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.

14 Chapter 3 identifies roles, relationships, and responsibilities of stakeholders. It examines principal steps involved in planning an airport passenger-rate data collection effort. It begins with the ques- tion of whether the potential benefits of the proposed effort outweigh the anticipated cost; describes different types of research (i.e., exploratory, descriptive, inferential); summarizes the questions each type addresses; and notes the ends to which the data might be used. 3.1 Roles and Responsibilities When an airport data collection event is first mentioned, it invariably raises numerous ques- tions: Who is asking for the data? How will it be used? What’s the budget? What’s the schedule? What kind of resources can be made available? Without answers to these fundamental questions, the success of your research is in jeopardy. This section will help the researcher establish the role of key stakeholders and their interrelationships within the team. Many entities can sponsor a data collection study, including airports, airlines, manufacturers, and various agencies. Likewise, there are many ways of managing and staffing the event and pro- moting involvement with stakeholders. There are therefore myriad ways of organizing a study. Exhibit 3-1 is an example of how a study could be arranged with the airport as the sponsor. 3.1.1 Client/Sponsor For airports, oversight is guided by a board, commission, or an authority consisting of appointed or elected officials. While these agencies typically provide oversight to airport man- agement and approve long-term plans and large capital expenditures, usually it is the airport director or manager who makes day-to-day decisions. Depending on the size of the airport, there may be several departments, each having its own manager. In such cases, passenger terminal-related studies would typically fall within the purview of the planning and/or engineering department and would be managed by its director. Regardless of the affiliation of the project sponsor(s), it is essential that the following ques- tions be answered clearly and unambiguously as they pertain to the sponsor at the beginning of any project: • Who has primary responsibility for defining the questions the study is intended to address? • What preference does this person or group have regarding ongoing involvement with the project? – What information would they like to receive, in what format, and with what frequency? – Who should be the principal point-of-contact (POC) on the client’s side for questions that might emerge related to the study’s focus, direction, etc.? C H A P T E R 3 Defining the Research: Purpose, Focus, and Potential Uses

Defining the Research: Purpose, Focus, and Potential Uses 15 • Who is the designated project manager, and what information would he or she like to receive, in what format, and with what frequency? • If the person given responsibility for day-to-day issues pertaining to access, authorizations, etc. is different from the project manager, who is that person, and what is the scope of issues he or she is authorized to address? • If problems or obstacles arise in implementing the study, and the project manager is not able or authorized to resolve them, what is the chain of persons through which the issue should be escalated? 3.1.2 Study Team The size of the study team will depend on the team’s depth and organization, and the size, duration, and complexity of the study itself. For a typical medium- to large-scale study, the roles listed in the following sections are the most typical. Multiple roles might be assumed by a single person or distributed across multiple persons. Titles vary as well, but the functions are largely universal. Project Manager The project manager is typically a mid-level to senior person who has the long-term, day-to- day relationship with his or her client counterpart. The need for the passenger-related process- ing rate study may initially originate from discussions between the project manager and those within the airport or airline. Survey Manager The survey manager is usually a mid-level staff person. His/her role on the project would be to oversee the day-to-day management of the data processing rate study, including leading the development of the scope, schedule, and budget; developing the team; and assigning roles and responsibilities. The survey manager would have the responsibility of ensuring the survey goals were adequately defined and met. Decision Maker Survey Manager Admin. Support Staffing Source (e.g., airport personnel, mkt. research firm) Surveyor Surveyor Surveyor Sponsor/Client (Airport) (Large Airport: Dir./Mgr.) Project Manager (Large Airport: Dir. Planning/Eng.) (Small Airport: Apt. Mgr.) Project Manager (Typ. oversees multiple tasks of which survey is but one part) Study Team (Typically, Consultant) Statistical Technical Expert Survey Assistant Data Analyst IT Analyst Other Stakeholders • Airlines • Agencies • Concessionaires Exhibit 3-1. Typical sponsor and study team roles (assuming an airport is the sponsor).

16 Airport Passenger-Related Processing Rates Guidebook Research and Statistical Expert A person(s) with expertise in research methodology and quantitative/statistical analysis should be consulted to develop, or provide comments and recommendations about, the overall methodology, the sampling plan, and so forth. Most of this person’s input would occur at the project’s initiation. A distinction is sometimes drawn in the consulting literature among differ- ent approaches to consulting. One such approach, generally referred to as process consultation might be of particular appeal.1 When acting in this role, the consultant not only provides tech- nical expertise related to the specific project, but also works with the client to develop expertise. This arrangement has the goal of, over time, reducing the reliance on the consultant. Survey Assistant The survey assistant has primary responsibility for assisting the survey project manager and secondarily to assist others on the project team throughout the duration of the study. Typically, this staff person will be at a junior level. The degree of assistance this person can provide is based on his/her level of education and current skill sets. Data Analyst The data analyst should not only be well-versed in technical analysis, but should also have a strong familiarity with the airport terminal environment. This person could be a terminal or air- port planner or aviation architect. The analyst is often largely responsible for documenting results, and responsibilities might extend to presenting findings to the client. Administrative Support Data collection efforts are inherently complex and, as such, often require a significant level of coordination and administration. The staff person serving this function would be responsible for such things as making travel plans, scheduling visits to the airport’s security office, buying supplies, shipping and receiving materials, scheduling meetings, preparing invoices and con- tracts, and editing/proofing the report. Data Collection Staff For small studies (e.g., small airports where only a few functional elements are being observed for a limited time period), airport/airline or consultant staffing may be used. For larger studies, typically examining multiple functional elements of a medium or large airport over a multi-day period, a market-research firm is frequently employed. The data collection staff reports directly to the survey manager. 3.2 Is the Study Needed? While the need for data collection is often justifiable, the benefit of validating the need, and avoiding what might be a costly, and possibly unjustified, effort well exceeds the relatively minor cost of pausing to consider a few basic questions (see Appendix C for more information). Exhibit 3-2 illustrates these questions. 3.3 Research Fundamentals This section summarizes a number of fundamental issues and terms related to the research process. (Additional detail is included in Appendix C.) 1 Schein, E. H. (1999). Process Consultation Revisited: Building the Helping Relationship. NY: Addison Wesley.

Research is a dynamic process with both deductive and inductive dimensions. This differs in some ways from what some present as the “traditional” approach to research, i.e., that theory drives hypothesis testing. Sometimes it does, but sometimes it doesn’t work this way. 3.3.1 Theory, Hypotheses, and Evidence The word “theory” often implies a formal set of laws, propositions, variables, and the like, whose relationships are clearly defined. A related implication is that theory may not be particu- larly germane to the everyday world of work. This view of theory is not incorrect, but neither is it complete. While theory can be abstract and complex in its detail, it does not necessarily have to be abstract, complex, or formal. It can be thought of more broadly and simply as an explanation of “how the world works.” For exam- ple, an organization might develop a mission or a value statement (or both); engrave the words in a medium intended to last millennia; and prominently display the statement in the workplace with the intent of communicating to all its perspective clients on issues pertinent to its view. In Defining the Research: Purpose, Focus, and Potential Uses 17 Question Things to Consider Have relevant data been collected at this airport in the past that might be used rather than collecting new data? Might you be able to get data from another airport similar in key ways to this airport? Are there data available that might help answer the research question? Might access to the data be blocked due to proprietary or security issues? Sometimes the data are perceived to be so sensitive that the “owner” of the data may not give permission to share it. Has the decision already been made, and the data are being collected to legitimize the decision? Is there anything to suggest that the study is an attempt to “prove” something true or false? What role will the results play in the decision being considered? To what extent will the decision makers be persuaded by the results? What will the decision makers accept as credible evidence? Before collecting data, make certain that the research plan will result in data that the sponsors will accept. It is better to learn beforehand, for example, that the proposed sampling plan does not meet the sponsor’s criteria for rigor. What is the cost of the potential investment that the data will help inform? What is the cost of conducting the research? Does the benefit equal or outweigh the cost? Cost should be considered not only in economic terms, but as safety, inconvenience, and so forth. Exhibit 3-2. Considerations to determine need for data collection.

2008, British Airways announced a new venture: OpenSkies. The “theory” OpenSkies used to define its clients is reflected in its advertising as shown in Exhibit 3-3. So, how does this relate to airport processing rate studies? It relates in the following two ways: 1. The published research literature may well contain formal theories relevant to what data to collect and how to collect it. For example, Appendix B includes a bibliography of recent research articles related to passenger and baggage processing in airports. It is intended to illustrate the scope and diversity of research available on a given topic. Before embarking on an investigation, review the literature to see how it might enhance the quality of the planned research. The Internet provides access to numerous sources for such scholarly documents. 2. Informally, the key decisions about how to go about collecting data are grounded in assump- tions about how things work (i.e., one’s own theory). For example, you might choose to col- lect passenger security screening data between 6:00 a.m. and 8:00 a.m. on a Monday because your experience is that this time period reflects peak checkpoint activity. While this “theory” may be correct in some circumstances, it may also be wrong in others. For example, at many vacation-oriented airports, the peak at the checkpoint occurs in the late morning due to check-out times at hotels. Another common view of research is of the stereotypical scientist, objectively testing hypothe- ses (or an “educated guess”) arising from theory. Exhibit 3-4 reflects this general approach to research. This is certainly one way in which research proceeds, but, similar to theory, it is not the only way. Before considering an “evidence first” approach, we wish to mention a variation on the tra- ditional approach displayed in Exhibit 3-4 that has been gaining dominance in recent years. In particular, this is a confidence interval (CI) approach rather than a hypothesis driven approach. In a hypothesis driven approach, the researcher’s primary interest is in testing a population parameter, and uses a sample drawn from the population. When the researcher takes a CI approach, the intent is to calculate an interval within which the population parameter is likely 18 Airport Passenger-Related Processing Rates Guidebook Exhibit 3-3. OpenSkies advertisement. Question key assumptions, even if they seem to be “common sense,” by checking with informants, look- ing at the literature, etc.

to fall. Hypotheses are stated before data collection; CIs are calculated after data are collected.2 In conducting passenger-processing rate research in airport environments, the CI approach is going to be the most appropriate in most instances. A markedly different approach to those described above is shown in Exhibit 3-5. In contrast to beginning with a theory and then collecting evidence to test the theory or estimate a popula- tion parameter within some CI, this approach begins with evidence for which one seeks poten- tial explanations, or “theories” to explain the evidence. This approach is subsumed under the broad heading of Bayesian Law, so named after the 18th Century English clergyman, Thomas Bayes, credited with developing the approach. Depending on where one begins can result in potentially dramatic conclusions (see Exhibit 3-6). This is important because limiting oneself to a particular perspective of how research should be conducted and how data ought to be gathered may impose unnecessary constraints. What is important is that the research is executed systematically and with rigor. The documented ways in which science proceeds are often idealized: portraying what is inherently a very dynamic and nonlinear process as logical and linear. 3.3.2 Research Questions and Purposes A basic issue in research is specifying the question the research will help answer. Penning a specific question also helps in determining what approach might be best used in seeking an Defining the Research: Purpose, Focus, and Potential Uses 19 Theory Drives questions & hypotheses Hypothesis: Installing n kiosks will reduce the average time of passengers waiting in line by 10% over check-in agents. Leading to a conclusion Drives data collection Followed by analysis Exhibit 3-4. Hypothesis driven approach. Evidence leads to speculation about possible explanations Which may or may not drive more data collection & analysis Theory Exhibit 3-5. Bayesian approach. 2 While these approaches are presented here as mutually exclusive, they might be integrated in practice.

answer. One classic text in research methodology5 suggests that a research question should express a relationship between two or more variables, and it should imply an empirical approach, that is, it should lend itself to being measured using data. A variable is, not surprisingly, some- thing that can vary, or assume different values. In the next section, illustrative questions are given, categorized by the purpose of research with which they are best matched. The five research purposes are presented as the following: 1. Explore, 2. Describe, 3. Test, 4. Evaluate, and 5. Predict. The distinctions among these purposes are not absolute, nor are they necessarily exclusive of one another. A research initiative might be directed at answering questions with multiple pur- poses. Indeed, this is but one of many ways of classifying research. In addition, the reader whose practice lies primarily in the arena of modeling and simulation might note their absence from this list. Although modeling and simulation applications require input data, for example, to gen- erate distributions and parameters for use as stochastic varieties in modeling, the techniques used to collect data are largely independent of specific applications (such as simulation and model- ing). Those issues unique to modeling are beyond the scope of this guidebook. Explore (Exploratory Research) Exploratory research is sometimes defined as “what to do when you don’t know what you don’t know.” Its aim is discovery and to develop an understanding of relevant variables and their interactions in a real (field) environment. Exploratory research, as such, is appropriate when the 20 Airport Passenger-Related Processing Rates Guidebook If your intent is to… And take action based on… Use… Example Test a hypothesis regarding a population parameter Whether you reject or fail to reject the null hypothesis Hypothesis testing approach The proportion of coach passengers checking in more than 60 min prior to scheduled departure is 80% H A : p > .80 3 H 0 : p .804 Estimate a population parameter The confidence interval selected CI approach Plus or minus 5%, what is the average time coach passengers check in prior to scheduled departure? Determine the likelihood of an event given some evidence The calculated probability Bayesian approach What is the probability that a passenger’s carry on- luggage will be subject to secondary security screening given that the passenger is boarding an international flight? Exhibit 3-6. Research approaches. 3 This is the research, or Alternative, hypothesis. It reads: The proportion is greater than 80%. 4 This is the null hypothesis. It is what is tested, and reads: The proportion is less than or equal to 80%. 5 Kerlinger F. & Lee, H. (2000). Foundations of Behavioral Research, 4th ed. NY: Harcourt Brace.

problem is not well defined. For example, passenger complaints about signs within a facility might prompt the following exploratory question: • “Where should signage be located to minimize passenger confusion?” As another example, if a new security checkpoint configuration is proposed, it may be too novel to rely on variables used in other studies. The question, therefore, might then be the following: • “How does a given alternative security checkpoint configuration affect capacity?” This type of research is often qualitative rather than quantitative. That is, it employs verbal descriptors of observations, rather than counts of those observations (see Appendix C for more information). Describe (Descriptive Research) Descriptive research, as the name implies, is intended to describe phenomena. While descrip- tive research might involve collecting qualitative data by asking open-ended questions in an interview, it typically employs quantitative methods resulting in reporting frequencies, calculat- ing averages, and the like. The following two questions illustrate the nature of descriptive research. Each implies that the relevant variables have been identified as well as the conditions under which the data should be collected. • “What is the average number of passengers departing on international flights on weekday evenings in July at a given airport?” • “How many men use a given restroom at a particular location at a given time?” Test (Experimental and Quasi-experimental Research and Modeling) Often, the intent of the research is not simply to describe something, but to test the impact of some intervention. In an airport environment, such research might be initiated to evaluate the relative effectiveness of a security screening technology in accurately detecting contraband. It is similar in approach to research conducted to assess the relative effectiveness of an experimental drug in comparison to a control (placebo) or another drug. Variables are often manipulated and controlled. This research lies largely outside the scope of this guidebook and, as such, will not receive much attention. Examples of questions that might be asked in this type of research include the following: • “What is the impact of posting airline personnel near check-in waiting lines on the average passenger waiting time?” In addition to the classic “experiment,” simulation modeling might be used, employing rep- resentative data to help answer questions such as the following: • “What would be the impact on processing time of a new security measure being considered?” • “How many agents are needed to keep passenger waiting time below an average of 10 min?” Evaluate (Evaluative Research) Sometimes, the intent of the research is to assess performance against some standard or stated requirement. Basically, evaluation research is concerned with seeing how well something is work- ing, with an eye toward improving performance, as illustrated by the following two questions: • “Is the performance of a given piece of equipment in the field consistent with manufacturer’s specifications?” • “On average, what proportion of passengers waits in a security checkpoint line longer than the 10-minute maximum threshold specified by an airline?” Defining the Research: Purpose, Focus, and Potential Uses 21

Predict Finally, research might be initiated to attempt to predict or anticipate potential emerging pat- terns before they occur. This is related to environmental scanning, insofar as it represents a delib- erate attempt to monitor potential trends and their impact. For example, in the early 1970s, one might have posed the following question: • “What would be the impact of an increase in the number of women in the workforce on air- port design?” There are numerous documented approaches to answering questions such as these. While well beyond the scope of this guidebook, here is one as illustrative: scenario planning. This method involves convening persons with relevant expertise to identify those areas that might most impact the industry (e.g., regulation, fuel costs, demographic changes), and then to systemati- cally consider what the best, worst, and might likely scenarios might be. The principal value of such an approach is that it facilitates deliberate consideration of future trends, and in so doing, presumably leaves people better prepared. When the goal of the research is to predict, data from multiple sources might be sought. The scenario planning example relies, to an extent, on the judgments of experts. Probabilities can also be drawn from historical data to help identify patterns and trends. Exhibit 3-7 is a summary of the key characteristics of each research type. 3.4 Developing the Research Plan Large research studies, particularly when funding is being requested, often require the researchers to adhere to a specific set of technical requirements. The Research Team is aware that the ad hoc and short timeline of many airport-planning research efforts makes developing a “for- mal” research plan impracticable. Nonetheless, even though you might not have the “luxury” of 22 Airport Passenger-Related Processing Rates Guidebook Research Purpose Characteristics Explore Primary purpose: to better define or understand a situation. Data will help answer the research question. The benefit of conducting the research justifies the cost. Qualitative data are recorded, using observation. Describe Primary purpose: to provide descriptive information about something. Test Primary purpose: to assess the impact of a proposed change in procedure or policy. Evaluate Primary purpose: to assess performance against requirements. Predict Primary purpose: to consider possible future circumstances with the purpose of being better prepared for emerging trends. Exhibit 3-7. Summary of research types.

developing such a plan, there are benefits to considering the issues described in this section, as well as documenting basic information. The following are the three major elements the Research Team believes worth documenting, regardless of the size of the research endeavor.6 1. Goals or aims. 2. Background and significance. 3. Research design and methods. Each is described in the sections that follow. 3.4.1 Goals or Aims Specify the question the research is intended to help answer or the specific purpose of the research. The experience of having to translate an intended purpose into words can help clarify your intent. In addition, a written statement can serve as a way of ensuring that your understand- ing of the purpose of the research is consistent with that of the sponsor and other stakeholders. Two examples follow: Statement of Purpose—Example 1 The purpose of this study is to aid decision makers in determining if extending the dwell time of the airport’s automated guideway transit system (AGTS) vehicles from 30 sec to 35 sec at the Concourse C station might improve overall system capacity by providing more boarding time for passengers. Statement of Purpose—Example 2 The goal of this study is to provide airport management with recent data showing the percent- age of arriving flights whose first checked bag reaches the claim device within the airport’s goal of 15 min. 3.4.2 Background and Significance Document what is already known, and specify how the proposed research initiative will add to this knowledge. Consider a “devil’s advocate” perspective by asking what the consequences of not doing the research might be. 3.4.3 Research Design and Methods In this section, describe how you will go about collecting and analyzing data. Additional infor- mation about these issues, including sampling strategies and sample size, is presented in Chapter 5 and in Appendix C. The research plan does not need be lengthy. It should, however, capture key information that, were it not documented and those familiar with the research were not available, would be diffi- cult to ascertain. Defining the Research: Purpose, Focus, and Potential Uses 23 6 This section is partly based on guidelines published by the Agency for Healthcare Research and Quality, Department of Health and Human Services. http://www.ahrq.gov/fund/esstplan.htm.