Airport Passenger-Related Processing Rates Guidebook (2009)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

24 This chapter begins with a general treatment of metrics and levels of measurement. Airport specific metrics are then considered, and entities, resources, and processes are defined as they relate to the air- port environment. 4.1 Metrics Overview The term metric is often used synonymously with the terms measure and indicator. A metric is an indicator insofar as it points to something else (as in the way a person uses an index finger for pointing). It also connotes performance. Simply, data are collected with a purpose—to assess how something is performing, should perform, etc. The amount of time to process a given type of passenger, for example, might be an indicator of service quality, or efficiency. Metrics are sometimes categorized as being counted or measured. One counts bags, passen- gers, number of domestic flights, etc. These largely objective measures are relatively easy to deal with, and are the focus in this Guidebook. By contrast, one may measure passenger satisfaction, agent courteousness, and other attributes, but these concepts are abstract and, as such, more dif- ficult to assess. They are largely outside the scope of this Guidebook. At the risk of stating the obvious, error is unavoidable if there is inconsistency in how funda- mental terms are defined or interpreted. Ambiguity will invariably result when collecting data in the airport terminal environment if terms are not fully defined. For example, Exhibit 4-1 shows three acceptable definitions for the term “passenger.” Note that while no definition is inherently correct or incorrect, failure to define the term will likely not only result in confusion but, ulti- mately, in a serious reduction in the value of the collected data. An operational definition specifies precisely how counts or measurements will be made. Developing consistent operational definitions not only reduces the likelihood of miscommunica- tion among persons associated with a particular data collection effort, but it increases the likelihood that data collected at different locations and at different times can be aggregated and compared. As noted in Chapter 1, the impetus for developing this Guidebook was the realization that while a large amount of data were available, definitions of the entities and other issues (e.g., time, location) were often missing; when definitions were documented, they were fre- quently defined differently in different studies. These limitations made comparisons across data sets impracticable. The metrics described in this Guidebook should be sufficient for many airport passenger- related data collection efforts. Inevitably, however, you will encounter situations for which the research questions will require you to develop additional metrics. When this occurs, it is sug- C H A P T E R 4 Designing the Methodology Do not assume that everyone defines terms similarly. Select or create clear unambiguous operational definitions.

gested that you document how the metric was defined. Doing so, and then sharing that defini- tion with others, will increase the likelihood of eventually having a common data base useful to all persons and organizations concerned with improving airport performance. 4.2 Levels of Measurement Before considering airport-specific metrics, consider another perspective on data—the level of measurement at which the data is collected. Consider this because the level at which you col- lect data impacts the calculation of sample size. It also might restrict the types of analysis that are permissible. Based on work by Stevens,1 social and behavioral psychologists customarily distinguish among four measurement levels or categories: nominal, ordinal, interval, and ratio.2 (As a pneumonic, you might consider the word NOIR, as in a bottle of Pinot Noir wine as shown in Exhibit 4-2). Each measurement level is briefly described below. 4.2.1 Nominal These are categorical data that can be counted. For example, if you count the number of busi- ness class passengers, you are collecting nominal level data. If, for simplicity, you decided to code men with the number 1 and women with the number 2, these numbers are arbitrary. You might have coded men as 0 and women as 6; the frequencies are meaningful, but the numerical substi- tute for a label has no quantitative meaning. The late New York Yankee Mickey Mantle wore the number 7. If you went to a sporting event, you might receive a program that matches each num- ber with the name of a team’s member. Imagine that you were counting passengers’ bags, and you classified them as small, medium, and large. As you noted each bag, you would classify it by size and increase by one the number of elements in the group by which you classified it. 4.2.2 Ordinal When data are at an ordinal level, the numbers, rather than being arbitrary as with nominal level data, can be meaningfully sequenced. The number reflects the location of an entity in an ordered sequence. In a marathon race, for example, the number “1” represents the person who was the first to complete the race. Coding pas- sengers by flight class is another example of ordinal level data. For example, first class passengers Designing the Methodology 25 Definition Example Number of Passengers Someone who traveled by air within the previous 6 months. Mr. Jones made one round-trip flight from Dallas to Florida. 1 A person traveling by air to a destination. Mr. Jones flew from L.A. to New York and back again. 2 A person getting on a flight. Mr. Jones flew round trip from Dallas to Orlando, making a connection in Atlanta in both directions. 4 Exhibit 4-1. Definitions of “passenger.” 1 Stevens, S. S. (1951). Mathematics, measurement, and psychophysics. In S. S. Stevens (Ed.) Handbook of Experimental Psychology. New York: Wiley. 2 Ghiselli, E. E., Campbell, J. P., & Zedeck, S. (1981). Measurement Theory for the Behavioral Sciences. San Francisco, CA: W. H. Freedman and Company. Pinot N ominal O rdinal I nterval R atio i l ti Exhibit 4-2. Use the pneumonic “NOIR” to remember nominal, ordinal, interval, and ratio.

26 Airport Passenger-Related Processing Rates Guidebook StronglyStrongly Disagree Disagree Neutral Agree Agree Exhibit 4-3. Ordinal agreement scale. Ordinal Mean Dispersion MeasuresLocation Measures Nominal Ratio Interval Standard Deviation Median RangeMode Measurement Level Exhibit 4-4. Appropriate use of summary statistics by measurement level. might be coded as 3, business class passengers as 2, and coach class passengers as 1. The numbers might correspond to the average ticket price. To use another example, an attitude survey of pas- sengers might employ a standard agreement scale for a set of items as shown in Exhibit 4-3. While the order from left to right reflects increasing agreement, there is no justification that the distances among the labels (the intervals) are equivalent. 4.2.3 Interval These data are measured using a consistent scale (i.e., the interval between the values 1 and 2 is the same as the interval between 22 and 23), but the scale lacks a true zero value. That the value 0° on a Celsius scale is equivalent to 32° on a Fahrenheit scale reflects that the scales are arbitrary in regard to a true zero. As such, it would be incorrect to state that 80° Fahrenheit is twice as hot as 40° Fahrenheit, because there is no absolute zero. To use another example, differences among calendars (e.g., the Jewish calendar, the Chinese calendar, and the Gregorian calendar) reflect differences in starting points. Practically, there is no unambiguous point at which time began, or was equal to 0. 4.2.4 Ratio Ratio level data have the same properties as interval data, but have one additional property: the presence of an absolute zero. Temperature is an often used example to distinguish interval and ratio scales. If you have $4 in your pocket and someone else has $2, you have twice as much money as the other person. This is ratio data and permits one to generate ratios. In the example given about a marathon in the previous section, were the data collected at an ordinal level, you would have noted each person’s name or ID number in the order in which they finished the race. Were you collecting ratio data, you might record the time each runner took to go from the start to the finish line. Someone who took a total of 10 min to complete the race would have taken half as much time, for example, as someone for whom 20 min were needed to run the entire race. Exhibit 4-4 presents common summary statistics that are permissible for each level of measure- ment. Explanations for each of the summary measures are presented in Appendix C of the Guidebook.

4.3 Introduction to Airport Specific Metrics This Guidebook is largely concerned with entities (objects to be counted such as passengers and baggage); resources (such as an agent or ticket kiosk); and processes (a function performed on an entity by a resource). In turn, measurements taken for entities and resources relate to processing rates and pro- cessing times. A processing rate is the number of entities (e.g., passengers, bags, etc.) processed by a single resource (e.g., metal detector, agent, x-ray machine, etc.) in a given unit of time (e.g., minutes and seconds). A processing rate can only be expressed from the perspective of a resource. Processing time is the duration of a transaction or other process for a particular entity and resource. Processing time can be expressed from the perspective of either the entity or the resource; these two times may or may not be equal. Average processing time is a summary sta- tistic of the mean time required for all entities to be processed within a given time period. For processing rate analysis, researchers must distinguish among and define entities, resources, and processes. As an example, Exhibit 4-5 illustrates entities and resources for the passenger check-in process. The two most common entities in the airport terminal environment are passengers and bags. Elements are processed by a resource—something perform- ing a function or process on the entity. Common resources include self-serve check-in kiosks, ticketing agents, metal detectors and x-ray machines, escalators, and others. Typical processes include checking in for a flight and the screening of a passenger or a bag. Failing to operationally define entities, resources, and processes at the start of a research project may result in problems. For example, one might ask if the entity should be defined as a passenger or a travel party. And if the data are collected by party, what definition of party should be used? (Travel parties may be self-defined, while airlines and CBP both have differing definitions.) Should the passenger secu- rity screening function be viewed as one overall process or should it be disaggregated between people and bags? This will be addressed in the following sections. 4.4 Defining Entities in the Airport Terminal Environment As noted previously, it is important to define key units and measurements when collecting data. The two most common entities in an airport terminal are passengers and bags. Each of these is examined below. 4.4.1 Defining Passengers and Travel Parties Depending on the function being analyzed, passengers are often processed at the airport in a cluster or a group, referred to as a travel party. The cost of recording data for both travel party and individuals can be relatively low, while the value of the additional data might be relatively high. In addition, resources usually process entities in parties.3 The question of how best to define a travel party can be complicated because airlines, CBP, and others have specific (and often different) definitions of what Designing the Methodology 27 Entity (Passenger) Entity (Checked Bag) Resource (Agent) THE CHECK-IN PROCESS Exhibit 4-5. Illustration of entity and resource in the check-in process. 3 If gross processing rates are being calculated (i.e., counting the total number of entities simultaneously processed by multi- ple resources over an extended period of time), party size information would be difficult to determine. Regardless of whether processing rates or other results will be presented in terms of travel parties or by individual passengers, we recommend recording data for both travel party and individuals.

constitutes a party. Additionally, the members of the party themselves may have their own, dif- ferent definition of what constitutes a party. The following sections consider how the definition of party differs by process. Defining Passengers and Travel Parties at the Check-in (Ticketing) Counter From the airline’s perspective, a travel party is generally defined as those traveling on one unique Passenger Name Record (PNR). These passengers may or may not be related as members of the same family, and the party can be of any size. Airlines might treat a large group (e.g., school class trip) as one party (i.e., process everyone at a particular counter location), even though members were in separate PNRs. It should be noted, however, that it can be difficult to distin- guish who is a member of a travel party and who is merely a well-wisher. Recommendation. Record both the size of the party as well as the number of bags checked. Defining Passengers and Travel Parties at the Security Screening Checkpoint Typically, passengers are processed individually at a security checkpoint. An exception is babies or children-in-arms. If you choose to record the amount of time it takes passengers to complete the entire checkpoint process, you will also have to account for the impact of party members waiting for others in the party to complete the process before proceeding to the gate. Recommendation. Unless you are monitoring passengers as they go through the entire screen- ing process, collect data on an individual passenger basis. Defining Passengers and Travel Parties at the Federal Inspection Service (International Arrivals) Area At the FIS, a party is defined as persons who both belong to a single family and who live in the same household. Persons who do not belong to the same family but who live in the same house- hold are processed separately. The party size is easy to determine by counting the number of peo- ple simultaneously presenting themselves to a CBP agent during a unique transaction. Recommendation. Define the party size as the number of people simultaneously approach- ing a CBP agent at passport control. By counting the total number of passengers exiting primary inspection in a given period and dividing this by the number of active agents, an overall throughput capacity can be estimated. Collecting this data can serve to validate individual processing times. Defining Passengers and Travel Parties at Other Terminal Functions At a gate, aircraft boarding and deplaning rates are typically expressed in passengers per minute. For this reason, and because it is often difficult to collect travel party information solely through observation when monitoring aircraft boarding and deplaning rates, it is considered to not be critical to collect party size information. At a bag claim device, processing information about passengers is also not expressed in terms of parties. It is difficult, however, to determine whether a person at a claim device is a passenger or a meeter/greeter. Finally, it should be noted that a group of passengers may have their own definition(s) of a travel party which can conflict with those described above. For example, three unrelated busi- ness people traveling together may define themselves as a party of one, two, or three depending on factors such as whether they traveled to the airport together and the degree to which they share the same travel itinerary. 28 Airport Passenger-Related Processing Rates Guidebook

Recommendation. For those processes that do not lend themselves to a formal definition of party size, use your own professional judgment (assuming you cannot interview the party). Dealing with Large Groups Dealing with large parties (typically, those having 10 or more passengers and which receive special processing procedures) is primarily limited to check-in facilities. Procedures for dealing with large parties vary by airline, station, party size, and agent preferences. For example, at the counter of an airline with a hub at the airport, a special counter may be set up for handling large parties, while at a small airport, these parties might be processed along with other passengers. In a recent passenger survey at a large international airport, only about four percent of passengers identified themselves as traveling in parties of 10 or more.4 It is not clear how this would trans- late into the percentage of PNRs with 10 or more passengers. Regardless of party size, in the check-in environment, one agent typically processes all the passengers listed on one PNR. Recommendation. When collecting check-in time data, document not only the size of the party, but also how the airline handled the check-in process for large parties. This practice ultimately will help determine how (or even whether) to include the check-in data for large groups. For airports which frequently experience large parties (e.g., vacation destinations), it is recommended to collect data specifically when large parties are likely to occur and to obtain suf- ficient data to make suitable inferences regarding performance. Collecting Data about Passengers as Entities—Conclusion There is no single best definition of a party that is appropriate across all airport-related processes. It is recommended that you use the standard party definition when observing functions for which a traditional definition is known (for example, at ticketing), and that you use profes- sional judgment in other situations (e.g., at baggage claim). For some functions, party size may not be critical (e.g., aircraft boarding and deplaning rates), and in others, party size information may or may not be needed, depending on the level of analysis being done (e.g., passenger security screening). Regardless of how you define party, clearly document the definition used to avoid ambiguity later. A summary of recommendations for defining parties is shown in Exhibit 4-6. 4.4.2 Defining Bags and Other Items as Entities Belongings that passengers take with them when they travel represent a second major type of entity to be considered. As with varying definitions of party size, the ways in which belongings are defined, classified, and processed also vary depending on the process of interest. Designing the Methodology 29 4 2007. Peak Week Survey Results, Hartsfield-Jackson Atlanta International Airport. Entity Focus Recommendation Airline check-in Use a definition that matches the airline’s PNR. Record number of bags and size of party. Security check-in Count individuals, not parties. Federal Inspection Service Define as the number of people simultaneously approaching a CBP agent at passport control. Other Use professional judgment. Exhibit 4-6. Recommended travel party definitions.

Defining Bags and Other Items at the Airline Check-in (Ticketing) Counter At the check-in counter, a distinction is made between carry-on bags (those brought by the pas- senger into the aircraft cabin) and checked bags (those which are stored in the belly hold of the air- craft). Since the number of bags checked impacts the overall length of a party’s check-in time, it is important to record this information. In fact, many airlines provide “express” kiosks for passen- gers not checking bags. Determining whether a bag is being checked can be made through obser- vation, as a passenger is observed handing a bag to an agent for tagging. (Note that bags might be checked at the gate; as such, do not record this information at another location—for example at baggage claim—if the purpose is to determine how many bags are being checked on a flight.) Recommendation. Record the number of checked bags as well as party size. Document bag- gage information at the source of baggage check-in. Defining Bags and Other Items at the Passenger Security Screening Checkpoint At the security checkpoint, the TSA is concerned with items in carry-on luggage and items on one’s person. The current procedure (at the time of developing the Guidebook) is that all items in a person’s possession are x-rayed while the passenger proceeds through a metal detector. The most common way of defining an item at the checkpoint, and the approach used by the TSA, is to classify it as either a bag or a bin (which can contain several individual items). Using this definition, any item that is laid directly on the x-ray feed belt is considered an item. For example, if a passenger placed a carry-on bag on the feed belt and then placed two bins down (one containing his jacket and shoes and one containing his laptop, its case, and his cell phone), and finally a basket for his loose change, the total number items (bags/bins) would be four. Another option is to define an item as anything laid down by the passenger for inspection, regardless of whether it is contained in a bin. Using the same passenger just described, counting this way would result in a total of seven items. Exhibit 4-7 compares how items on the x-ray belt would be counted and grouped using these two methods. In this case, however, clear definitions of what constitutes an “item” must be provided. For example, is a pair of shoes one item or two? Will loose change be counted as one item or will each coin be counted? By observing the specific, individual items going through the x-ray, it is easier to see the impact of TSA requirements and passenger response on checkpoint capacity as they change over time. For example, the impact of the recent restriction on liquids may not be seen as clearly with the “bags/bins” definition; however, adhering to a practice of routinely counting all items, would result in data that would enable one to analyze and explain a decrease in checkpoint throughput resulting from the additional time needed for passengers to divest a plastic bag prior to security, 30 Airport Passenger-Related Processing Rates Guidebook 1 3 1 3 52 4 6 X-Ray 2 4 7 Counting bags and bins Counting individual items Exhibit 4-7. Two methods of counting x-rayed items at the security checkpoint.

return it to the carry-on following inspection, as well as the increase in time needed for agents to scan these items. Recommendation. We recommend that both operational definitions be used when collect- ing TSA checkpoint x-ray processing rates (i.e., bags/bins and individual items), unless precluded by time or resource constraints. Defining Bags and Other Items at a Federal Inspection Service (International Arrivals) At the FIS, passengers enter the passport control area with only their carry-on luggage. After passport control, they then retrieve their checked bags and proceed to declaration and baggage inspection. If passengers have items that need inspection, these items are typically grouped as customs or agricultural, and agents usually provide separate inspection stations for these two groupings—both employing different inspection processes and equipment. Recommendation. When observing the FIS in the passport control area, use the classifica- tion scheme employed by the U.S. CBP. That is, classify items to be inspected into one, or both of the following: • Customs (typically, checking bags for items and their proper declaration), or • Agricultural (looking for contraband animal/vegetable products). You can estimate the average number of checked bags per party by observing individual par- ties retrieving bags from FIS bag claim carousels. Collecting Data about Bags and Other Entities—Conclusion A summary of recommendations for defining baggage is shown in Exhibit 4-8. 4.5 Defining Resources This section examines how resources might be addressed on a process by process basis. 4.5.1 Defining Resources at Airline Check-in There are several resources in the ticketing process, including self-serve kiosks, agents, scales (to weigh bags), document “spitters,” and electronic readers. Unfortunately (for the researcher Designing the Methodology 31 Entity Focus Recommendation Airlines Define as carry-on or checked. Wait until bags are tagged by an agent before classifying the bag as a carry-on element or one intended to be conveyed in the aircraft belly. Security check-in Each bag counts as one. The contents of each tray may be counted as one entity, or each item contained within the bin as a single entity. FIS Agricultural or customs. Exhibit 4-8. Baggage definitions.

at least), definitions of resources are somewhat blurred in the check-in process, as reflected in the use of flush-mounted kiosks, where a passenger may engage either one of the resources (kiosk or agent) or both of them. Many passengers use online check-in and avoid these resources altogether. Exhibit 4-9 displays two types of kiosks in relation to a traditional counter design. For example, a passenger may check in using the kiosk, ignoring the agent nearby. Or, a pas- senger may step up to the counter, interact directly and exclusively with the agent, but block the kiosk for use by other passengers. Alternatively, the passenger may begin with an agent, who then walks the passenger through the kiosk process, or, conversely, start at the kiosk but end up need- ing agent assistance. In terms of disaggregation, a passenger may check-in at a kiosk and then walk to a designated bag drop area at the counter. As such, the researcher must define the process, a topic considered in Section 4.6. 4.5.2 Defining Resources at the Passenger Security Screening Checkpoint At a gross level, the passenger security checkpoint consists of lanes in which passengers first divest their belongings and place them on a belt for x-ray screening then proceed through a metal detector to be tested for metals. Traditionally, each lane has consisted of one x-ray machine and one walk-through metal detector. More recently, as increased scrutiny of bags has slowed x-ray throughput, some checkpoints have a pair of x-ray machines serving only one metal detector. These two layouts are shown in Exhibit 4-10. To help reduce confusion, it is recommended that a checkpoint lane be defined as an x-ray device, not a walk-through metal detector. Recently, TSA has begun installing explosive detection trace portals to screen people. Passengers stand inside portals while puffs of air are blown over them; dislodged particles are then sucked into the top of the machine and analyzed for explosives. Once clear, the passenger can step out of the device. TSA also uses a manual trace detection process on bags. With this process, the out- side of a bag is swiped and the sample is placed in a machine for analysis. Finally, TSA also uses physical inspection of people and bags; in this instance, the resource is the agent. 4.5.3 Defining Resources at the FIS (International Arrivals) The first resource arriving international passengers encounter is the CBP agent at passport control. These agents are typically positioned in booths with a queuing area in front of them. In 32 Airport Passenger-Related Processing Rates Guidebook “Island” Kiosk Configuration “Traditional” Counter Configuration “Flush-mounted” Kiosk Configuration Exhibit 4-9. Ticket counter layouts.

most instances, booths are assigned to handle either United States’ residents or visitors (who take much longer to process). At busier airports, some booths may be reserved for crew and diplo- mats. For any formal study of FIS processing rates, it is desirable to collect separate observations for residents and visitors. While there is secondary processing of arriving international passen- gers, usually of those whose paperwork is not in order, there will probably not be a need to col- lect observations of these resources due to their infrequency of use. Once clearing passport control, passengers claim their luggage at carousels. These carousels are considered to be resources for presenting bags for passenger claiming. Upon claiming their bags, passengers proceed to their third resource, which is primary cus- toms processing. Depending on the volume of passengers, there are usually one to three agents who scan passengers’ forms. Should passengers have items requiring further inspection, they are directed to a more formal customs and agricultural inspection area. Both of these processes have specific protocols and machinery. 4.6 Defining Processes A fundamental question is whether one should examine a process from the perspective of the entity or the resource. For many processes, the length of the process can differ depending on which perspective is used. To take an example, refer to Exhibit 4-11. In preparing this guidebook, one of the Research Team members calculated processing time in two ways: (1) using a resource perspective and (2) using an entity perspective. As shown, a travel party may spend 3.5 min with an agent and then proceed to the security checkpoint. From the agent’s perspective, however, there may be an additional 0.3 min of processing time required to complete that transaction and before the agent can process another passenger, and beyond that, it may take an addi- tional 0.2 min before the next party at the front of the queue is aware of the available agent and make its way to them. Designing the Methodology 33 X -r ay X -r ay M et al D et . X -r ay X -r ay M et al D et . M et al D et . Exhibit 4-10. Two layouts of x-ray devices and metal detectors. Traditional two-lane arrangement with two x-ray devices and two metal detectors (left) compared to a two-lane arrangement with just one metal detector. At the outset, define whether the perspective of a resource or an entity will be used.

To determine processing rates (i.e., the number of entities that a resource can process in a given unit of time), one must define the processing time from the perspective of the resource. As shown in the exhibit, if one were to calculate processing rates using the processing time from perspective of the entity (passenger), one would erroneously conclude that the processing rate was 13 percent higher. In summary, the importance of documenting how the data are being collected is emphasized. In the following section, the processes and their functional elements most commonly studied at airports are described. 4.6.1 Airline Check-in Related Processing With the advent of new technologies, the airline check-in process has become disaggregated, decentralized, and amorphous—at least from a data collection perspective. (For more informa- tion on this see Section 2.2.6.) For this reason, it is essential that the researcher rigorously define the check-in functions being observed. In the airline check-in area (and at curbside check-in), several passenger transactions are com- mon. In descending order of frequency, these include the following: • Checking in (obtaining a boarding pass); • Checking/dropping a bag; • Rebooking (due to missing a flight, or having a flight canceled); • Upgrading/changing seats; • Purchasing a ticket; and • Obtaining information or an airline pass to go through security. In most instances, when observing check-in activity, only the first two processes (checking in and checking or dropping a bag) are typically defined, most likely because they are the most com- mon and the easiest to distinguish through observation. 4.6.2 Passenger Security Screening Processing As noted previously, there are two broad processes occurring at a security screening checkpoint— the screening of passengers and the screening of their bags. Secondary screening activities include wandings and pat-downs of passengers and hand inspection and trace detection of bags. Most processing rate studies at the checkpoint focus on overall throughput and use passenger counts as they pass through the metal detector as a proxy for the overall capacity of checkpoint, since passengers do not leave a checkpoint without their belongings. 34 Airport Passenger-Related Processing Rates Guidebook 0:00 1:000:30 4:00 4:30 Proc. Time = 4:12 Proc. Rate = 14.3 Pax/Hr Proc. Time = 3:42 Proc. Rate ≠ 16.2 Pax/Hr Agent signals for next passenger Agent signals for next passenger Passenger notices agent’s signal & begins walking to agent Passenger reaches counter Passenger leaves Counter & is free to move to next process Watching Agent Watching Pax Source: Observation made at Washington Dulles, fall 2007. Exhibit 4-11. The processing time from the perspective of the entity and the resource is often different.

4.6.3 FIS (International Arrivals) Processing The key processes at the FIS include passport control and primary customs. Other processes include baggage claim, secondary customs inspection and agricultural inspection. In addition, some passengers must pay a duty. Finally, at some FIS facilities, passengers recheck bags for con- necting flights and often must proceed through a TSA checkpoint to enter the domestic portion of the terminal. 4.6.4 Baggage Processing Processes related to checked bags include their transfer to and from the aircraft and their inspection by TSA. For outbound bags, a skycap or agent prints and applies a “bag tag” and weighs the bag. The agent then places the bag on a take-away belt for screening by TSA and sorting at the airline’s outbound bag make up area. At that point, they are loaded onto bag carts and transferred to the aircraft, where they are then placed in the plane’s belly hold. Upon a flight’s arrival, checked bags are offloaded, placed on bag carts and driven to the ter- minal to be placed on bag claim devices for retrieval by passengers. Designing the Methodology 35