling
process in the model is illustrated in (Figure 3-5).
Figure 3-5 Pedestrian assemble process.
Chapter 3: Developing an Agent-Based Passenger Flow Model with Group Dynamics in an Airport Terminal 39
'Inject' Read flight
schedule
agents according to arrival rate
Passenger source
Waver source
Calculate passenger group size
Calculate waver number
Group Assemble
Pedestrian group
Choose passenger arrival pattern
Calculate passenger arrival rate
Pedestrian arrival in the model has two modes: ‘Individual arrival’ and ‘Group
arrival’. Under the ‘Individual arrival’ mode, there will be no group assembles in the
model. The group size will be automatically set to 1. This kind of setting enables the
comparison between dynamics of passengers who are travelling alone and those who
travelling in groups and therefore allows the investigation of the effect of group
dynamics. The model is also able to ‘switch off’ wavers in order to examine the
influence of wavers on passenger behaviour.
3.2.3 Pedestrian Characteristics
In the real-world, airport passengers have their personal characteristics which
influence their behaviour and activity choices in the airport. Those characteristics
include physical related factors (e.g. age, gender, mobility) and psychological related
factors (e.g. familiarity with the airport and departure processes). All these factors
need to be considered in order to develop a complete model. Due to limited access to
detailed passenger information in the reference airport, the characteristic data used in
this model is referred to aviation organisation reports, previous research and
assumptions which will be stated later in this section.
Age, gender, residential status and travel purpose are four basic characteristics
of passengers in the model. These four factors can influence advanced passenger
characteristics such as mobility and shopping preference through defining certain
rules. How these rules are defined will be explained later in this section. Table 3-4
and Table 3-5 summarise the distribution of airport passengers’ age and gender
provided by the global passenger survey carried by IATA (2013b). According to the
country of residence, passengers in the model are divided into Australian resident
and overseas visitors. Australian Bureau of Statistics (ABS) provided the information
of departure passengers’ country of residence in 2012-2013 financial year (refer to
Table 3-6) (ABS, 2013). ABS had also investigated passengers’ main reasons for
their journey. Those reasons include convention/conference, business, visiting
friends/relatives, holiday, employment, education, and ‘other & not stated’. For
convenience, these travel reasons are categorised into two travel purpose: business
and leisure (see Table 3-7).
40 Chapter 3: Developing an Agent-Based Passenger Flow Model with Group Dynamics in an Airport Terminal
Age range <25 25-34 35-44 45-54 55-64 >65
Percentage in total passengers 10% 31% 23% 18% 11% 7%
Table 3-4 Age distribution of global airport passengers (IATA, 2013b).
Gender Male Female Percentage in total passengers
59% 41%
Table 3-5 Gender distribution of global airport passengers (IATA, 2013b).
Country of residence
Australian resident
Overseas visitor Percentage in total passengers
58% 42%
Table 3-6 Country of residence of airport departure passengers in 2012-2013 financial year in
Australia (ABS, 2013).
Travel purposes Detailed reasons Percentage Total
Business
Convention/conference 3% Business Employment 10% 2%
15%
Education 1%
Leisure
Visiting friends/relatives 23% Holiday 58% 85% Other & not stated 4%
Table 3-7 Percentage of departure passengers travelling for business and leisure purpose in 2012-2013
financial year in Australia (ABS, 2013).
The four basic characteristic factors: age, gender, country of residence and
travel purpose will be initialised to each agent according to the percentage rate
showed in the above tables. The age and gender are assigned to each agent when the
agent enters the system. Since passenger groups usually share common features of
country of residence and travel purpose, these two factors are initialised to each
agent after the pedestrian group had finished assembling and will assume passengers
in the same group have a common country of residence and travel purpose. Based on
the four basic characteristic factors, two advanced characteristics: speed and activity
preference can be defined.
Previous research had shown that passenger walking speed in airport terminals
can be influenced by passenger characteristics such as age, gender, travel purpose
and group size. Table 3-8 summarises the influence of these four factors on
Chapter 3: Developing an Agent-Based Passenger Flow Model with Group Dynamics in an Airport Terminal 41
passenger walking speed in airport terminals. Calculation of agents’ speeds in the
model has following three steps:
(1) Using Table 3-8, calculate the agent’s speed under each one of the four
factors separately: v
age
, v
gender
, v
purpose
, and v
group
.
(2) The agent’s speed is initialised by the average speed calculated in step 1:
v = average(v
age
,v
gender
,v
purpose
,v
group
).
(3) Agents in a group with group size i will adjust their speeds so that each
agent (numbered by 1, 2, 3... i) can have the same speed:
v
1
= v
2
= ⋯ = v
i
= average(v
1
,v
2
...v
i
).
Influence factor
Source Category
Mean
Standard speed (m/s)
deviation (m/s)
Age Finnis and Walton (2008)
Age < 15 1.38 0.24 15 < age < 30 1.46 0.22 30 < age < 55 1.49 0.23 Age > 55 1.37 0.28
Gender Schultz, et al. (2010)
Male Female 1.40 0.22 1.27 0.22 Travel purpose
Business 1.36 0.22 Leisure 1.00 0.23
Group size
Schultz, et al. (2010)
Group size = 1 1.36 0.23 Schultz, et al. (2010)
Group size = 2 1.06 0.21 Group size >= 3 0.96 0.19
Table 3-8 Influence of age, gender, travel purpose and group size on passenger walking speed in
airport terminals.
, P
gender
, P
purpose
, P
residence
and P
group
.
(2) The probability for the agent to conduct this activity is initialised by the
average speed calculated in step 1:
P = average(P
age
,P
gender
,P
purpose
,P
residence
,P
group
).
(3) Repeat step (1) and (2) to calculate the probability for all the four activities
listed in Table 3-9.
Activities
Influence Factors
Food and beverage (%)
Specialty retails (%)
Duty-Free shops (%)
Airline services (%) Age <25 15 8 8 2 25-35 10 12 10 5 35-50 10 11 12 8 >50 8 10 11 5 Gender Male 10
8 10 5 Female 12 15 12 5 Travel purpose
Business 10 5 6 8 Leisure 13 15 9 5 Country of residence
Australian 10 8 9 7 Foreign 12 12 15 8 Group size 1 8 8 10 5 2 10 12 12 6 >=3 10 15 12 6
Table 3-9 Passenger activity preference in airport.
,p
2
and p
3
× p
2
×
p
3
× p
2
× p
3
,a
2 and a
3
+ a
2
+ a
3
For passengers travelling in groups, they will wait for stalled group members
before they enter check-in queues, and after check-in, they will wait in the waiting
area until the group is reassembled and then move on to other destinations as a
group. If the group contains wavers, wavers will not enter check-in queues. Instead,
they wait outside the queues until passengers finish the check-in process and will
move on together after the group is reassembled.
Figure 3-8 Illustration of airport check-in area and check-in process.
There are possible failures for both passengers and bags. Processing failure
rates for passengers at their first attempts listed in Table 3-11 are estimated through
the observations conducted by Kirk, et al. (submitted). Passengers who fail the metal
detector are required to pass through it again or undertake an extra body check. If
any bags fail the X-ray check, the passenger will be asked to unpack their bag. Time
statistics of each detailed activity at security have been collected by Kirk (2013)
(refer to Table 3-12) and the activity time in the model was set accordingly.
Domain Problem Fail at 1st attempt Security Bags through X-ray 15% Metal detector 10%
Table 3-11 The percentage of passengers failing certain mandatory activities at security (Kirk, et al.,
submitted).
Activity Minimum (min) Maximum (min) Average (min) Unpacking 0 2.58 0.83 Being scanned 1
(second) 5 (second) 2 (second) Interacting with staff 0 1.48 0.25 Repacking 0 7.55 0.83 Extra security
check 0.37 4.02 1.53
Table 3-12 Time passengers spend in each activity at security (Kirk, 2013).
It is unlikely that passengers who travel in a group can finish the security
process simultaneously. Passengers who finish the process earlier will wait for the
remaining group members at the small waiting area between security and customs
processes. There are seven desks with X-ray machines and metal detectors available
Chapter 3: Developing an Agent-Based Passenger Flow Model with Group Dynamics in an Airport Terminal 49
at security checking area. In the simulation, not all seven desks are available at all
times. Available desks can be automatically calculated by the program. In the default
setting, two desks are available at all times and two additional desks are opened each
time the passenger number in the queue increases by fifteen.
3.4.3 Passenger Interaction at Customs
The security control process is closely followed by the customs process. Figure
3-10 shows the sequence of the customs operation process. Similar to the check-in
and security control process, passengers queue before being processed by customs
personnel. Once finished document checking, passengers will wait for their group
members in the waiting area right behind the customs area and then move on to their
next discretionary period. At customs, there are six desks available to process
passengers. In the simulation two desks are available at all times to process
passengers, and two additional desks are opened each time the queue length increases
by fifteen passengers.
Figure 3-10 Illustration of airport customs area and detailed processing sequence.
The model results also suggest that the time passengers spend in queuing can
be influenced by group structure. It can be seen that passengers travelling alone
spend approximately 5 minutes less in the queue when compared with passengers
travelling in groups. A possible explanation for witnessing such a trend could be the
congestion caused by people waiting to regroup with their fellow travellers around
the queuing area. In essence, ignoring group dynamics in agent based modelling may
yield results that may not accurately represent the real-world observations.
4.4 PASSENGER BEHAVIOUR AT SECURITY AND CUSTOMS
Figure 4-5 compares two passenger flow conditions at the same time point in
the simulation. Since wavers are not allowed to enter the airside, the comparison is
only between passengers travelling: (a) alone and (b) in groups. In the model,
72 Chapter 4: The Impact of Group Dynamics on Airport Passenger Activities
passengers who travel in groups wait for stalled group members until all members
complete the security check. Those who finish the check earlier will wait in the
narrow area (marked in red) between security check and customs queue.
(a)
(b)
Figure 4-5 Passenger’s behaviour at security process. Passenger travelling: (a) alone; (b) in groups.
The security checking time depends not only on passengers, but also on their
luggage. If their luggage fails at the X-ray scanner, passengers may be required to
open their luggage for further inspection. Therefore, security checking time for group
members can vary significantly. As a result, time can be long between the first
member and the last member of the group passing through the security check.
Moreover, the waiting members can lead to congestion in the area between security
and customs, since there is no room specially designed for waiting in this area.
Chapter 4: The Impact of Group Dynamics on Airport Passenger Activities 73
Consequently, longer dwell time and queuing time can be found at security and
customs for group travellers (Figure 4-6).
8.00
6.00
4.00
2.00
0.00
Figure 4-7 illustrates the density maps at security and customs area. The
density maps recorded the highest pedestrian density values during the day (24
hours) in this area. The waiting area between security and customs is highlighted in
74 Chapter 4: The Impact of Group Dynamics on Airport Passenger Activities
the red rectangle. Compared to the scenario that passengers are travelling alone,
more serious blockage can be found in the waiting area when passengers are
travelling in groups. Figure 4-8 demonstrates the comparison of pedestrian density in
the waiting area during the full simulation run (24 hours in the simulation). Because
of the waiting behaviour in pedestrian groups, the density values when passengers
are travelling in groups (average 0.14 Ped/ m2) are higher than the scenario that
passengers are travelling alone (average 0.03 Ped/ m2).
(a)
(b)
Figure 4-7 Density maps at security and customs area for the two different scenarios. Passenger travelling (a) alone;
(b) in groups.
Chapter 4: The Impact of Group Dynamics on Airport Passenger Activities 75
1
0.9 ) 2 m / d e p ( y t i s n e d n a i r t s e d e P
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0 200 400 600 800 1000 1200 1400
Simulation time (min)
Alone Group
Figure 4-8 Pedestrian density in the waiting area between security and customs.
300.0
250.0
200.0
150.0
100.0
50.0
0.0
Alone In groups
In groups with wavers Landside 37.4 98.6 174.0 Airside 49.0 110.0 92.6 Total 86.4 208.6 266.6
Figure 4-10 Retail visits in the airport for three different scenarios. Passenger travelling: (a) alone; (b) in groups; (c)
in groups with wavers.
Figure 4-10 compares passenger retail choices for three different scenarios by
counting the shop visits. The visit count in Figure 4-10 calculates the sum of total
number of pedestrians (include 222 passengers and their companions if there are any)
who entered retail shops (includes (1) food and beverage; (2) specialty retails; (3)
duty-free shops; and (4) airline services) on the landside, the airside and the entire
departure terminal respectively. The visit of a passenger group to a shop is counted
as the total number of people in the group. It is obvious that compared with
Chapter 4: The Impact of Group Dynamics on Airport Passenger Activities 77
passengers travelling individually, more visits to the retail shops can be seen on both
landside and airside when passengers travel in groups. On average, when passengers
are travelling alone, there are 37.4 and 49.0 visits for retail shops at landside and
airside, respectively. The figures grow to 98.6 and 110.0 when passengers are
travelling in groups. For those travelling in groups with varying number of wavers,
the numbers are 174.0 and 92.6. It can also be seen in Figure 4-10 that the existence
of wavers contributes to higher probabilities of shopping at the landside. Without
wavers, in either passenger travel alone or with other travel companions, they are
more likely to choose retails at the airside. This can be explained by the fact that
passengers tend to complete their mandatory activities before their discretionary
activities (Popovic, et al., 2010). Therefore, they prefer to pass all necessary
processes such as security and customs and ensure that they have flexible time for
discretionary activities at the airside before boarding an aircraft.
4.6 RESULT ANALYSIS AND DISCUSSION
Through the simulation of the international airport, it is shown that agent-based
simulation can be used to analyse group dynamics of pedestrians in a complex
environment. The results in this study suggest that the group dynamics have
significant influence on passenger activities in the airport terminal in terms of dwell
time and activity preferences and therefore influence the airport operation.
Although the group behaviour in the simulation is defined by simple rules
(refer to Section 3.3 and 3.4), some general conclusions can be made. In airport
processing activities such as check-in, security and customs, the group dynamics can
potentially lead to congestion and longer dwell times. Such scenarios can lead to
congestions and potential flight delays which can contribute to a lower level of
service (LOS) and poor passenger experience. Furthermore, they may also leave the
passengers with less time for discretionary activities which may not be favourable for
airport retail operators. Group dynamics can also be a major factor, if not the only
one, which affects passenger discretionary activities and retail choices. It shows that
passengers with groups are more likely to choose retail activities than those who are
travelling alone; and the presence of wavers can contribute to higher landside retail
opportunities.
78 Chapter 4: The Impact of Group Dynamics on Airport Passenger Activities
4.7 CHPATER SUMMARY
This chapter has analysed the influence of group dynamics on passenger
processing activities (check-in, security, and customs) and discretionary activities
using an agent-based model. Results from both visual and statistical aspects show
that when group dynamics exist, pedestrians behave differently from the situations
where group dynamics are absent in an airport departure terminal. It is a common
phenomenon that many of the pedestrians are walking in groups in the airports.
Therefore, in a complete and realistic pedestrian flow model, the group dynamics
should be considered.
The agent-based model not only assists in understanding pedestrian behaviour
in an airport, but also provides an essential tool to assess the performance of airport
design and the quality of the pedestrian facilities in the terminal. The next chapter
provides an example to illustrate how this agent-based pedestrian flow model can be
used for investigating the effectiveness of an evacuation process in the airport. The
evacuation case study presented in the next chapter considers the group dynamics as
well and analyses the influence of group dynamics on pedestrian evacuation.
Chapter 4: The Impact of Group Dynamics on Airport Passenger Activities 79
80 Chapter 4: The Impact of Group Dynamics on Airport Passenger Activities
Chapter 5: Case Study – Impact of
Passenger Group Dynamics on Airport
Evacuation Process
5.1 INTRODUCTION
The safety of passengers is a major concern to airports. In the event of crises,
having an effective and efficient evacuation process in place can significantly aid in
enhancing passenger safety. Hence, it is necessary for airport operators to have an in-
depth understanding of the evacuation process of their airport terminal. Although
evacuation models have been used in studying pedestrian behaviour for decades,
little research has been done in considering the evacuees’ group dynamics and the
complexity of the environment.
In this chapter, the agent-based model is used to simulate a passenger
evacuation process in an international airport departure terminal. Due to limited
access to detailed evacuation strategy in the airport, part of the evacuation procedure
is based on assumptions. For example, different evacuation exits were allocated to
passengers based on their location and the security level to ensure a more effective
evacuation process. In order to simplify the model, the simulation scenario is an
evacuation drill instead of a real evacuation event where panic behaviour should be
considered. It is assumed that participants of the evacuation practice are only
passengers and their fellow companions, airport staff are not included in the
experiment.
The remainder of this section is organised as follows. Section 5.2 demonstrates
the configuration of the model and the procedure of the evacuation process in the
model. Section 5.3 describes the behaviour response of airport passengers to the
evacuation. Section 5.4 provides the simulation results and analysis, while Section
5.5 concludes the findings using the agent-based simulation and points out the
limitations.
Chapter 5: Case Study – Impact of Passenger Group Dynamics on Airport Evacuation Process 81
5.2 CONFIGURATION AND PROCEDURE DURING EVACUATION
Different from other building environments, an airport is considered as a
complex system that comprises multiple stakeholders and social interactions (Wu &
Mengersen, 2013). For example, before boarding the flight, passengers are required
to pass mandatory processes which include check-in, security process and customs.
Therefore, the security level varies in the airport, which needs special consideration
during the evacuation.
The layout of the airport departure terminal used in the simulation is shown in
Figure 5-1. As can be seen from the terminal layout, three emergency exits (marked
as red circles) are available on both landside (level 4) and airside (level 3) of the
terminal. In the event of an evacuation, passengers will be notified by an emergency
alarm, and then they will make their way to the nearest exit under the guidance of
building wardens and airport staff. Passengers will remain at designated assembly
points until it is safe for them to re-enter the terminal.
In our simulations, we presume there are three security levels (these could be
adapted based on the operating conditions of the airport). Passengers who have not
been examined by the security personnel are categorised as having security level 1
status; passengers that passed security but not the customs have security status level
2; passengers that pass both security and customs possess security status level 3. In
our model, it is assumed that only certain exits are accessible to people depending on
their security level status as described below.
The landside of the terminal is the public area. The crowd on the landside is
treated to be on security level 1, along with all outgoing passengers who have not
cleared the security check. They will choose one exit among the three located on
level 4 that has the minimum walking distance while evacuating the airport.
Situations are more complex on the airside of the terminal. On the airside, there are
two mandatory processes: security and customs and different security levels are
imposed on them. Passengers belonging to security level 2 will evacuate through exit
2 on level 3, and passengers with security level 3 will evacuate through exit 3 on
level 3.
82 Chapter 5: Case Study – Impact of Passenger Group Dynamics on Airport Evacuation Process
(a)
(b)
Figure 5-1 Airport environment defined in our simulation. The exits are marked as red circles. (a) check in area and
retail (landside); (b) Security, Customs, Boarding and retail (airside).
In the simulation, once the emergency ceases passengers returning into the
terminal will keep their security level status intact, so that they can continue to finish
their remaining processes rather than doing them from the beginning. However, the
Chapter 5: Case Study – Impact of Passenger Group Dynamics on Airport Evacuation Process 83
simulation is flexible and different policies can easily be implemented and tested
(e.g. the policy that all passengers must be re-scanned on entry, regardless of their
security levels when they are evacuated). Table 5-1 summarises the corresponding
exits for pedestrians with different security levels.
Security Level Domain Emergency exits
1 Security unchecked
Exit 1,2 and 3 on Level 4 Exit 1 on Level 3
2
Security checked; Customs unchecked
Exit 2 on Level 3
3 Customs checked Exit 3 on Level 3
Table 5-1 Exits assigned for passengers of different security levels.
level 4 (landside)
160.00 140.00 120.00 100.00 80.00 60.00 40.00 20.00 0.00
Response Movement Waiting
Total evacuation Alone 10.03 57.58 26.05 93.68 In groups 28.55 76.04 41.58 146.18
level 3 (airside)
200.00 180.00 160.00 140.00 120.00 100.00 80.00 60.00 40.00 20.00 0.00
Response Movement Waiting
Total evacuation Alone 10.08 98.23 39.85 146.76 In groups 24.29 99.88 54.58 173.45
Chapter 5: Case Study – Impact of Passenger Group Dynamics on Airport Evacuation Process 89
(a)
(b)
Figure 5-6 Pedestrian density map during the evacuation process. (a) level 4; (b) level 3.
On both level 4 and level 3, passengers spent the majority of their evacuation
time on the evacuation movement. Because of the group dynamic, larger groups are
supposed to move slower than small groups or individual travellers (Santos &
Aguirre, 2004; Schultz, et al., 2010). This behaviour is well illustrated on level 4.
Passengers travelling in groups on level 4 spend approximately 20 seconds longer in
moving during the evacuation. However, the movement times under the two different
settings on level 3 are very close. A possible explanation for such a phenomenon is
that the pathways to the exits on level 3 are narrower than those on level 4 (as can be
seen from the highlighted areas in Figure 5-6). Severe congestion was observed all
90 Chapter 5: Case Study – Impact of Passenger Group Dynamics on Airport Evacuation Process
along passageways through to exit on level 3. Thus, passengers who are travelling
alone had to slow down due to congested passageways and their speed was
comparable to those travelling in groups. On the other hand, passengers on level 4
had more open space, which allowed those individual travellers to advance towards
the exit quickly.
5.5 DISCUSSION
The simulation of the evacuation process in the airport terminal shows that the
agent-based model can be applied in evaluating the performance of pedestrian
facilities. Based on passenger’s locations in the airport, three security levels are
differentiated, which require passengers to evacuate through different exits. The
simulation results suggested that passengers with group dynamics spend longer time
in making decisions, moving to the exits and waiting for other group members during
the evacuation.
This simulation technique prevents the potential risks in real practical trials and
reduces research expense. Moreover, the simulation results provide valuable
information such as how passengers react to an evacuation signal, which route to
choose in the evacuation and the average time for passengers to finish the
evacuation. The simulation is also to report congestions through the 3D visual
demonstration during the evacuation. The evacuation model offers an expedient way
for airport managers to propose and test evacuation plans. Given the information of
flight schedule and passenger number, the evacuation simulation can be run at any
time of day and the simulation results will provide valuable information for them to
respond proactively to any potential congestion.
However, a few limitations of this case study need to be acknowledged. First,
the proposed model is not designed for the extreme evacuation situation. Under
extreme cases, there is no guarantee that the pre-defined escaping routes are safe for
the evacuees. Second, the evacuation subjects in the model are all passengers.
However, in the real-world, there are large numbers of airport staff that need to be
considered as well. As part of future research, we would like to consider a phased
evacuation approach i.e. only areas directly threatened will be evacuated first and
areas at lesser risk will be evacuated later. Furthermore, different exit strategies
employed at various airports could also be trialed using this framework. The
Chapter 5: Case Study – Impact of Passenger Group Dynamics on Airport Evacuation Process 91
dissemination of evacuation information among passengers and further addition of
attributes to agents (such as age, gender, spatial cognition) will also be explored
along with the inclusion of airport staff into the model.
5.6 CHAPTER SUMMARY
This chapter demonstrated a case study of simulating an airport evacuation
process using an agent-based model. The evacuation strategy applied in this
simulation divides passengers into three different security levels and evacuates
passengers through different exits. The model is able to capture passenger behaviour
during evacuation visually and collect passenger evacuation time statistically. This
enables airport operators and building planners to design evacuation strategy and
terminal layout according to the simulation results. By comparing evacuation time of
individual passengers and passengers in groups, the impact of group dynamics during
airport evacuation process can be seen. Experimental results demonstrated that group
dynamics can significantly impact pedestrian behaviour during the evacuation and
the total evacuation time in the airport.
92 Chapter 5: Case Study – Impact of Passenger Group Dynamics on Airport Evacuation Process
Chapter 6: Conclusions
6.1 INTRODUCTION
This chapter summarises the main areas covered by this research, discusses the
major findings and how they contribute to new knowledge of pedestrian modelling.
In this chapter, the research questions raised at the beginning of this thesis will be
answered. Limitations of this research will be considered and future research
direction will be recommended.
6.2 THESIS SUMMARY
Nowadays, airports are becoming more and more passenger focused and
passenger experience in airports is highly valued. Under this context, this research is
motivated to investigate passenger behaviour in airport terminals using a passenger
flow model. The literature review in Chapter 2 revealed that grouping is a common
phenomenon among pedestrians, while most research failed to consider the group
dynamics when developing pedestrian models. In order to reflect the actual
passenger flow condition, the group dynamics must be included in the model.
Chapter 3 demonstrates the detailed procedure of building a passenger flow
model using an agent-based method. The model defined pedestrian characteristics
such as age, gender and travel purposes. Pedestrian group attributes were
implemented in their characteristics as well.
In Chapter 4, the model was used to analyse the impact of group dynamics on
airport passenger activities. Simulation results were compared under three different
scenarios: passenger travelling (1) alone; (2) in groups; and (3) in groups with
wavers. The results from both visual and statistical analysis showed that the group
dynamics can significantly influence passenger behaviour in terms of dwell time at
processing facilities and discretionary activity choices. Therefore, the group
dynamics can influence the passenger experience and level of service in the airport.
Publications arising from this work can be found in (Cheng, Reddy, Fookes, &
Yarlagadda, 2014a), (Cheng, Reddy, Fookes, & Yarlagadda, in press) and (Cheng,
Reddy, Fookes, & Yarlagadda, submitted).
Chapter 6: Conclusions 93
In Chapter 5, a case study was conducted to demonstrate how the agent-based
passenger flow model can be used by airport operators to examine the efficiency of
an evacuation strategy. By comparing evacuation time of individual passengers and
passengers in groups, the impact of group dynamics during an airport evacuation
process was analysed. The simulation results shows that group dynamics can
significantly impact passenger behaviour during airport evacuation process and the
total evacuation time. A publication arising from this work can be found in (Cheng,
Reddy, Fookes, & Yarlagadda, 2014b).
6.3 RESEARCH CONTRIBUTION
The primary objective of this research was to investigate the influence of group
dynamics on passenger behaviour in airports. By simulating pedestrian flow in an
international airport terminal using an agent-based model, the influence of group
dynamics on pedestrian behaviour has been found and analysed. The main
contributions of this thesis are embodied in the following four aspects:
1. Improve the understanding of group dynamics among pedestrians.
The literature review in this thesis had suggested that crowds consist of both
individual pedestrians and people in groups. The percentage of people in groups
within a crowd ranges from 40% to 70% at different occasions. Researchers also
discovered that the size of a pedestrian group can influence pedestrian walking
dynamics such as speed, group formation, and avoidance behaviour. As a result, one
should pay attention to group dynamics when developing pedestrian flow models.
2. Provide a more realistic agent-based passenger flow model by
incorporating group dynamics.
An agent-based modelling approach was applied in developing the pedestrian
model in this thesis because of its advantages in modelling pedestrian behaviour in
complex environments. As the simulation environment becomes more and more
complex, research had suggested a trend to combine multiple modelling approaches
into one simulation. This statement can be supported by the proposed model in which
heterogeneous agents are created according to the agent-based mindset, while the
pedestrian movement is governed by the customised social force algorithm.
94 Chapter 6: Conclusions
The most significant feature of the proposed model is that it considered
pedestrian group dynamics in a complex environment (an international airport
departure terminal in this case). Pedestrians in the model have their own attributes
which allow them to interact within groups, with pedestrians from other groups and
the airport environment. The model is able to collect statistics and present real-time
3D visualisations during the simulation. This advanced feature enables the statistical
validation and face validation of the model, and provides data for analysing
pedestrian group dynamics.
3. Demonstrate the influence of group dynamics on passenger flow in an
airport departure terminal.
By incorporating group dynamics into the agent-based model, simulation
results of different scenario settings had shown that the group dynamics have
significant influence on passenger activities in the airport terminal in terms of dwell
times, waiting behaviour and choices of discretionary activities.
In airport mandatory processing activities such as check-in, security control
and customs, passengers with group dynamics had a larger dwell time in the
processes. The waiting behaviour of passengers travelling in groups can lead to
higher pedestrian density in the waiting areas around the processing areas, which can
cause congestions and a lower level of service (LOS). Moreover, longer dwell time
in processing activities means less time for passengers to undertake discretionary
activities, which is not favourable for airport retail operators. Passenger group
dynamics can also influence the choices of retail activities. It was found that
passengers travelling in groups are more likely to undertake retail activities than
those who are travelling alone. The existence of wavers contributes a higher landside
retail opportunity.
4. Introduce the potential application of the agent-based pedestrian flow
model in design and management of pedestrian facilities.
From the airport design and management perspective, the model can be used to
test the impact of design changes, allocate optimal queuing area based on passenger
traffic, measure peak capacity and analyse the effectiveness of airport processing and
discretionary facilities. Running such a simulation provides flexibility and insights
into the design process at no extra cost. An application of the model in analysing
Chapter 6: Conclusions 95
potential effects of airport self-service technology can be found in Kirk, et al.
(submitted).
Furthermore, if terminal operators could run the simulation beforehand by
simply importing flight schedules and passenger quantity, then simulation results can
provide valuable information for them to respond proactively to potential
congestions. The density map in Appendix C shows the pedestrian density in the
airport terminal after a whole day simulation. According to the density map, airport
operators can easily predict critical congestion areas, and therefore take preventive
measures.
Hence, we envisage that such a simulation tool can be critical to airport
designers and operators. An airport with minimal congestion will in turn assist in
providing a positive experience for passengers using the airport. However, the
potential application of the model is not limited to the airport terminals only. It can
be extended to simulate pedestrian behaviour in all public pedestrian environments
such as railway stations, shopping centres, and theme parks.
6.4 RESEARCH LIMITATIONS
The limitation of this thesis lies in the lack of access to detailed airport data,
which leads to difficulties in model development, validation and calibration. For
agent-based models, the accuracy of decision making processes and human activities
determine the reliability of the model. Passenger attributes such as gender, age and
culture background all have influence on the passenger’s behaviour, retail choice and
environment cognition. However, due to the lack of access to such empirical data,
the model was considerably simplified by using data obtained from previous research
papers or empirical assumptions while simulating the human behaviour.
6.5 RECOMMANDTIONS
As was mentioned in the previous section, a limitation of this research was the
access to airport passenger characteristic data. Therefore, it is recommended that
future research be undertaken in the following areas:
1. Field data collection of pedestrian characteristics
Since the environment can influence pedestrian behaviour and group dynamics
such as the proportion of people in groups, the group size and group speed, it is
96 Chapter 6: Conclusions
necessary to collect further field data from the reference environment in order to
build a realistic model. To build a pedestrian model for an international airport
terminal, one needs to consider field data such as basic pedestrian status (e.g. arrival
time distribution, passenger group size, waver status) and personal information (such
as gender, age, nationality, destination, travel purposes, etc).
2. Explore passenger discretionary activities and retail behaviour in
airports
A large proportion of airport revenue is generated in the airport retail segment.
High-quality airport retail service can contribute to higher passenger satisfaction and
enhanced airport experience, which in turn attracts more passengers and benefits the
airport. Therefore, future study investigating airport passenger retail activity would
be very interesting. The agent-based model can be used to simulate the retail and
purchasing behaviour of passengers with different characteristics.
3. Investigate the application of agent-based pedestrian flow model
Future research should also concentrate on the application of agent-based
pedestrian flow model. The model can be used as a tool for airport operators to
propose regulations, test the level of service (LOS) in each processing unit and sub-
system, and design new airports.
Chapter 6: Conclusions 97
98 Chapter 6: Conclusions
Bibliography
AAA. (2012, May). The economic and social contribution of Australia’s airports.
Retrieved from
ABS. (2013). Overseas Arrivals and Departures. Retrieved from
ACCC. (2013). Airport Monitoring Report 2011-2012. Retrieved from
http://transition.accc.gov.au/content/item.phtml?itemId=1110935&nodeId=2b
65e50d1e38784c95922f94ebc5b389&fn=Airport%20Monitoring%20Report %202011-12.pdf
Ahyudanari, E. (2003). Methodology to Determine Airport Check-in Counter Arrangements
Master. The University of New South Wales, Sydney, Australia.
Ahyudanari, E., & Vandebona, U. (2005). Simplified model for estimation of airport check-in
facilities. Journal of the Eastern Asia Society for Transportation Studies, 6, 724-735.
Ando, K., Ota, H., & Oki, T. (1988). Forecasting the flow of people. Railway
Research Review, 45(8), 8-14.
AnyLogic. (2013). Retrieved 2013from http://www.anylogic.com/
Asano, M., Iryo, T., & Kuwahara, M. (2010). Microscopic pedestrian simulation model
combined with a tactical model for route choice behaviour. Transportation Research Part C:
Emerging Technologies, 18(6), 842-855. Retrieved from
http://www.sciencedirect.com/science/article/pii/S0968090X10000197.
doi:http://dx.doi.org/10.1016/j.trc.2010.01.005
Ashford, N. J., Mumayiz, S., & Wright, P. H. (2011a). Airport Engineering : Planning, Design
and Development of 21st Century Airports. Retrieved from
http://UQL.eblib.com.au/patron/FullRecord.aspx?p=698766
Ashford, N. J., Mumayiz, S., & Wright, P. H. (2011b). Airport Engineering:
Planning, Design and Development of 21st Century Airports: Wiley.
Aveni, A. F. (1977). The Not-So-Lonely Crowd: Friendship Groups in Collective Behavior.
Sociometry, 40(1), 96-99. Retrieved from http://www.jstor.org/stable/3033551.
Bibliography 99
Bandini, S., Federici, M., Manzoni, S., & Vizzari, G. (2006). Towards a Methodology for
Situated Cellular Agent Based Crowd Simulations. In O. Dikenelli, M.-P. Gleizes & A. Ricci
(Eds.), Engineering Societies in the Agents World VI (Vol. 3963, pp. 203-220): Springer Berlin
Heidelberg.
Bauer, D., Seer, S., & Brändle, N. (2007). Macroscopic pedestrian flow simulation for designing
crowd control measures in public transport after special events. In Proceedings of the 2007
summer computer simulation conference (pp. 1035-1042): Society for Computer Simulation
International.
Blue, V. J., Embrechts, M. J., & Adler, J. L. (1997, 12-15 Oct 1997). Cellular automata modeling
of pedestrian movements. In Systems, Man, and Cybernetics, 1997. Computational Cybernetics
and Simulation., 1997 IEEE International Conference on (Vol. 3, pp. 2320-2323 vol.2323).
Bonabeau, E. (2002). Agent-based modeling: Methods and techniques for simulating human
systems. Proceedings of the National Academy of Sciences of the United States of America,
99(Suppl 3), 7280-7287. Retrieved from http://www.pnas.org/content/99/suppl.3/7280.short.
doi:10.1073/pnas.082080899
Camillen, F., Capri, S., Garofalo, C., Ignaccolo, M., Inturri, G., Pluchino, A., . . . Tudisco, S.
(2009, 26-27 Sept. 2009). Multi agent simulation of pedestrian behavior in closed spatial
environments. In Science and Technology for Humanity (TIC-STH), 2009 IEEE Toronto
International Conference (pp. 375- 380).
Castillo-Manzano, J. I. (2010). Determinants of commercial revenues at airports: Lessons
learned from Spanish regional airports. Tourism Management, 31(6), 788-796. Retrieved from
http://www.sciencedirect.com/science/article/pii/S0261517709001599.
doi:http://dx.doi.org/10.1016/j.tourman.2009.08.005
Castle, C. J. E., Waterson, N. P., Pellissier, E., & Bail, S. (2011). A Comparison of Grid-based
and Continuous Space Pedestrian Modelling Software: Analysis of Two UK Train Stations. In R.
D. Peacock, E. D. Kuligowski & J. D. Averill (Eds.), Pedestrian and Evacuation Dynamics (pp.
433-446): Springer US.
Ceder, A. (1979). An algorithm to assign pedestrian groups dispersing at public gatherings based
on pedestrian-traffic modelling. Applied Mathematical Modelling, 3(2), 116-124. Retrieved from
http://www.sciencedirect.com/science/article/pii/0307904X79900386.
doi:http://dx.doi.org/10.1016/0307-904X(79)90038-6
100 Bibliography
Cheng, L., Reddy, V., Fookes, C., & Yarlagadda, P. (submitted). Impact of Passenger Group
Dynamics on Passenger Flows in Airports Using an Agent- Based Model.
Cheng, L., Reddy, V., Fookes, C., & Yarlagadda, P. K. D. V. (2014a). Agent-based modelling
simulation case study : assessment of airport check-in and evacuation process by considering
group travel behaviour of air passengers. Paper presented at Measurement Technology and its
Application III, Shanghai, China. Retrieved from http://eprints.qut.edu.au/72311/
Cheng, L., Reddy, V., Fookes, C., & Yarlagadda, P. K. D. V. (2014b). Impact of passenger
group dynamics on an airport evacuation process using an agent- based model. Paper presented
at 2014 International Conference on Computational Science and Computational Intelligence, Las
Vegas, Nevada, USA. Retrieved from http://eprints.qut.edu.au/69071/
Cheng, L., Reddy, V., Fookes, C., & Yarlagadda, P. K. D. V. (in press). ANALYSIS OF
PASSENGER GROUP BEHAVIOUR AND ITS IMPACT ON PASSENGER FLOW USING
AN AGENT-BASED MODEL. Paper presented at 4th International Conference on Simulation
and Modeling Methodologies, Technologies and Applications (SIMULTECH), Vienna, Austria.
Coleman, J. S., & James, J. (1961). The Equilibrium Size Distribution of Freely- Forming
Groups. Sociometry, 24(1), 36-45. Retrieved from http://www.jstor.org/stable/2785927.
Correia, A. R., Wirasinghe, S. C., & de Barros, A. G. (2008). A global index for level of service
evaluation at airport passenger terminals. Transportation Research Part E: Logistics and
Transportation Review, 44(4), 607-620. Retrieved from
http://www.sciencedirect.com/science/article/pii/S1366554507000543.
doi:10.1016/j.tre.2007.05.009
Crawford, G., & Melewar, T. C. (2003). The importance of impulse purchasing behaviour in the
international airport environment. Journal of Consumer Behaviour, 3(1), 85-98.
Daamen, W., & Hoogendoorn, S. (2004). Pedestrian traffic flow operations on a platform:
observations and comparison with simulation tool SimPed. In Computers in Railways IX
(Congress Proceedings of CompRail 2004, Southampton: WIT Press, Dresden, Germany (pp.
125-134).
Daamen, W., & Hoogendoorn, S. P. (2003). Research on pedestrian traffic flows in
the Netherlands. Proceedings Walk, 21, 101-117.
Bibliography 101
Farmer, J. D., & Foley, D. (2009). The economy needs agent-based modelling. [Editorial
Material]. Nature, 460(7256), 685-686. Retrieved from <Go to ISI>://WOS:000268670300020.
doi:10.1038/460685a
Filippidis, L., Galea, E. R., Gwynne, S., & Lawrence, P. J. (2006). Representing the Influence of
Signage on Evacuation Behavior within an Evacuation Model. Journal of Fire Protection
Engineering, 16(1), 37-73. Retrieved from http://jfe.sagepub.com/content/16/1/37.abstract.
doi:10.1177/1042391506054298
Finnis, K. K., & Walton, D. (2008). Field observations to determine the influence of population
size, location and individual factors on pedestrian walking speeds. Ergonomics, 51(6), 827-842.
Retrieved from http://dx.doi.org/10.1080/00140130701812147.
doi:10.1080/00140130701812147
Freathy, P., & O'Connell, F. (2012). Spending time, spending money: passenger segmentation in
an international airport. The International Review of Retail, Distribution and Consumer
Research, 22(4), 397-416. Retrieved from http://dx.doi.org/10.1080/09593969.2012.690778.
doi:10.1080/09593969.2012.690778
Fukui, M., & Ishibashi, Y. (1999). Self-Organized Phase Transitions in Cellular Automaton
Models for Pedestrians. Journal of the Physical Society of Japan, 68(8), 2861-2863. Retrieved
from http://ci.nii.ac.jp/naid/110001955643/en/. doi:10.1143/JPSJ.68.2861
Gatersleben, M. R., & Van der Weij, S. W. (1999). Analysis and simulation of passenger flows
in an airport terminal. In Simulation Conference Proceedings, 1999 Winter (Vol. 2, pp.
1226-1231): IEEE.
Gwynne, S., Galea, E., Owen, M., Lawrence, P., & Filippidis, L. (1999a). A review of the
methodologies used in evacuation modelling. Fire and Materials, 23(6), 383-388.
Gwynne, S., Galea, E. R., Owen, M., Lawrence, P. J., & Filippidis, L. (1999b). A review of the
methodologies used in the computer simulation of evacuation from the built environment.
Building and Environment, 34(6), 741-749. Retrieved from
http://www.sciencedirect.com/science/article/pii/S0360132398000572.
doi:http://dx.doi.org/10.1016/S0360-1323(98)00057-2
Harney, D. (2002). Pedestrian Modelling: Current Methods and Future Directions.
Road and transport research, 11(4), 38-48. Retrieved from http://dx.doi.org/.
102 Bibliography
Harrison, A., Popovic, V., Kraal, B. J., & Kleinschmidt, T. (2012). Challenges in passenger
terminal design : a conceptual model of passenger experience. Paper presented at DRS 2012
Bangkok : Reforming Traditions, Reshaping Boundaries. Retrieved from
http://eprints.qut.edu.au/53779/
Helbing, D., Buzna, L., Johansson, A., & Werner, T. (2005). Self-Organized Pedestrian Crowd
Dynamics: Experiments, Simulations, and Design Solutions. Transportation Science, 39(1), 1-24.
Retrieved from http://transci.journal.informs.org/content/39/1/1.abstract.
doi:10.1287/trsc.1040.0108
Helbing, D., Farkas, I., Molnàr, P., & Vicsek, T. (2002). Simulation of pedestrian crowds in
normal and evacuation situations. In M. Schreckenberg & S. D. Sharma (Eds.), Pedestrian and
Evacuation Dynamics (pp. 21-58): Springer.
Helbing, D., Farkas, I., & Vicsek, T. (2000). Simulating dynamical features of escape panic.
[10.1038/35035023]. Nature, 407(6803), 487-490. Retrieved from
http://dx.doi.org/10.1038/35035023.
Helbing, D., & Molnár, P. (1995). Social force model for pedestrian dynamics. Physical Review
E, 51(5), 4282-4286. Retrieved from http://link.aps.org/doi/10.1103/PhysRevE.51.4282.
Henderson, L. (1971). The statistics of crowd fluids. Nature, 229, 381-383.
Henein, C., & White, T. (2005). Agent-Based Modelling of Forces in Crowds. In P. Davidsson,
B. Logan & K. Takadama (Eds.), Multi-Agent and Multi-Agent- Based Simulation (Vol. 3415,
pp. 173-184): Springer Berlin Heidelberg.
Hoogendoorn, S., & Daamen, W. (2004). Design assessment of Lisbon transfer stations using
microscopic pedestrian simulation. In Computers in railways IX (Congress Proceedings of
CompRail 2004) (pp. 135-147).
IATA. (2004). Airport Development Reference Manual. Montreal, Canada:
International Air Transport Association.
IATA. (2013a, June). Annual Report 2013. Retrieved from
IATA. (2013b). IATA Global Passenger Survey. Retrieved from
James, J. (1953). The distribution of free-forming small group size. American
Sociological Review, 18, 569-570. doi:10.2307/2087444
Bibliography 103
Johansson, A., Batty, M., Hayashi, K., Al Bar, O., Marcozzi, D., & Memish, Z. A. (2012).
Crowd and environmental management during mass gatherings. The Lancet infectious diseases,
12(2), 150-156.
Karamouzas, I., & Overmars, M. (2010). Simulating the local behaviour of small pedestrian
groups. Paper presented at Proceedings of the 17th ACM Symposium on Virtual Reality
Software and Technology, Hong Kong.
Kirk, P. J. (2013). Passenger experience at airports : An activity-centred approach
(PhD). Retrieved from http://eprints.qut.edu.au/60803/
Kirk, P. J., Cheng, L., Popovic, V., Kraal, B., & Fookes, C. (submitted). Improving airport
passenger experience: Using agent-based modelling to understand potential effects of
self-service technology.
Kitazawa, K., & Batty, M. (2004). Pedestrian Behaviour Modelling An Application
to Retail Movements using a Genetic Algorithm.
Klügl, F. (2008). A validation methodology for agent-based simulations. In Proceedings of the
2008 ACM symposium on Applied computing (pp. 39-43): ACM.
Klügl, F., & Rindsfüser, G. (2007). Large-scale agent-based pedestrian simulation:
Springer.
Klüpfel, H. (2007). The simulation of crowd dynamics at very large events — Calibration,
empirical data, and validation. In N. Waldau, P. Gattermann, H. Knoflacher & M. Schreckenberg
(Eds.), Pedestrian and Evacuation Dynamics 2005 (pp. 285-296): Springer Berlin Heidelberg.
Kormanová, A. (2012). Combining Social forces and Cellular automata models in
pedestrians’ movement simulation (Vol. 10).
Köster, G., Seitz, M., Treml, F., Hartmann, D., & Klein, W. (2011). On modelling the influence
of group formations in a crowd. Contemporary Social Science, 6(3), 397-414. Retrieved from
http://dx.doi.org/10.1080/21582041.2011.619867. doi:10.1080/21582041.2011.619867
Kraal, B. J., Popovic, V., & Kirk, P. J. (2009). Passengers in the airport : artefacts and activities.
Paper presented at Design : Open 24/7. Retrieved from http://eprints.qut.edu.au/31029/
Lewin, K., & Cartwright, D. (1952). Field theory in social science: Selected
theoretical papers: Tavistock London.
104 Bibliography
Livingstone, A., Popovic, V., Kraal, B. J., & Kirk, P. J. (2012). Understanding the airport
passenger landside retail experience. Paper presented at DRS 2012 Bangkok – Research:
Uncertainty, Contradiction and Value, Chulalongkorn University, Bangkok. Retrieved from
http://eprints.qut.edu.au/54334/
London First. (2008). Imagine a world class Heathrow. Retrieved 12 October 2009, from
http://www.london-
first.co.uk/documents/Imagine_a_world_class_Heathrow_SUMMARY_web_ final.pdf
Ma, W. (2013). Agent-based model of passenger flows in airport terminals (PhD).
Retrieved from http://eprints.qut.edu.au/63457/
Ma, W., Fookes, C., Kleinschmidt, T., & Yarlagadda, P. (2012). Modelling passenger flow at
airport terminals : individual agent decision model for stochastic passenger behaviour. Paper
presented at SIMULTECH 2012, Rome. Retrieved from http://eprints.qut.edu.au/52640/
Ma, W., Kleinschmidt, T., Fookes, C., & Yarlagadda, P. K. D. V. (2011). Check-in processing :
simulation of passengers with advanced traits. Paper presented at Proceedings of the 2011 Winter
Simulation Conference. Retrieved from http://eprints.qut.edu.au/42377/
Ma, W., & Yarlagadda, P. K. (2012). A micro-simulation of airport passengers with advanced
traits. Paper presented at 28th Congress of the International Council of the Aeronautical
Sciences, Brisbane, QLD. Retrieved from http://eprints.qut.edu.au/54425/
Ma, W., Yarlagadda, P. K., & Fookes, C. (2012). Using advanced traits of passengers to
facilitate route-choice decision-making. Paper presented at Proceedings of the 4th International
Conference on Computational Methods (ICCM2012), Crowne Plaza, Gold Coast, Qld. Retrieved
from http://eprints.qut.edu.au/54427/
Macal, C. M., & North, M. J. (2005, 4-7 Dec. 2005). Tutorial on agent-based modeling and
simulation. In Simulation Conference, 2005 Proceedings of the Winter (pp. 14 pp.).
Macal, C. M., & North, M. J. (2006, 3-6 Dec. 2006). Tutorial on Agent-Based Modeling and
Simulation PART 2: How to Model with Agents. In Simulation Conference, 2006. WSC 06.
Proceedings of the Winter (pp. 73-83).
Macal, C. M., & North, M. J. (2010). Tutorial on agent-based modelling and simulation. Journal
of Simulation, 4(3), 151-162. Retrieved from ProQuest Central.Retrieved from
Bibliography 105
http://gateway.library.qut.edu.au/login?url=http://search.proquest.com/docvie
w/749101997?accountid=13380. doi:10.1057/jos.2010.3
Macal, C. M., & North, M. J. (2011, 11-14 Dec. 2011). Introductory tutorial: Agent- based
modeling and simulation. In Simulation Conference (WSC), Proceedings of the 2011 Winter (pp.
1451-1464).
Manenti, L., Manzoni, S., Vizzari, G., Ohtsuka, K., & Shimura, K. (2012). An Agent-Based
Proxemic Model for Pedestrian and Group Dynamics: Motivations and First Experiments
Multi-Agent-Based Simulation XII. In D. Villatoro, J. Sabater-Mir & J. Sichman
(Eds.), (Vol. 7124, pp. 74-89): Springer Berlin / Heidelberg.
MASON. (2013). Retrieved 08/12, 2013from
http://cs.gmu.edu/~eclab/projects/mason/
Mehran, R., Oyama, A., & Shah, M. (2009, 20-25 June 2009). Abnormal crowd behavior
detection using social force model. In Computer Vision and Pattern Recognition, 2009. CVPR
2009. IEEE Conference on (pp. 935-942).
Moore, R. L. (1953). Pedestrian Choice and Judgment. OR, 4(1), 3-10. Retrieved
from http://www.jstor.org/stable/3006905. doi:10.2307/3006905
Moussaïd, M., Perozo, N., Garnier, S., Helbing, D., & Theraulaz, G. (2010). The Walking
Behaviour of Pedestrian Social Groups and Its Impact on Crowd Dynamics.
[doi:10.1371/journal.pone.0010047]. PLoS ONE, 5(4), e10047. Retrieved from
http://dx.doi.org/10.1371%2Fjournal.pone.0010047.
NetLogo. (2013). Retrieved 2013from http://ccl.northwestern.edu/netlogo
Osaragi, T. (2004). Modeling of Pedestrian Behavior and Its Applications to Spatial Evaluation.
Paper presented at Proceedings of the Third International Joint Conference on Autonomous
Agents and Multiagent Systems - Volume 2, New York, New York.
Pan, X., Han, C., Dauber, K., & Law, K. (2007). A multi-agent based framework for the
simulation of human and social behaviors during emergency evacuations. AI & SOCIETY,
22(2), 113-132. Retrieved from http://dx.doi.org/10.1007/s00146-007-0126-1.
doi:10.1007/s00146-007-0126- 1
Parisi, D. R., Gilman, M., & Moldovan, H. (2009). A modification of the Social Force Model can
reproduce experimental data of pedestrian flows in normal conditions. Physica A: Statistical
Mechanics and its Applications, 388(17), 3600-3608. Retrieved from
106 Bibliography
http://www.sciencedirect.com/science/article/pii/S0378437109004075.
doi:http://dx.doi.org/10.1016/j.physa.2009.05.027
Pelechano, N., Allbeck, J. M., & Badler, N. I. (2007). Controlling individual agents in
high-density crowd simulation. Paper presented at Proceedings of the 2007 ACM
SIGGRAPH/Eurographics symposium on Computer animation, San Diego, California.
Perng, S.-W., Chow, C.-C., & Liao, W.-C. (2010). Analysis of shopping preference and
satisfaction with airport retailing products. Journal of Air Transport Management, 16(5),
279-283. Retrieved from http://www.sciencedirect.com/science/article/pii/S0969699710000256.
doi:http://dx.doi.org/10.1016/j.jairtraman.2010.02.002
Pluchino, A., Garofalo, C., Inturri, G., Rapisarda, A., & Ignaccolo, M. (2013). Agent-based
simulation of pedestrian behaviour in closed spaces: a museum case study. arXiv preprint
arXiv:1302.7153.
Popovic, V., Kraal, B., & Kirk, P. J. (2010). Towards airport passenger experience models. Paper
presented at Proceedings of 7th International Conference on Design & Emotion, Spertus
Institute, Chicago, Illinois. Retrieved from http://eprints.qut.edu.au/38095/
Popovic, V., Kraal, B. J., & Kirk, P. J. (2009). Passenger experience in an airport : an
activity-centred approach. Paper presented at IASDR 2009 Proceedings, COEX, Seoul.
Retrieved from http://eprints.qut.edu.au/30226/
Qiu, F., & Hu, X. (2010). Modeling group structures in pedestrian crowd simulation. Simulation
Modelling Practice and Theory, 18(2), 190-205. Retrieved from
http://www.sciencedirect.com/science/article/pii/S1569190X09001555.
doi:10.1016/j.simpat.2009.10.005
Rastogi, R., Thaniarasu, I., & Chandra, S. (2011). Design Implications of Walking Speed for
Pedestrian Facilities. Journal of Transportation Engineering, 137(10), 687-696. Retrieved from
http://ascelibrary.org/doi/abs/10.1061/%28ASCE%29TE.1943- 5436.0000251.
doi:doi:10.1061/(ASCE)TE.1943-5436.0000251
Repast. (2013). Retrieved 08/12, 2013from http://repast.sourceforge.net/
Reynolds, C. (1999). Steering Behaviors for Autonomous Characters. In Game
Developers Conference 1999 (pp. 763-782): Miller Freeman Game Group.
Santos, G., & Aguirre, B. E. (2004). A critical review of emergency evacuation
simulation models.
Bibliography 107
Schadschneider, A. (2001). Cellular automaton approach to pedestrian dynamics-
theory. arXiv preprint cond-mat/0112117.
Schadschneider, A., Klingsch, W., Klüpfel, H., Kretz, T., Rogsch, C., & Seyfried, A. (2011).
Evacuation Dynamics: Empirical Results, Modeling and Applications. In R. A. Meyers (Ed.),
Extreme Environmental Events (pp. 517- 550): Springer New York.
Schelhorn, T., O'Sullivan, D., Haklay, M., & Thurstain-Goodwin, M. (1999).
STREETS: An agent-based pedestrian model.
Schultz, M., Lehmann, S., & Fricke, H. (2006). Pedestrian Dynamics in Airport Terminals
Considering Emergency Cases. Proceedings of Internation Council of the Aeronautical Sciences.
Schultz, M., Lehmann, S., & Fricke, H. (2007). A discrete microscopic model for pedestrian
dynamics to manage emergency situations in airport terminals. In N. Waldau, P. Gattermann, H.
Knoflacher & M. Schreckenberg (Eds.), Pedestrian and Evacuation Dynamics 2005 (pp.
369-375): Springer Berlin Heidelberg.
Schultz, M., Schulz, C., & Fricke, H. (2010). Passenger Dynamics at Airport Terminal
Environment. Pedestrian and Evacuation Dynamics 2008, 381-396. Retrieved from
http://dx.doi.org/10.1007/978-3-642-04504-2_33. doi:10.1007/978-3-642-04504-2_33
Singh, H., Arter, R., Dodd, L., Langston, P., Lester, E., & Drury, J. (2009). Modelling subgroup
behaviour in crowd dynamics DEM simulation. Applied Mathematical Modelling, 33(12),
4408-4423. Retrieved from
http://www.sciencedirect.com/science/article/pii/S0307904X09000808.
doi:10.1016/j.apm.2009.03.020
Song, W., Yu, Y., Wang, B., & Fan, w. (2006). Evacuation behaviors at exit in CA model with
force essentials: A comparison with social force model. Physica A: Statistical Mechanics and its
Applications, 371(2), 658-666. Retrieved from
http://www.sciencedirect.com/science/article/pii/S0378437106003633.
doi:http://dx.doi.org/10.1016/j.physa.2006.03.027
Stefanik, M., Kandera, B., & Badanik, B. (2012). Aspects influencing arrival behavioural pattern
of air travellers. Archives of Transport, 5(1), 35-40.
Swarm. (2013). Retrieved 8/12, 2013from http://www.swarm.org/
Takakuwa, S., & Oyama, T. (2003, 7-10 Dec. 2003). Simulation analysis of
international-departure passenger flows in an airport terminal. In Simulation
108 Bibliography
Conference, 2003. Proceedings of the 2003 Winter (Vol. 2, pp. 1627-1634 vol.1622).
Tarawneh, M. S. (2001). Evaluation of pedestrian speed in Jordan with investigation of some
contributing factors. Journal of Safety Research, 32(2), 229-236. Retrieved from
http://www.sciencedirect.com/science/article/pii/S0022437501000469.
doi:http://dx.doi.org/10.1016/S0022-4375(01)00046-9
Teknomo, K. (2006). Application of microscopic pedestrian simulation model. Transportation
Research Part F: Traffic Psychology and Behaviour, 9(1), 15- 27. Retrieved from
http://www.sciencedirect.com/science/article/pii/S1369847805000689.
doi:http://dx.doi.org/10.1016/j.trf.2005.08.006
Teknomo, K., Takeyama, Y., & Inamura, H. (2000). Review on microscopic pedestrian
simulation model. In Proceedings Japan Society of Civil Engineering Conference: Citeseer.
Tosic, V. (1992). A REVIEW OF AIRPORT PASSENGER TERMINAL OPERATIONS
ANALYSIS AND MODELING. Transportation Research Part a-Policy and Practice, 26A(1),
3-26. Retrieved from <Go to ISI>://WOS:A1992HK16400002.
doi:10.1016/0965-8564(92)90041-5
Vizzari, G., Manenti, L., Ohtsuka, K., & Shimura, K. (2012). An agent-based approach to
pedestrian and group dynamics: experimental and real world scenarios. In Proceedings of the 7th
International Workshop on Agents in Traffic and Transportation.
Von Neumann, J., & Burks, A. W. (1966). Theory of self-reproducing automata.
Weidmann, U. (1992). Transporttechnik der fußgänger.
Wolfram, S. (1983). Statistical mechanics of cellular automata. Reviews of modern
physics, 55(3), 601.
Wu, P. P., & Mengersen, K. (2013). A review of models and model usage scenarios for an
airport complex system. Transportation Research Part A: Policy and Practice, 47, 124-140.
Retrieved from http://eprints.qut.edu.au/48262/. doi:10.1016/j.tra.2012.10.015
Xing, P., Lees, M., Nan, H., & Viswanthatn, T. V. (2012). Validation of agent-based simulation
through human computation: an example of crowd simulation. In Multi-Agent-Based Simulation
XII (pp. 90-102): Springer.
Bibliography 109
Yuan, W., & Tan, K. H. (2007). An evacuation model using cellular automata. Physica A:
Statistical Mechanics and its Applications, 384(2), 549-566. Retrieved from
http://www.sciencedirect.com/science/article/pii/S0378437107006085.
doi:10.1016/j.physa.2007.05.055
Zheng, X., Zhong, T., & Liu, M. (2009). Modeling crowd evacuation of a building based on
seven methodological approaches. Building and Environment, 44(3), 437-445. Retrieved from
http://www.sciencedirect.com/science/article/pii/S0360132308000577.
doi:10.1016/j.buildenv.2008.04.002
Zhou, S., Chen, D., Cai, W., Luo, L., Low, M. Y. H., Tian, F., . . . Hamilton, B. D. (2010).
Crowd modeling and simulation technologies. ACM Transactions on Modeling and Computer
Simulation (TOMACS), 20(4), 20.
110 Bibliography
Appendices
A. FLIGHT SCHEDULE
Flight schedule used in the model:
Flight Departure
Time
Passenger Quantity
Departure Gate
Entrance Number
Check-In Counter Number EK433 3:30 222 81 1 1 QF123 6:20 134 82 2 2 EK434 7:25 222 83 3 3
DJ080/EY6117 8:20 134 80 4 4 DJ186/EY6189 8:20 134 77 5 5 DJ066/EY6194 8:30 134 79 6 6
DJ175/VA175 9:00 134 85 7 7 HT531 9:00 128 78 8 8 SQ256/VA5668 9:10 241 75 9 9 DJ4191 9:15 134
86 10 10 IE701 9:30 120 84 11 1 QQ371 9:30 51 79 12 2 CX146/AY5832 9:35 241 83 13 3 DJ4197
10:30 134 76 14 4 DJ181 10:30 134 86 15 5 PX004/QF349 10:40 180 82 16 6 VA007/DL6795 11:20 222
84 17 7 NZ136/CA5104 11:15 251 80 18 8 QF015/AA7380 12:30 258 75 19 9 QF051/9W4051 13:40
258 78 20 10 TG474/LH9759 14:20 251 85 21 1 SQ236/LH9773 14:40 265 77 22 2 NZ734/CA5139
16:45 134 82 23 3 PX026/QF379 17:40 99 80 24 4 QF125/AA7303 17:40 134 81 25 5 DJ068/EY6195
17:45 134 86 26 6 DJ082/EY6190 18:00 134 76 27 7 DJ188/EY6188 18:30 134 75 28 8 HA444 18:35
180 85 29 9 EK435 20:45 222 84 30 10 SB153 21:20 134 76 31 1 FJ922/QF347 22:40 134 83 32 2
BR316/QF329 22:45 243 85 33 3 SQ246/LH9783 23:45 265 86 34 4 MH134/KL4101 23:50 265 77 35 5
Table A-1 Flight timetable in the model.
Appendices 111
B. IATA LEVEL OF SERVICE (LOS) FRAMEWORK
Description LOS Flow Delay Comfort A. Excellent Free
None Excellent B. High Stable Very Few High C. Good Stable Acceptable Good D. Adequate Unstable
Acceptable for short time Adequate E. Inadequate Unstable Unacceptable Inadequate F. Unacceptable
Total system breakdown Unacceptable
Table B-2 IATA LOS Framework (IATA, 2004).
LOS standards (square meters per occupants) Sub-system A ..B.. ..C.. ..D.. E.. ..F.. Check-in queue area
1.8 1.6 1.4 1.2 1 Total Wait/circulate 2.7 2.3 1.9 1.5 1
System Hold room 1.4 1.2 1 0.8
0.6
Breakdown Bag clain area 2 1.8
1.6 1.4 1.2 Government inspection 1.4 1.2 1 0.8 0.6
Table B-3 IATA LOS Congestion Standards (Ashford, Mumayiz, & Wright, 2011a).
112 Appendices
C. PEDESTRIAN DENSITY MAP OF THE AIRPORT DEPARTURE
TERMINAL
(a)
(b)
Figure C-1 Pedestrian density map of the airport departure terminal after whole day simulation. (a) level 4; (b) level
3.
Appendices 113