Sie sind auf Seite 1von 4


Degree of Importance of Airport

Passenger Terminal Components and
Their Attributes
This paper applies the Analytical Hierarchy Process (AHP) to determine the importance that users assign
to the various components of an airport passenger terminal (APT) and their attributes. A survey of depart-
ing passengers was employed to obtain the user perceptions at São Paulo/ Guarulhos International Airport,
the biggest airport in South America. The results indicate that the most important components of an APT
for departing passengers are the check-in counter and the departure lounge.

By Michelle Bandeira, Anderson Correia and Chan Wirasinghe

Airport managers have to struggle with
the decision of prioritizing resources.
Although they are motivated to offer a
reasonable level of service (LOS) to pas-
sengers, there is a growing worldwide
tendency for cost reduction. In this sce-
nario, an effort to determine the impor-
tance that passengers attribute to airport
components will be useful, as it will be
one indication of where airport managers
should invest their limited resources such
as funds, employees and their own atten-

This paper provides a methodology to

obtain the importance that passengers
assign to the various components and
attributes of an airport passenger terminal Photo 1: Denver’s Airport Passenger Terminal
(APT). The Analytical Hierarchy Process criticized (Müller and Gosling (1991); steps to develop such measures, using
(AHP) is employed in order to obtain Ashford (1988); Correia and Wirasinghe objective variables, i.e. variables that can
quantitative weights representing the rel- (2004)). One of the main issues is the be measured using a physical scale: time,
ative importance of components and their lack of passenger input in determining space, distance, etc. They used multiple
attributes (processing time, courtesy, etc). the standards. In addition, these studies least squares regression analysis to
We can suppose, through the process of do not indicate the degree of importance regress an overall LOS identified by pas-
pair-wise comparisons between the com- of each APT component or attribute to sengers (dependent variable) against
ponents and their attributes, that these the overall level of service. Thus, it is individual component LOS.
weights could be further used to obtain a impossible to obtain an overall APT level
global LOS measure of an APT as a of service for a given group of passengers This paper intends to analyze LOS as a
function of the LOS of individual com- (e.g. departures) using those manuals. function of objective and subjective
ponents (check-in, departure lounge, etc).
Correia and Wirasinghe
Literature Review (2004) provide a com-
Several air transport agencies have pro- prehensive literature
posed level of service (LOS) standards review on airport LOS
for airport passenger terminals, including measures. It was con-
the ACI – Airports Council International cluded that obtaining the
(ACI, 2000), IATA - International Air degree of importance of
Transport Association (IATA, 1995), the APT components
FAA – Federal Aviation Administration and their attributes was
(TRB, 1987) and Transport Canada critical. Correia (2005)
(1979). Despite their efforts, the pro- and Correia, Wirasinghe Figure 1: Average daily aircraft movement at São Paulo
posed LOS standards have been widely & de Barros (2006) took International Airport (Source: Infraero, 2000)

e-zine edition, Issue 37 1

Components Characteristics sengers were asked to indicate the
Parking Courtesy, Security, degree of importance of courtesy com-
Availability of pared to comfort at the departure
Parking Spots lounge. The scale provided to the user
will be explained in the next section.
Departure Hall Security, Orientation, Figure 2: n x n square matrix
Information, The number of factors to be analyzed similar family income, etc). It might
Comfort, Services should be kept into a minimum, in be the case that given groups of pas-
order to prevent passengers getting sengers have different perceptions of
Check-in Processing and
confused. Additionally, the question- the importance of components and
Waiting Time,
naire should be simple because pas- attributes.
sengers in the departure lounge usual-
Departure Courtesy, Comfort ly do not have too much time to The AHP – Analytical Hierarchy
Lounge answer it. A pilot survey was done to Process
Concessions Courtesy, Variety of test the questionnaire. One of the con- The Analytical Hierarchy Process was
Stores clusions was that too many questions first presented by Saaty (1980). It is
Table 1: Components and their Attri-butes led the passenger to answer the last one of the first multi-criteria decision
as proposed by an Expert Panel ones without much consideration. methods. Its main objective is to repre-
sent the decision model in the most
measures, such as courtesy and comfort. realistic way, including subjective and
The management of São Paulo/
The AHP method was selected for the objective factors. Using this method, it
Guarulhos International Airport did
analysis, since it is able to incorporate is possible to structure a problem
not allow the surveying of passengers
both objective and subjective measures. through hierarchical levels and make
about the importance of security
The weights found for each component alternative comparisons through a
screening and passport control. Thus,
and their respective attributes will indi- quantitative importance scale. Table 2
these components were not included in
cate the degree of importance that pas- presents the fundamental scale of the
the analysis.
sengers attribute to them. method.
Several socio-economic variables
Data Collection were collected during the survey in According to Table 2, the maximum
A passenger survey was taken at São degree of importance of a given com-
addition to the passenger ratings of the
Paulo/ Guarulhos International Airport in ponent or attribute is 0.90. That will
importance components and their
the period of November 9th-15th, 2006. happen when this component is
Over 100 randomly selected passengers extremely more important than anoth-
were interviewed at the departure lounge er component.
1.gender; age; family income;
during peak hours (08:00-10:00 and
2.airline; destination (national or inter-
17:00-19:00) as identified from the aver- The AHP method consists of an n x n
national); purpose of trip (business,
age daily aircraft movements at São square matrix, where rows and
leisure, or combined);
Paulo Airport (Figure 1). columns are associated with the n cri-
3.annual frequency of air trips; degree
of familiarity to major Brazilian air- teria (e.g. the APT components and
The Airports Council International their attributes). The Aij represents the
level of service manual (ACI, 2000) relative importance of row i criteria
was employed to pre-select the most over column j criteria (Figure 2).
These socio-economic variables will
important components and character-
be useful in future research to analyze
istics of components, according to the If all passenger judgments are consis-
groups of passengers that have com-
opinion of managers of 512 ACI air- tent for any i, j, k,
mon characteristics (same destination,
ports. Final selection of variables to be
included in the model was concluded Degree of Saa
with specialists of the Department of Importance ty
Air Transport and Airports of the Comp. Comp. Scal
Aeronautical Institute of Technology 1 2 e Importance Degree
(ITA) during meetings in May-July 0.90 0.10 9 Component 1 is extremely more important than Component 2
2006. 0.80 0.20 7 Component 1 is very important compared to Component 2
0.70 0.30 5 Component 1 is important compared Component 2
Table 1 presents the proposed vari- 0.60 0.40 3 Component 1 is less important compared to Component 2
ables indicated by the expert panel.
0.50 0.50 1 Both components have the same importance
0.40 0.60 1/3 Component 2 is less important compared to Component 1
A questionnaire was developed, in
which passengers could compare the 0.30 0.70 1/5 Component 2 is important compared Component 1
importance of one component (or 0.20 0.80 1/7 Component 2 is very important compared to Component 1
attribute) over another component (or 0.10 0.90 1/9 Component 2 is extremely more important than Component 1
attribute). This process is called pair-
Table 2: Comparison among the scale used in this research with the scale of Saaty.
wise comparisons. For instance, pas-

e-zine edition, Issue 37
They were obtained by application of
the methodology previously presented
in this paper. According to the Saaty
In practice, these values can be slight- scale (Table 2), the sum of values on a
Here, given level should be close to 0.90
ly different from what is expected.
i: 1, ..., m (number of criteria) (depending on the degree of consisten-
However, the method allows a given
degree of inconsistency (see Gomes et. cy).
A questionnaire was developed in
al., 2004), which will be evaluated
which passengers are asked to deter-
here to check if it is within acceptable According to Figure 3, the check-in
mine the importance of one APT com-
limits. counter (0.33) is the most important
ponent/attribute over another using the
component for departing passengers
Saaty scale. All calculations were done
Verbal judgments are transformed into that have been interviewed at São
according to the theoretical framework
a quantitative scale as indicated in Paulo/ Guarulhos International
described in this section. The results
Table 2. Such values are normalized Airport. The most important attributes
are presented in the next section.
through Equation 2. The priorities vec-
tor of sub criteria i (Aij) related to cri-
teria (Ci) are calculated through
Equation 3.

Figure 3: Weights of Importance for Components and Its Characteristics

i: 1, ..., n; Descriptive Analysis of Passen- for each component are presented in
v: vector; ger Responses bold. For instance, security is the most
A: criteria of second level (sub criteria – important attribute for the parking
e.g. courtesy);
A brief descriptive analysis was
accomplished using data obtained component with a degree of impor-
n: number of criteria of the a given level.
from passenger responses. In this case, tance of 0.48.
Equations 4 and 5 provide the weights the most common passenger groups
of importance of criteria. are represented by: male passengers, The values of Figure 3 could be further
age 21-30 years old, family income of employed to obtain a global LOS
US$ 10,000-20,000, frequency of measure for departing passengers at
travel – twice a year, familiarity with the APT. In this case, the AHP method
São Paulo airports only (domestic air- must be used to rate the LOS for each
port and international airport), busi- component. Note that the degree of
ness passengers, flying GOL airlines, importance and rating are different
and flying to domestic destinations. things: for instance, a passenger might
consider the check-in counter the most
The next section presents the degrees important component; however he/she
Here, could evaluate it as the component
of importance, which are valid for pas-
j: 1, ..., m with the relatively worst LOS at the
w: vector representing the weights of sengers at São Paulo/ Guarulhos
International Airport. The results APT, in comparison with the remain-
C: criteria being evaluated might differ from the perceptions of ing components.
m: number of criteria at a given level passengers flying at different airports,
especially if their socio-economic pro- Conclusions
Finally, the AHP-Group method will file differs significantly from those The work presented here represent a
be employed in order to obtain the surveyed. new step in airport passenger terminal
results for a group of s decision-mak- level of service modeling, since it indi-
ers, in our case, 103 passengers Degrees of Importance of cates the quantitative degree (weight)
(Gomes et. al, 2004). Given a group of Components and Their Attri- of importance of APT components and
s decision-makers, where each one is butes their attributes. Further research is
represented by the parameter k, being developed in order to analyze
Figure 3 presents the weights indicat-
k=1,..., s, the weights of criteria should these weights as a function of socio-
ing the degree of importance for the
be calculated by Equation 6. economic characteristics (family
group of 103 interviewed passengers.
income, destination, etc).

e-zine edition, Issue 37 3

Acknowledgements de Barros, A. G. (2006), A Global Index for About the Authors
The authors would like to thank FAPESP Level of Service Evaluation at Airport Michelle Bandeira and Anderson Correia
(São Paulo State Research Council) and Passenger Terminals, In: 10th Annual work at the Aeronautical Institute of
CAPES (Brazilian Federal Research World Conference - Air Transport Technology, São José dos Campos, SP –
Research Society (ATRS), 2006, Nagoya, Brazil. Chan Wirasinghe works at the
Agency) for the support provided in this
10th, p. 1-15. Schulich School of Engineering,
research and the Administration of São
Gomes. L. F. A. M. et al (2004), Tomada de University of Calgary, Canada.
Paulo/ Guarulhos International Airport for
Decisões em Cenários Complexos:
allowing the surveys at restricted areas of Introdução aos Métodos Discretos do
the TPS. apoio Multicritério de Decisão, São Paulo:
Ed. Thomson.
References IATA (1995), Airport development refer-
Airports Council Internacional (2000), ence manual, International Air Transport
Quality of service at Airports: Standards & Association, 8th Edition, Geneva.
Measurements, ACI World Headquarters, Infraero (2000), Aeroporto Internacional
Geneva, Switzerland, 2000. de São Paulo/ Guarulhos, GUA-GRL-900,
Ashford, N. (1988), Level of Service RE-512/Rn.
Design Concept for Airport Passenger Müller, C. and Gosling, G. D. (1991), A
Terminals: A European View, Framework for Evaluating Level of
Transportation Research Record 1199, Service for Airport Terminals,
TRB, National Research Council, Transportation Planning and Technology,
Washington D.C., pp 19-32. Vol. 16, pp 45-61.
Correia, A. R. (2005), Evaluation of level Saaty, T. L. (1980), “The Analytic
of service at airport passenger terminals: Hierarchy Process”, McGraw-Hill. New
individual components and overall per- York, 1980.
spectives, Ph.D. Thesis, University of Transport Canada (1979), A Discussion
Calgary, Canada. paper on Level of Service definition and
Correia, A. R. and Wirasinghe, S. C. Methodology for Calculating Airport
(2004), Evaluation of level of service at Capacity, Report TP 2027.
airport passenger terminals: a review of Transportation Research Board (1987),
research approaches, Transportation
Special Report 215: measuring airport
Research Record 1888, TRB, National
landside capacity, TRB, National Research
Research Council, Washington D.C., pp. 1-6.
Council, Washington D.C.
Correia, A. R., S. C. Wirasinghe, S. C. and

Become a Book Reviewer, Editor or Associate Editor!

Aerlines Magazine is always looking for people with an interest in the Air Transport sector willing and able to contribute to our rap-
idly growing e-magazine! Please do not hesitate to contact us if you are interested to join us. But please read the following first:

Aerlines had the following editorial positions available:

- As a reviewer, you are reading judging incoming manuscripts according to our editorial standards;
- As a contributor, you are entitled to regularly publish your research findings or thoughts (by means of a column or weblog on our site);
- As a lay-out editor, you are responsible for the lay-out of the website and articles for new editions of Aerlines;
- As a web editor, you are responsible for the technical maintenance and regular updates of our content management system.
Advanced knowledge in this field is necessary.

Required & desired knowledge, skills and experiences: Interest in the (global) air transport industry, interest in aviation sci-
ence and research is desirable, good knowledge of the English language and the ability to contribute regularly as we publish
four editions of Aerlines per year.

What is in it for you? Aerlines is a voluntary non-profit association which means that we do not offer a financial compensation.
However, we do offer you the opportunity to get in touch with a large, diverse and international aviation network.

More information: If you are interested in this position, you may contact for more information:

e-zine edition, Issue 37 4