Development and application of an airport terminal simulation model - A case study of CKS Airport

大學]

On: 28 April 2014, At: 05:22 Publisher: Routledge

Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Transportation Planning and

Technology

Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/gtpt20

Development and application

of an airport terminal

simulation model — a case

study of CKS airport

J.‐T. Wong a & T. C. Liu a a

Institute of Traffic and Transportation , National Chiao‐Tung University , Taipei, Taiwan, 10012 , ROC

Published online: 21 Mar 2007.

To cite this article: J.‐T. Wong & T. C. Liu (1998) Development and application

of an airport terminal simulation model — a case study of CKS airport, Transportation Planning and Technology, 22:1, 73-86

To link to this article: http://dx.doi.org/10.1080/03081069808717620

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever

relation to or arising out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Photocopying permitted by license only the Gordon and Breach Science Publishers imprint. '• Printed in India.

DEVELOPMENT AND APPLICATION

OF AN AIRPORT TERMINAL

SIMULATION MODEL - A CASE STUDY

OF CKS AIRPORT

J.-T. WONG and T.C. LIU

Institute of Traffic and Transportation, National Chiao-Tung University, Taipei, Taiwan 10012, ROC

(Received 16 February 1998; In final form 23 February 1998)

Technology advancement, terminal user behavior and changes of service characteristics of terminal facilities all have great impact on airport terminal operations. Consequently, the adequacy of traditional airport terminal planning concepts and standards have recently been challenged and are worth being explored.

To investigate the associated impact of environmental changes on terminal operations, a simulation model which takes into consideration air travel patterns, facility operational characteristics, flight delays, passenger behavior and needs, etc. is developed and verified. The simulation logic and results show that the model works well for the case of CKS (Chiang Kai-Shek International Airport, Taipei) operations. Also, the exploration demon-strates that passenger arrival patterns, numbers of group passengers, flight delays and load factors all have very significant influences on terminal space requirements. This result strongly suggests that local characteristics should not be neglected in planning terminal operations.

Keywords: Airport terminal; Simulation; Space requirement

1. INTRODUCTION

In general, passenger behavior, flight delays and service characteristics are not taken sufficiently into account in the airport terminal planning process. Terminal space requirement is roughly set by means of stan-dard formulae or procedures. As a result, terminal space is often not adequate for actual operations.'1'21 In reality, passengers may arrive

73

much before their flight time due to the inconvenience of the ground access system. Flights may be delayed due to weather conditions or airport congestion. In such cases, passengers may overcrowd the terminal. Therefore, if these related factors are not carefully explored during the planning process, the terminal may be operated in a less than acceptable manner. It will be either not fully utilized or over-crowded. It is for this reason that in this study a variety of factors, along with their impact on terminal operations, are investigated.

An airport terminal is an air transportation facility. Transportation is its very basic and most important function. It provides convenient processing, mode transfer and a comfortable holding area for air pas-sengers.'31 To meet the service requirement, there must be adequate

terminal facilities and space. Traditionally, these facilities and space are directly related to the number of passengers and are used only for transportation purposes. However, airports are becoming more than just a place for aircraft to land and take off. In a modern airport, commercial space not directly necessary for air transportation is attracting great attention. Increasingly, airports around the world are shifting to private ownership. With sparkling shopping malls, high-rise hotels and connecting business plazas, some airports have evolved into commercial hubs for surrounding communities, competing against nearby metropolitan centers for travelers' cash once spent mostly downtown. One example is Frankfurt Airport in Germany. Housing a hotel and convention center under one roof, it aims to create an "airport city" with sophisticated urban functions. The other example is Amsterdam Schiphol Airport in the Netherlands. Its terminal includes a hotel, tax-free shopping center, a business center with relevant facil-ities, etc. Yet another example is Changi Airport in Singapore. It has around 100 shops in terminals 1 and 2. The facility is like a huge shopping center. Aside from the shops, restaurants, hotel and other facilities also contribute to the mini-city atmosphere. In the case of Taiwan's Chiang Kai-Shek International Airport, the plan is that it will also be converted into an "airport city" as part of the government's intention to make Taiwan a regional operations center. As a con-sequence, these land-side facilities are becoming the major source of income for many major international airports.

More commercial space available in an airport may be a trend. However, there are no hard and fast rules to follow. Utilization of

this space is very much dependent on the philosophy of the airport management. It completely depends on the airport operator's decision. When it is considered desirable to provide more space for duty-free shopping, recreation, and business offices, an airport terminal requires more space. In other words, the space for ancillary facilities varies with the business goals of the airport operator.

For transportation planners, their major concern is the impact of environmental changes on terminal operations and the associated space requirement. Therefore, the main theme of this paper will focus on passenger traffic characteristics and their impact on terminal operations. Furthermore, inbound and outbound operations in the CKS (Chiang Kai-Shek, Taipei) terminal are separate. They can be analyzed independently. Since outbound operation is much more affected by external factors, in this study we will focus only on the space and facilities needed for processing, circulation and waiting of outbound passengers.

2. FACTORS AFFECTING TERMINAL SPACE REQUIREMENT

Terminal space includes processing, holding, circulation and ancillary units. After reviewing the relevant literature,11"81 we recognize that

most studies do not consider comprehensively factors that affect these terminal space requirements. Empirical and statistical formulae are commonly applied to size individual processing facilities within the terminal buildingJ2'41 A simple queueing approach with the

assump-tions of Poisson arrival and exponential service time distribution is frequently adopted.15"71 However, the arrival and service patterns vary

with the time of day. They cannot adequately be described as having a fixed distribution. Besides, some important factors cannot be carefully taken into account in the queueing approach. As a result, the simple queueing approach and statistical formulae may not be appropriate for space requirement decision-making. To plan these space units prop-erly for passengers at a specified 'level of service' (LOS), one should clearly understand the various factors which affect the space required (only for transportation purposes) and how the space is affected.

These factors include:

(i) Passenger flow Given a set of terminal operating conditions, peak occupancy and waiting time in a terminal is simply determined by the peak passenger flow. Thus, the most important element in com-puting terminal space is peak-hour passenger flow into and out of the terminal. To estimate the number of peak passengers, US-FAA has suggested a ratio of typical peak hour passengers (TPHP) to annual passengers as shown in Table I. However, this may differ significantly from the actual number. The actual number of peak-hour passengers is based on actual flight schedules which are determined by many factors such as market demand, availability of aircraft, etc.

(ii) Passenger characteristics Passenger characteristics include pas-senger arrival patterns and paspas-senger behavior in the terminal. Differ-ent passenger characteristics will result in differDiffer-ent terminal space occupancy.'1'21 Ashford[10] has pointed out that passengers on different

flight routes may have different arrival patterns. Similar results were found in our survey. Table II reveals that passengers with different trip purposes, ground access times and personal characteristics exhibit many differences in their preferred terminal arrival time. Passengers on European/Australian routes arrive at the terminal much earlier than those on other routes. This is because most passengers surveyed on European/Australian routes were traveling as leisure/group passengers. On the contrary, passengers on the Hong Kong route arrive at terminal not so early as their counterparts on other routes. This is because of the very high frequency on the route and a cooperative agreement between China Airlines and Cathay Pacific Airways. Once a flight has been missed, a passenger can get a seat soon on another flight without much difficulty. As to the behavior of group passengers, who share a great proportion of the market, an even greater difference is found. Table III

TABLE I Ratio of TPHP to annual passengers"1]

Annual passengers TPHP IAnnual passengers (%)

> 20,000,000 0.030 10,000,000-19,999,999 0.035 1,000,000-9,999,999 0.040 500,000-999,999 0.050 100,000-499,999 0.065 < 100,000 0.120

TABLE II Arrival patterns of passengers on different flight routes Flight route Hong Kong Japan North American European/Australian TABLE III Type of passenger Individual Group

Arrival time before flight departure (min) Mean 81.5 110.6 103.6 147.1 Standard deviation 28.3 33.4 31.8 28.5

Arrival patterns of individual and group passengers

Arrival time before flight departure (min) Mean 97.4 162.5 Standard deviation 30.1 12.9

shows this difference is statistically highly significant. Group passen-gers are asked to arrive at the terminal much earlier. They generally arrive at the terminal together by chartered bus. The earlier the passen-gers arrive at the terminal, the longer they occupy the terminal space. Consequently, to maintain a reasonable level of service when there is a large number of group passengers, providing more space should be considered. This implies that passenger characteristics should be seri-ously taken into account.

(iii) Service characteristics Along with technology advancements and revolutions in airport management, airport operators frequently introduce new facilities and operating strategies to improve terminal service quality and solve capacity deficiency problems. Because of such changes, passengers must adjust their patterns of behavior while using the terminal. For example, common use of terminal equipment reduces waiting time at the check-in counter. It also reduces the time passengers may spend in the check-in lobby. The use of smart cards for ticketing, check-in and security checks will further reduce the processing time and increase the passenger flow rate. If a terminal is further equipped with an efficient people-mover, the time needed to process its air passengers will be significantly reduced. Momberger[121

has pointed out that a passenger can complete the whole process within 45 minutes if new technology is applied in the terminal. Thus, change of operating characteristics should not be ignored.

(iv) Flight delays When flight delays occur, a terminal will be crowded with not only passengers for scheduled flights but also passengers of delayed flights. Those delayed passengers obviously uti-lize more terminal space than originally envisaged. Jovanovic[131

analyzed Ljubljana Airport flight delays and concluded that the prob-ability of a change in the flight departure sequence is about 9.26%. The number of passengers in the central hall will accordingly increase between 50% and 80%. Our study, employing flight delay data from CKS airport,1141 shows that the number of peak passengers in the

check-in area based on actual flight departures is 9.12% more than the number based on the flight schedule with no delays taken into account. Since the possibility of flight delays cannot be eliminated, it is better to have this factor considered in determining the amount of terminal space.

(v) Design standard (LOS) Planned LOS for future terminal operation is one of the key factors in determining adequate space. The higher the LOS is set, the more space should be allocated. Traditionally, terminal space planned by traffic engineers is based on studies for other modes of transportation. However, those facilities are differ-ent from the airport terminal facility.'151 Davis and Braaksma'161

developed a standard for passengers moving from one area to another within the transportation terminal as shown in Table IV. Ashford'171

reviewed standards for a variety of airport terminals. It appeared that standards for different airports were different. The space for each passenger varied from 1 to 3 m2. Since this has a great influence on

the terminal space requirement, an airport authority should consider LOS, cost and its goal very carefully so as to make a more informed decision.

TABLE IV Platoon flow LOS criteria for transportation terminals'16'

LOS A + A B C D E F Speed (m/s) >1.4 1.3 ~ 1.4 1.2~ 1.3 1.1 ~ 1.2 1.0~l.l 0.7 ~ 1.0 <0.7 Flow (ped./min/m) < 3 7 37~46 46~57 57~68 68~75 75~57 < 5 7 Area module (m2) >2.3 1.7 ~ 2.3 1.3~1.7 1.0~1.3 0.8 ~ 1.0 0.7 ~ 0.8 <0.7

3. TERMINAL SIMULATION MODEL

A terminal consists of well-organized space units. These are linked on the basis of passenger handling procedures. In order to manage and apply the simulation model, the whole model is divided into four parts as shown in Fig. 1. The first part is from the terminal entrance to the entrance of the emigration procedures area. The second part is from the entrance of the emigration procedures area to the security-check area. The third is from the security-check area to the entrance of the departure lounge. The last part is from the entrance of the departure lounge to the boarding ramp.

Based on the known or simulated flight schedule, aircraft type, load factor and passenger arrival pattern, a variety of passenger-related characteristics are generated. Every minute, passenger activities and the movement in each block of the terminal are scanned and the related statistics are computed.

To be clear, the logic of our simulation model is described as follows. Since every block of the simulation model consists of proce-dures for processing, holding and circulation, it is worthwhile address-ing the concept of passenger handladdress-ing in these three space units. First, to handle the passengers at the processing facility, service policies governing the number and time of counter opening and closing are set on the basis of a field survey. Given the passenger arrival pattern and the facility service policy, the passengers are processed and move within the terminal during the course of the simulation period. Second,

part 3', Departure;part 4 Departure Lounge Security <i — check 'FT"") ; part 3 Emigration procedures part 4 Departure J—g» Lounge 'Departure Ancillary facility ±. parti I UA SQ CX PR MH CI NW CP OF VN

FIGURE 1 The simulation model structure.

to assess passengers using ancillary facilities such as shops and restau-rants, a survey is conducted so as to collect information about how passengers spend slack time. The survey result is then used for the model input to simulate passenger activities in the ancillary facility. With minute-by-minute scanning set in the simulation model, time variation of space occupancy and passenger flow at the relevant area and facility can thus be recorded and analyzed.

4. MODEL VERIFICATION AND VALIDATION

Simulating terminal space requirement involves many complex vari-ables. On any given day, passengers are arriving continuously for check-in, emigration procedure, security check, etc. The occupancy of space in every unit is continuously changing. Uncontrollable factors alter the simulation result. Due to these uncontrollable factors such as flight delay, passenger arrival pattern and passenger characteristics, etc., it is hard to control very accurately the parameters of the model input. Consequently, it is difficult to validate the model output. None-theless, the model can be verified by tracing the simulation output. After verifying the correctness of the model logic, we use the data col-lected from the field to validate the model. By tracing the model output, we find the pattern of open and closed check-in counters well within our expectations. Moreover, compared with the number of passengers computed from flight timetables, the outputs of passenger flow and occupants of associated spaces also have expected results.

Finally, we validate the model via comparison between the real data and the model outputs of passenger arrival patterns both at the terminal entrance and the boarding area. The results are shown in Figs. 2 and 3. Estimates of R2 of the simple regression equations equal

0.77 and 0.93, respectively. Moreover, if the simulated results fit perfectly to the observed, the coefficients of the regression equation should equal 1. As expected, both t values suggest that under 95% confidence level, the hypothesis that the coefficients are equal to 1 could not be rejected. Furthermore, we adopted Theil's inequality coefficient.

If U=0, there is a perfect fit. If U= 1, the predictive performance of the model is as bad as it possibly could be. The U values of the

i i l i l i i i l i i i s i

FIGURE 2 Passenger arrival at the terminal entrance.

Passengers 3500 . 3000 • 2500 2000 1500 . - - - -Observed V -\ \

:\V

i I I i

Simulated 3000 2500 2000 1500 1000

. , - ,

\ v < > - - i - . "

FIGURE 3 Passenger arrival at the departure lounge.

simulated results shown in Figs. 2 and 3 are 0.099 and 0.098, respec-tively. From these results and statistics, we can see that the simulation model works satisfactorily. In the following section, we will apply it to analyze the impact of various affecting factors.

5. MODEL APPLICATION

Factors to be investigated include annual number of passengers, route distribution, load factor, group passenger and arrival pattern. The basic input of the experiments is listed in Table V, which corresponds to the 1995 survey situation. The flight schedule of the representa-tive case will be applied through the experiments except those cases listed in Table VI. In those cases, the timetables are generated on the

TABLE V Basic input of the simulation experiments

Item

Annual outgoing passengers

Regional flights: intercontinental flights Load factor

Arrival pattern of group passengers Arrival pattern of individual passengers

No. of group passengers: No. of individual passengers No. of male: No. of female

Input 6.5 million 80%: 20% 60% 260-95 min 180-15min 50%: 50% 70%: 30%

TABLE VI Outputs of flow and occupant for various annual passengers

Annual outbound passengers (million) Occupant (pax) Check-in area Ancillary area Emigration procedures Departure lounge Flow (pax/h) Check-in Emigration area Departure lounge 0.5 162 67 19 116 166 227 227 1 303 174 32 204 32 349 319 5 1277 868 145 583 1481 1685 1618 10 2588 1601 318 1191 2759 3365 3349 20 30 40 5626 7950 10492 3167 4620 6391 665 916 1238 2542 3642 4586 5684 8453 11454 6855 9914 13377 6732 9812 13329

same pattern as the existing timetable. The simulation results are summarized in Tables VI-X. Some points drawn from these tables are stated briefly below:

(i) The increases of terminal occupants and passenger flow are approximately proportional both to the increase of annual pas-sengers and to the increase of load factors.

(ii) Differences of peak passenger flow and occupants among cases with different ratios of intercontinental flights may reach as high as 30% or more.

(iii) The proportion of group passengers in a terminal will have a tremendous impact on the space requirement. It, however, has far less impact on the peak passenger flow.

(iv) The passenger arrival pattern also has very significant impacts on the space requirement. Similarly, it has only little impact on the peak passenger flow.

(v) Table VIII shows that since during peak season the load factor can reach as high as 90% or more, the check-in area - which at CKS occupies 2336 m2 - can be crowded with more than 2500

passengers. This means the space allotted to each passenger

during the peak period is less than 1 m2. The terminal obviously is

operating at near capacity. Again, this result meets the current terminal situation. To alleviate congestion, a second terminal is under construction and scheduled for operation in 1999.

TABLE VII Outputs of flow and occupant for various route distributions

Region/'Inter- 100:0 90:10 80:20 70:30 60:40 50:50 40:60 30:70 20:80 10:90 0:100 continental flights Occupant (pax) Check-in area 2215 1828 1767 1838 2025 1868 2171 2170 2069 2457 2419 Ancillary area 1312 1178 997 1082 1133 1242 1213 1334 1355 1507 1533 Emigration 246 230 197 190 209 209 205 243 233 257 280 procedures Departure lounge 904 835 837 790 1090 967 1000 1145 1157 1202 1480 Flow (pax/h) Check-in 2193 1816 1743 1971 2090 2071 2149 2149 2226 2524 2503 Emigration area 2589 2088 2128 2151 2408 2421 2666 2627 2472 2582 2981 Departure lounge 2544 2108 2119 2178 2404 2467 2687 2623 2482 2806 2929

TABLE VIII Outputs of flow and occupant for various load factors

Loadfactor 60% 70% 80% 90% 100% Occupant (pax) Check-in area Ancillary area Emigration procedures Departure lounge Flow (pax/h) Check-in Emigration area Departure lounge

TABLE IX Outputs of flow and occupant for various ratios of group passenger

Ratio of group passenger 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Occupant (pax) Check-in area 653 850 1066 1197 1434 1683 1804 2093 2672 2926 2945 Ancillary area 931 968 997 1008 1019 1111 998 1057 996 837 881 Emigration procedures 195 210 195 189 196 202 197 196 220 190 205 Departure lounge 431 535 717 643 761 790 788 1031 1016 1083 1064 Flow (pax/h) Check-in 1822 2029 1976 1748 1749 1731 1718 1749 1749 1752 1765 Emigration area 1985 2215 2151 2014 2033 2048 2011 2211 2124 2127 2148 Departure lounge 2188 2331 2291 2058 1998 1995 1944 2162 2130 2011 2020 1691 1076 201 807 1723 2007 2094 1938 1166 218 960 2095 2313 2346 2225 1387 231 1061 2208 2377 2317 2568 1460 257 1151 2580 3006 3017 2940 1677 339 1658 3040 3672 3659

TABLE X Outputs of flow and occupant for various passenger arrival patterns

Time of the first arrival {minutes before departure)

Group 260 230 200 170 140 Non-group 180 150 120 90 60 Occupant (pax) Check-in area Ancillary area Emigration procedures Departure lounge Flow (pax/h) Check-in Emigration area Departure lounge 1863 1478 1241 889 626 1006 1069 951 951 888 216 209 194 190 198 886 666 591 468 415 2058 1786 1884 1760 1708 2514 2074 2363 1995 2084 2489 2142 2474 2160 2233 Vehicles 80 _ — „ _ 70 60 30 40 30 20 10 :3 9 :0 9

V H

FIGURE 4 Vehicle arrival at the curbside.

In addition, this simulation tool was applied to curb planning. Simulating passenger arrivals and thus transferring them to vehicle arrivals, we then can assess curb utilization demand. To demonstrate applicability of the model, a case study was conducted. The results from the volume count and simulation are shown in Fig. 4. Again, it suggests that the performance of the model is acceptable.

6. CONCLUDING REMARKS

The simulation results demonstrate that the traditional procedures used by traffic engineers may not be in agreement with real operations. It also implies that the transferability of planning standards and pro-cedures should be very carefully examined. Local conditions should be deliberately incorporated, especially for those airports with a large proportion of group passengers.

During the study process, we found that searching a set of inputs which can represent the airport conditions may be difficult but is nevertheless important. Since the various changes can have significant impacts on terminal operation, current operations may not be appro-priate for the model input. Therefore, it is always worth pondering over the input issue when a terminal simulation is to be used.

Setting a planning guidance suitable for most airports is an impor-tant task. The key, however, lays in the understanding of the related factors and their impacts. Thus, more simulation runs with careful examination and interpretation are urgently needed.

A cknowledgments

The authors wish to acknowledge the financial support of the National Science Council, Taiwan, ROC for the funding of the research project on which this paper is based.

References

[1] T.J. Foster, N.J. Ashford and N.N. Ndoh, "Knowledge based decision support in airport terminal design," Transportation Planning and Technology, 19, 165-185 (1995).

[2] A.R. Odoni and R. de Neufville, "Passenger terminal design," Transportation

Research A, 26A(1), 27-35 (1992).

[3] N. Ashford and P.H. Wright, Airport Engineering, 3rd Edition, John Wiley & Sons Ltd., New York (1992).

[4] V. Tosic, "A review of airport passenger terminal operations analysis and modeling,"

Transportation Research A, 26A(1), 3-26 (1992).

[5] F.X. McKelvey, "Use of analytical queuing model for airport terminal design,"

Transportation Research Record 1199, Transportation Research Board,

Washington DC (1988).

[6] N. Ashford, M. O'Leary and P.D. McGinity, "Stochastic modeling of passenger and baggage flows through an airport terminal," Traffic Engineering and Control, 17(5), 207-210(1976).

[7] H.P. Piper, "Estimation of passenger flow in departure halls," Airport Forum, 20(3), 50-53, June (1990).

[8] S.A. Mumayiz, "Overview of airport terminal simulation models," Transportation

Research Record 1273, 11-20, Transportation Research Board, Washington DC

(1990).

[9] IATA, Airport Terminals Rejerence Manual, 7th Edition, International Air Transport Association (1989).

[10] N. Ashford, N. Hawkins, M. O'Leary, D. Bennetts and P. McGinity, "Passenger behavior and design of airport terminals," Transportation Research Record 588, 18-26, Transportation Research Board, Washington DC (1976).

[11] N. Ashford, H.P.M. Stanton and C.A. Moore, Airport Operations, John Wiley & Sons Ltd., New York (1984).

[12] M. Momberger, "Speeding up air travel on the ground," Airport Forum, 25(3), 32-34, June (1995).

[13] T. Jovanovic, "Modeling a real airport flight schedule for outgoing traffic," Airport

Forum, 20(4), 50-54, August (1990).

[14] J.T. Wong, "Modeling flight delay at CKS airport," Chiao Ta Management Review,

15(1), 19-37 (1995).

[15] C. Müller and G.D. Gosling, "A framework for evaluating level of service for airport terminals," Transportation Planning and Technology, 16, 45-61 (1991).

[16] D.G. Davis and J.P. Braaksma, "Level-of-Service standards for platooning pedes-trians in transportation terminals," ITE Journal, 57(4), 31-35, April (1987). [17] N. Ashford, "Level of service design concept for airport passenger terminals - a

European view," Transportation Research Record 1199, Transportation Research Board, Washington DC (1988).