Simulation of Departure Terminal in Soekarno-Hatta International Airport

Simulation of Departure Terminal in Soekarno-Hatta International Airport D. Novrisal 135, N. Wahyuni 24, N. Hamani 2, A. Elmhamedi 1, T. P. Soemardi 5 1 LISMMA (Laboratoire d Ingénierie des Systèmes Mécaniques et des MAtériaux), Université de Paris 8, France 2 LTI (Laboratoire des Technologies Innovantes), Université de Picardie Jules Verne, France 3 Department of Industrial Engineering, University of Mercubuana, Jakarta, Indonesia 4 Department of Industrial Engineering, University of Sultan Ageng Tirtayasa, Cilegon, Indonesia 5 Department of Mechanical Engineering, University of Indonesia, Depok, Indonesia (nadia.hamani@u-picardie.fr) Abstract - Congestion problem in the airport declines the passengers satisfaction. The passengers satisfaction is an important matter to the sustainability operation of the airport. Airport management has to develop a company program to improve the quality of service to the passengers. To regulate the passengers processes, it is important to define the performance of existing system. The objective of this research is to measure the performance the existing system of airport departure terminal and provide a suggestion to airport management of Soekarno-Hatta International Airport. Discrete-event simulation method was developed to measure the system. ProModel simulator software, from ProModel Corporation, was utilized to simulate the existing system of Departure Terminal in the Airport. To measure the performance, authors focus on some indicators i.e. passengers processing time, passengers waiting time, and utilization of each station. Keywords Modeling, discrete-event simulation, performance measurement I. INTRODUCTION The airport management is a complex system related to the passengers, airplanes and baggage. The passengers are the most important entities in the Airport. The sustainability of an airport depends on passengers because passengers are the users of airport service. Passengers satisfaction has to be the main goal of the airport service. The airport must have adequate facilities and public area to serve the passengers properly. Passengers satisfaction depends on the performance of airport in serving the passengers. Performance of facility in the airport can be measured by simulation techniques. Many researchers develop simulation techniques to measure the capacity of the airport, utilization of each area, passengers waiting time, and passengers processing time in the airport system. The main objective of this research is to measure the performance of the existing system of airport departure terminal and provide a suggestion to airport management of Soekarno-Hatta International (SHI) Airport. The paper is structured into seven main sections; Section 1 is related to the context and the problem statement; Section 2 is related to literature review; Section 3 related to research methodology; Section 4 presents modeling and simulation of departure terminal in the airport; Section 5 presents to results of the research; and Section 6 is the conclusion. A. The Departure Terminal SHI Airport is located on the Java Island, near the capital city of Indonesia. SHI Airport is the main airport in Indonesia and one of South East Asia s busiest airports [1]. SHI Airport was operated in 1985 and served more than 53 million passengers in the year of 2012 [1]. SHI Airport has three main terminals and six sub terminals. Every terminal in the airport consists of departure and arrival terminal. Congestion problems usually happen in departure terminals [2]. The research was done in the departure of terminal 1b which has several areas: - 4 stations in the screening area I - 42 stations in the check-in area - 4 stations in the airport tax area - 4 stations in the screening area II The departure process [2] will begin when passengers enter to the screening area I to check the ticket, the passengers and the baggage. If they do not have a problem with the baggage and the ticket, the passengers will go to check-in area to confirm the flight and the baggage will be weighted and numbered. In the next step, the passengers will pay for airport tax in the airport tax area. The last step of departure process is screening area II for screening process of the passengers and their cabin baggage before they go to the waiting room. B. Problem Statement Guizzi, Murino, and Romano [3] defined the congestion as a complex traffic in airside and landside of the airport. Passengers are subjected to increasing levels of congestion because of three factors; first is demand fluctuation, second is network problem, and third is flight schedule [4]. Many airports in the world have congestion problem particularly in the check-in and screening area. Savrasovs, Medvedev, Sincova [5] found that the check-in and equipment areas are the main problem zone in Riga International Airport because of the passengers number growth. Khadgi [6] focused on check-n area and the baggage screening area because from that lower level of the airport, the author had the various operations for the average time for passengers, processing times, idle times for the agents, and maximum queue content of baggage. 978-1-4799-0986-5/13/$31.00 2013 IEEE

Congestion problem in the airport could be the problem of decreasing of passengers satisfaction especially in the lower level of the airport. SHI Airport develops a company program to improve service quality to passengers. The objective of the company program is to prevent the congestion, increase capacity, simplify the processing procedure, etc. To regulate passenger process, it is important to define the performance of the existing system. In SHI Airport, we could find a long queue length in the peak hours. Passengers spent their time for waiting in a line. To improve this situation, we would like to measure the performance using simulation. II. LITERATURE REVIEW Simulation is a powerful tool to measure the performance of the airport system. Simulation method can predict the effect of various scenarios in the system without actually changing the real system. The researchers use simulation in different fields of the airport; check-in area, screening area, baggage handling system area, parking area, passengers access entry, runways, aprons, and taxiways. There are several studies which have been done by simulation techniques in airports. Thomet and Mustoufi [7] present the simulation techniques to measure the performance of new passenger terminal at Curacao airport. The authors use the TERMSIM software computer to confirm that the performance of the new terminal of Curacao airport in 2031 regard to International Air Transport Association (IATA) Level Of Service C (LOS C) [7]. In Olaru and Emery [8], discrete simulation using Extend V.6 platform software and generic algorithms were developed to optimize the airport facility plan of departure terminal at Broome International Airport according to IATA LOS C. In Savravosh, Mendvedev, and Sincova [5], simulation method was used to measure and to improve the performance of Baggage Handling System (BHS) in the airport of Riga. The authors develop different scenarios of passengers flow and number of officers to improve the system. Microscopic discrete-event simulation approach by Exstendsim software is used. Appelt, Batta, Lin, and Drury [9] developed simulation method to identify delays and to improve the efficiency by using different scenarios in the check-in procedure at the Buffalo Niagara International Airport. Basically, waiting time and processing time in the system are evaluated. Freivalde and Lace [10] were studying maximum capacity, found bottleneck, and suggested solutions to the management of Riga airport. The authors used GRADE software computer to simulate the first results, and for the further results simulation frame in Microsoft Excel computer software is used. Khadgi [6] wanted to solve how to reach the customer satisfaction in check-in counter, security checkpoint, and baggage claim with simulation. Using FlexSim simulation tool, Khadgi simulated three cases; Case 1 simulated changing the method of baggage screening procedure, Case 2 simulated changing the number of passengers, Case 3 simulated multiplied flights. Ahyudanari and Vandebona [4] using simulation to make a time block concept utilize the counting periods adopted in data collections for simplifying the analysis process. They consider two issues i.e. the expected quality of service at airport check-in areas and the congestion caused by flight scheduling. III. METHODOLOGY There are several steps to simulate the existing system of SHI Airport. The methodology of this research is shown in Fig. 1. 1) Objective: The first phase of the research methodology is determining the objective of the research. The objective of the research intends to measure the performance of existing system of departure terminal at SHI Airport and to provide suggestions to the airport management. 2) Model Description: After we determine the objectives, the next phase of the research is describing the model. The aim of this phase is to illustrate the real problem occurring in the system. 3) Modeling and Simulation: The third phase of the research is modeling and simulating the existing system. The simulation was done by ProModel Simulator version 7.5 from ProModel Corporation [11]. 4) Results: The last phase of the research is obtaining the results of simulation. The authors used some indicators, they are waiting time, processing time, and utilization of stations to measure the performance of existing system. From the output of ProModel Simulator, authors will provide suggestions to the management of airport related to the system improvement. The authors used some indicators to measure the performance. These are waiting time, processing time and utilization of locations. Objective Model Description Modeling and Simulation Results Fig. 1. Research Methodology

IV. MODELING AND SIMULATION In this research, we proposed to use ProModel simulator version 7.5 to simulate the existing system of departure terminal. Benson [12] describes ProModel simulator as: - A powerful yet easy to use simulation tool for modeling all types of manufacturing systems ranging from small job shops and machining cells to large mass production and flexible manufacturing systems, - It combines the flexibility of a general purpose simulation language with the convenience of a datadriven simulator, - Windows based system with an intuitive graphical interface and object-oriented modeling constructs that eliminate the need for programming. In this research, we used four basic elements of simulation by ProModel i.e. Entity, Location, Arrival, and Processing [12]. Entity is anything that a model processes; parts, products, people, materials, documents, or phone calls should be modeled as entities. The entities of this simulation research are the airport passengers, and their carried baggage. Locations are places in the system where entities are processed, storage, or some other activity. The location of this model is the screening area I, check-in area, airport tax area, and screening area II. We also define the queue before each station. So, there are nine defined locations that have been built. They are entrance in each station and the station itself. Arrival is any new entities coming into the system. In this research, the arrival was the airport passengers and their baggage when they come in to the system. We exclude the passengers transfer as an arrival because they are not coming into the system. Last element of the simulation is processing. Processing defines the routing of entities through the system and the operations that take place at each location they enter. To build simulation in the ProModel simulator, we need several kinds of data. The data are related to processing time, passengers arrival distribution, and layout of the terminal. The data collection of processing time was done in the screening area I, check-in area, airport tax area, and screening area II. Data was collected by direct time measurement using stopwatch and the random sampling method. Table 1 presents the data of processing time of each area in departure terminal. Passengers arrival distribution is describing the pattern of passengers arrival per unit of time in a specific time period. IATA Earliness Distribution approach is used to find the pattern of passenger s arrival distribution. IATA Earliness Distribution method is combining the flight schedule and the percentages of number passengers coming into the airport before departure time of airplane [4]. TABLE I. PROCESSING TIME No Location Processing Time (seconds) 1 Screening Area I 29.94 2 Check-in Area 145.58 3 Airport Tax Area 12.29 4 Screening Area II 13,53 To build passengers arrival distribution, we need the airplane schedule and the number of passengers. We collected the data from the company, then we combined the data with IATA Earliness Distribution to find passengers pattern. Fig. 2 presents passengers arrival distribution of SHI Airport in a normal day [2]. As we can see in Fig. 2, the peak time of passengers arrival is in the morning around 05:00-09:00 and in the afternoon around 15.00. In the other side, there are no passengers at all in the early morning around 01.00 to 03.00 or a few number of passengers in the night. This period is the peak-off time. Third data that we collected from the company is the layout of departure terminal. In this area, the entity will flow starting from entrance through all the stations until exit the system. The layout of the departure terminal is presented in Fig. 3. To evaluate the performance of the existing system, the authors utilized some indicators. Indicators are instrument that we used to know the achievment or performance of the system. In this research, the main indicators are entity processing time, entity waiting time and utilization of location. Fig. 2. Passengers Arrival Distribution [2]

A. Entity Activity Fig. 3. Layout Departure Terminal V. RESULTS The results obtained through ProModel Simulator summarized in two categories; Entity Activity and Utilization of Location. The entity activity is used as one of the most important characteristic for determining passengers satisfaction level. Entity activity shows how much time the passengers in the system related to average time in the system, average time in waiting time, average time in the operation, etc. The result of the simulation is shown in Table II. Utilization of location determined loading level of the location in the airport. Utilization of locations can be used to find bottleneck in the system. From this table we could define how much time passengers spent in each station (Screening I, Check-in, Airport Tax, and Screening II). Name TABLE II ENTITY ACTIVITY Passengers Total Exits 15670 Current Quantity in System Average Time in System Average Time in Move Logic Average Time Waiting Average Time in Operation Average Time Blocked 0 25.63 4 15.92 3.57 2.14 Total exits are the passengers that completely exit the system. From Table II, in 24 hours of simulation, there are 15670 passengers in the terminal 1b of SHI Airport. It means that the capacity of terminal 1b is insufficient to serve the passengers. The capacity of terminal 1b now is around 8219 passengers per day [1]. That number was clearly evident that there s over capacity problem. Current quantity in system is the total number of passengers remaining in the system at the time the simulation ends. The result indicates that all passengers are exit from the system. Average time in the system is the average total time the passengers spend in the system. From the table above, the average time in the system is 25.63 minutes. If we consider the processing time in each station, there was no processing time more than 3 minutes, so, one passenger has to spend 25.63 minutes in the system which indicates the delay and bottleneck in one station/location. Average time in move logic is the average time the passengers spend traveling between locations, including any delays incurred in move logic. The result show that average time in move logic is 4 minutes. That means the passengers spent time for moving from one location to the next location for 4 minutes. Average time waiting is the average time the passengers spend waiting for a resource; it also includes the time waiting in queue behind blocked passengers. The average time waiting in this simulation result is 15.92 minutes. It can be indicator of the bottleneck that passengers have to wait for 15.92 minutes in the system. Average time in operation is the average time the passengers spent processing at a location or moving on a queue. From the table, the average time in operation is 3.57 minutes. In this part, a passenger in each location has to spend for 3.57 minutes to be served. Average time blocked is the average time the passengers spent waiting for a destination location to have available capacity from the table above (Table II). The results show that the average time blocked is 2.14 minutes. B. Utilization of Locations Utilization is the percentage of capacity occupied, on average, during the simulation. From the obtained results of ProModel simulator, the utilization of each location in the departure terminal is presented in Fig. 4. As shown on Fig. 4, there was no utilization close to 100%. This is because there are periods of peak time and off-peak. In the peak time, when the number of passengers increases, the station will get very busy, whereas in the off-peak, there is a time that none of passengers is coming, so that there is no activity in the station.

terminal because check-in area has the highest utilization. It is the bottleneck of the system with the longest time passenger in the queue line (more than 61%). After measurement of the performance of the existing system, further work will focus on improving the baggage handling system of departure terminal at SHI Airport. We have to study the effect of increasing the number of check-in stations on the performance of the airport using simulation methods. The objective of improvement is to increase passenger satisfaction by reducing processing time and waiting time in the system. REFERENCES Fig. 4. Utilization of locations From Fig. 4, we also could observe the Check-in station has the highest number of utilization. On the contrary, the utilization of Screening II has the lowest number. The reason is there are passengers queuing in Airport Tax station, so the flow of passengers is stagnating in the Airport Tax station. Therefore passengers who have passed the Airport Tax station will pass Screening II more easily without significant queuing. Based on results of simulation, the longest time passenger was in the queue line more than 61%. This means that the airport has a congestion problem. We suggest that the management of the airport has to improve the system to reduce the processing time and waiting time. Passenger s satisfaction depends on the processing time in the system because the less time they spend time in the system, more they are satisfied [13]. Based on the utilization chart, we could make a suggestion to the management of airport to increase the number of check-in stations. Check-in area was the bottleneck of the system because it has the highest utilization. The improvement can be reached by increasing the capacity of the airport. VI. CONCLUSION Discrete event simulation has been used to measure the performance of the existing system of departure terminal at Soekarno-Hatta International Airport. ProModel simulator version 7.5 of ProModel Corporation is used in order to simulate the system. The study consists of four steps: determining the objective, description of the model, modeling and simulation, and results. The results obtained through ProModel Simulator were summarized in two categories; Entity Activity and Utilization of Locations. The key results belonging to the category Entity Activity are summarized in Table II. After simulating the Utilization of Locations, we suggest to the airport management to increase the number of check-in number in departure [1] PT. Angkasa Pura II [Electronic Resources]: www.angkasapura2.co.id [2] D. Novrisal, N. Hamani, A. El Mhamedi, and T. Soemardi, Data collection of the baggage handling system at Soekano-Hatta International Airport, in Proc. 8ème Conférence Internationale Conception et Production Intégrées, CPI 2013, Tlemcen, Algérie. [3] G. Guizzi, T. Murino, and E. Romano, A Discrete Event Simulation to model passenger flow in the Airport Terminal, in Proc. 11th WSEAS international conference on MAthematical methods and Computational Techniques in Electrical Engineering, MACTEE'09, Vouliagmeni, Athens, Greece, pp. 427-434. [4] E. Ahyudanari, and U. Vandebona, Simplified model for estimation of airport check-in facilities, Journal of the Eastern Asia Society for Transportation Studies, vol. 6, pp. 724-735, March 2005. [5] M. Savrasovs, A. Mendvedev, and E. Sincova, Riga Airport baggage handling system simulation, in Proc. 23rd European Conference on Modeling and Simulation, ECMS 09, Madrid, Spain, pp. 384-390. [6] P. Khadgi, Simulation analysis of passenger check-in and baggage screening area at Chicago-Rockford International Airport, NIU Engineering Review Journal, vol. 1, no. 1, pp. 29-34, Spring 2009. [7] M. A. Thomet, and F. Mostoufi, Simulation-aided airport terminal design, Bechtel Technology Journal, vol. 1, no. 1, pp. 43-48, December 2008. [8] D. Olaru, and S. Emery, Simulation and GA-optimasion for modeling the operation of airport passengers terminal, in Proc. 29 th Conf. of Australian Institutes of Transport Research, CAITR 07, Adelaide, Australia. [9] S. Appelt, R. Batta, L. Lin, and C. Drury, Simulation of passengers Check-in at a medium-sized US Airport, in Proc. 39th Winter Simulation Conference, WSC'07, Washington, DC, USA, pp. 1252-1260. [10] L. Freivalde, and L. Lace, Improvement of passengers flow management in an Airport Terminal, in Proc. 5 th International Scientific Conference on Business and Management, BM 08, Vilnius, Lithuania, pp. 659-664. [11] ProModel Corporation, ProModel versions 7.5 users guide, 2008. [12] D. Benson, Simulation modeling and optimization using ProModel, in Proc. 29th Conference on Winter Simulation, WSC 97, Atlanta, G, USA, pp.587-593. [13] I. O. Manataki, and K. Zografos, A generic system dynamics based tool for airport terminal performance analysis, Transportation Research part C, vol. 17, pp. 428-433, 2009.