B

Effects of Airplane Seat Dimensions on Egress: Civil Aerospace Medical Institute Submission to the Federal Aviation Administration Institutional Review Board

Principal Investigator: David Weed

Collaborating Investigators: Melissa Beben, David Ruppel, Kelly Guinn

PURPOSE AND OBJECTIVE

This research project was requested by AIR-600 to meet the legislative requirements established in Public Law 115-254: FAA Reauthorization Act of 2018, Section 577; Minimum Dimensions for Passenger Seats. Specifically, this project will be investigating seat pitch, as defined by the distance between a point on one seat, to the same point on the seat forward of it, and seat width, as measured by the distance between the seat armrests, to determine what effects they have on passenger egress from a narrow-body transport-category airplane. The specific influence of these two factors on airplane egress has not been studied as their own factors in previous egress research projects. The results of this project, which will consider the effect of the above factors both individually, and within the context of the airplane evacuation event as a whole, will inform the rulemaking on minimum dimensions for passenger safety, as required by legislation.

DESCRIPTION OF THE RESEARCH

Hypothesis

Aircraft evacuation should be considered as a system. Currently, the bottleneck within that system is the egress rates of the doors combined with

evacuation slides/assist means. As long as ergonomic minimums at the passenger seats are respected—that is, people can get into and out of the seat—the rate at which the passengers can move from their seats into the main aisle is ultimately immaterial to the evacuation flow as a whole. Manipulation of the usable egress space between seats may provide a statistically significant difference in evacuation speed, but the practical significance of this difference will be determined by the consideration of overall egress speeds and the expected flow rates of the door/slide evacuation system.

Research Question 1

What percentage of the population (based on body type) would a smaller seat pitch/width constitute a violation of the ergonomic minimums and a safety risk?

Research Question 2

Does egress time (DV₁:DV₂) differ as a function of seat pitch (V₁) and seat width (V₂)?

Variable Breakdown

DV₁: Individual participant egress times: measured as the period lasting from the timestamp at which one subject is completely through the exit opening until the time the next subject is completely through the opening.

DV₂: Total group egress times: measured as the timestamp of when the buzzer sounds to start the evacuation, to the timestamp at which the last subject is completely through the exit opening.

IV₁: Pitch (A) (4 levels):

A₀: 32 in (control)

A₁: 30 in

A₂: 28 in

A₃: 34 in

IV₂: Seat width (B) (2 levels): as defined by the distance measured between the inside faces of the armrests while fully deployed.

B₀: 18 inches (control)

B₁: 16 inches

HISTORICAL BACKGROUND

The number, placement, and size of the doors on the aircraft determine the maximum capacity of passengers that can be in an airplane cabin (14 CFR 25.807). This passenger allowance per door pair is based on the designated flow rates of the doors, which correspond to how many passengers can be expected to utilize a door and attached egress assist means to safely reach the ground in 90 seconds. These numbers have been substantiated throughout the years as each aircraft type has had to successfully demonstrate that, at a maximum passenger load based on the exits, the aircraft can be fully evacuated in 90 seconds with half the exits blocked in dark-of-night conditions (Appendix J to 14 CFR part 25). The data from these evacuation demonstrations are proprietary and belong to the aircraft manufacturers, but the Cabin Safety Research Team has been allowed to observe historical demonstration videos as well as directly observe these demonstrations on some occasions. Results of these demonstrations, the overall certification process, and changes to aircraft interiors proposed by the manufacturers are what have historically influenced the focus of evacuation research at CAMI and in other facilities around the world. This research has mostly focused on the access to, and immediately around, the exits in an effort to maximize the efficiency of evacuations, as well as aspects of the evacuation research methodology itself to ensure the research scenarios were as close as possible to real-world events while reasonably maintaining scientific rigor and participant safety. McLean et al. (2001) provide an analysis of the effort just surrounding the Type-III overwing exit over a span of approximately a decade.

These data provide a roadmap of the series of events that occur during an emergency evacuation of an airplane. First, an event occurs, followed (usually) by the call to evacuate from the cabin crew as they open the doors. At this call, the passengers stand and start to file into the aisle leading toward an exit of their choice. This aisle, which is minimally required to be 20 inches wide, cannot hold the entirety of the airplane cabin load, so miniature queues form in each seat row. Additionally, the exits take time to prepare and become functional, including time to deploy each slide (maximum of 10 seconds per regulation) and for the cabin crew to verify they are usable before passengers start to egress from the airplane. This in turn causes the main aisle to form a queue, which precludes more passengers from moving into the aisle than those already in aisle seats. Once the evacuation properly begins, with passengers able to leave the airplane, the evacuation moves relatively quickly, with people replenishing the queue in the aisle as openings appear, until the evacuation finishes. It is primarily this observation, the formation of the queues, and flow rates of the doors that ultimately form the hypothesis that, as long as ergonomic minimums are

maintained, the seat pitch of the aircraft will not be a practical significant factor in an evacuation. There may be statistically significantly faster evacuations with larger seat pitch, but the door flow rates are still the same; it should not matter if configuration A can get people to the exit faster than configuration B, because the evacuation system (door and slide) can only accommodate so many people. This project will be utilizing a simulated Type-I exit, with an opening 32 inches wide by 72 inches tall. The Type-I exit on transport category airplanes are rated for 45 passengers, which, assuming an 80-second time frame, gives an expected egress rate of 1.77 seconds per passenger for comparison to the egress rates achieved in this project. The Type-I exit configuration was chosen to represent the doors found on Boeing 737/Airbus A320 airplanes, as these are the aircraft with the most restrictive economy-class sections, as well as to facilitate comparison to previous egress research (see the “Egress Assist Means” section).

This research project builds upon a foundation of aircraft egress research, that, while not focusing specifically on the effects of seat pitch and width on evacuations, has provided a significant trove of lessons learned and best practices applicable to this project. Specifically, that historical background has already established means of replicating, in a laboratory setting, an approximation of a “real-world” evacuation while maintaining both scientific and safety control of the event being observed.

FLIGHT ATTENDANT ASSISTANCE

This project will request the use of current flight attendants and former flight attendants currently working as FAA Cabin Safety Inspectors to assist in the evacuation trials.

Flight attendants are highly trained safety professionals who have accepted a role of responsibility and influence should an emergency event occur on an airplane. Previous research has investigated the effects flight attendants have on evacuations and how much flight attendant assertiveness plays a role in facilitating a rapid egress event (Muir and Cobbett 1995; McLean et al. 2002). This project will request volunteer flight attendants to be present and assist with this evacuation research as they would assist during a real aircraft evacuation. The research team has reached out to two airlines based in Texas and received positive feedback for supplying volunteer flight attendants for this project. The flight attendants working for this project will receive a pre-test briefing session to describe the tests to be performed and their role in them, as well as a standardized set of commands to use during the evacuation trials. All flight attendants who facilitate the evacuation trials will be considered on duty during testing days and covered by their companies’ insurance in the unlikely event they

sustain injury during the evacuation trials. All former flight attendants currently working as FAA Cabin Safety Inspectors who assist with this project will be on duty during trial days as well.

MOTIVATION

This project will manipulate participant motivation during the evacuation trials by utilizing a competitive compensation approach.

Research projects manipulating the motivation of participants with extra payments have shown improved performance during egress events thought to be closer to real-world scenarios compared to trials with no extra compensation available (Muir et al. 1989). Muir further refined this motivation manipulation by comparing a competitive approach (the first x number of participants out of the exit get more money) to a cooperative approach (if the group as a whole gets out in 90 seconds, everyone gets more money) and concluded that the cooperative approach would resemble a precautionary or certification evacuation while the competitive approach could be seen as closer to an evacuation during an emergency (Muir and Cobbett 1995). Also of note is that the cooperative approach provided the fastest overall egress times, while the competitive approach was faster than no compensation. This improvement was attributed to increased efficiency in the aisles and at the exits, with the participants working together instead of against each other.

The research project Access to Egress (McLean et al. 2001; Corbett et al. 2003; McLean and Corbett 2004) also utilized a competitive approach to influencing participant motivation during evacuation trials. This research project had reported injuries ranging from bruises to a broken leg, but the overall occurrence of injuries was low: of the 2,544 participants in the project, 58 (2%) received injuries, with 11 of the injuries deemed serious (Corbett et al. 2003). This project also noted that studies and events involving aircraft evacuation are chaotic, with higher-than-normal potential for injury; restricted operating areas and passageways where people intersect can further compound this potential, such as at and around the Type-III exit. The report goes on, however, to identify that the occurrence of injuries in the low-motivation group (no compensation for egress speed) was abbreviated due to alterations in the test, specifically the alterations of the commands given by the flight attendants at the exits. The paper also describes the safety precautions taken to mitigate the risks during evacuation research that this current project is expanding upon.

The current research project proposes to use a competitive approach to achieve a high-motivation evacuation test environment. Based on feedback and recommendations from the IRB, this project will offer an incentive

similar to early research by Muir, in which the first 70% of participants out the exit of each trial will receive a bonus equal to a quarter of the day’s total possible pay. This should provide an increase in competitive evacuation behavior during the evacuation trials while also enabling, with assigned seating for each trial, approximately 99% of the participants to receive some kind of bonus during the project. This project will include safety monitors within and outside the simulator with the authority to stop the trial should someone receive an injury in the egress path and be unable to remove themselves without causing more harm to themselves or others. The simulator will also be outfitted with egress ramps with pads adjacent to the ramps to mitigate any falls outside of the simulator. The door that will be used for this project is 72 inches tall by 32 inches wide, simulating an airplane Type-I exit, leading to a ramp outside of the simulator. The FlexSim is equipped with a first aid kit, and this project will seek cooperation with the CAMI clinic and/or the OKC ARFF to have medical first responders present during the evacuation trials.

Risk Assessment and Reasoning

The use of a high-motivation condition for airplane evacuation has been studied multiple times across different research institutions, including CAMI. The Access-to-Egress project is the project that easily comes to mind at CAMI when thinking about the risks associated with evacuation research in general, and with high motivation research specifically. As noted in Corbett, McLean, and Whinnery (2003), the high-motivation condition accrued the most injuries, when considering both major and minor injuries together, though the major injuries were more evenly spread across both motivation conditions. One point to consider is that the major injuries primarily centered around the use of the Type-III exit, which requires a person to navigate a step-up and step-down to traverse. Type-III exits are outside the scope of the current proposal and will not be used. This research project is focused on the effect of aircraft cabin seating dimensions and how they may influence an evacuation, as a whole. This project has chosen to use a floor-level, Type-I exit, which is more representative of the majority of aircraft doors on narrow-body airplanes. Another factor in Access-to-Egress was, during the evacuation trials, especially in the high-motivation groups, people made their own exit paths by pushing seat backs down and walking across them. This caused further congestion and conflict close to the exits (McLean and Corbett 2004). A recurrence of this risk will be prevented by two factors. First, the current project will lock out the reclining capability of the simulated aircraft seating during the test, since seat pitch and legroom are of primary focus in this study. Second, the simulated aircraft seating in

our new FlexSim (testing facility) lack the ability to break forward, which was not the case for the repurposed aircraft seats in Access-to-Egress. This eliminates another potential area for injury risk and conflict by limiting participants to the paths we want them to take. Finally, Access-to-Egress was the first evacuation project, which used the Type-III exit, and reported injuries in exacting detail, to the point that injury analysis and descriptions constituted half the subject matter of the publication (Corbett et al. 2003). Previous evacuation research projects are either vague or silent on the numbers and types of injuries experienced during their studies. If injuries are mentioned at all in past evacuation research, it is usually in the context of describing the effect they may have had on data collection (trials stopped). Therefore, we have no available comparator other than Access-to-Egress to conceptualize the frequency and severity of injuries typically expected for a smaller study not using a Type-III exit. Our assumption is that participants in a high-motivation study may be more likely to encounter minor injuries than in regular, day-to-day activities. However, this current project proposal does not include the use of a Type-III exit or break-over seats, which were determined to be primary contributors to the major injuries seen in Access-to-Egress.

Egress Assist Means

This project will use ramps with handrails and surrounding padding at the egress doors for participant disembarkation.

The egress portal of the aircraft in an evacuation scenario must be considered as a system that contains both the physical door, the size of the opening, and the egress assist means (slide, stairs, and so on). Each pair of doors on an aircraft allows a number of passengers (passenger credit) to be on the aircraft based on the efficacy of the door and evacuation assist means to disembark passengers (flow rate). Previous research has shown that evacuation rates for doors with slides were lower than those of doors with ramps (McLean et al. 1996, 1999). However, while evacuations using ramps are faster, they are also significantly safer than using aircraft evacuation slides in research and real-world applications.

This study will not be using airplane evacuation slides. This is due to safety considerations and to remove the slide as an unnecessary confounding variable for this test. The normal course of an evacuation is well documented: An event occurs, passengers stand to the best of their ability, and those with aisle seats inhabit the aisle and a queue forms as people wait for an exit portal to become available, this queue is replenished by those sitting in the middle and outboard seats until the airplane is empty. By not using a slide, this study will eliminate the variable hesitation at the doorway

that has been documented by previous research, and allow a better look at how much the seat dimensions influence the participants’ ability to move from a sitting position to the main aisle and toward the exit portal. Post-test analysis will compare the data from this project to the data from the previous ramp versus slide studies, to determine the practical, in addition to the statistical, differences in egress times caused by the manipulation of seat dimensions. A discussion of practical significance will center on an evaluation of the ability of the cabin seat configuration to deliver passengers to the exits; we know that the door/ramp combination is already faster than a door/slide. We also know what flow rate a Type-I sized door/slide combination can accommodate (1.77 seconds per person, as discussed previously, or approximately .8 seconds per person according to the 1999 study).

This study is designed to pair an evaluation of each experimental seat pitch with a control pitch that represents the average current seat pitch in use. This should allow both a comparison of control evacuation rates with expected egress rates for the same size door with slides, as well as comparison of changes between the control pitch versus the experimental pitch scenarios. The ramps to be purchased for this project are approximately 10 feet long, by 36 inches wide, with a 12 degree rise over the run. The pads to be placed to the sides of the ramps are 1 foot thick, by 6 foot wide, by 12 foot long.

EXPERIMENTAL METHOD

Equipment and Facilities

This project will require the use of the MMAC visitor center room for participant arrival and initial check in. This project will require the use of CAMI rooms 117 and 127 for participant briefing, seating, testing, and demographic information collection. This project will use the Flexible Aircraft Cabin Simulator (FlexSim), a modular narrow body aircraft simulator designed from the ground up for airplane cabin evacuation research. This project will use large open-sided tents for participant gathering between trials. This project will require usage of portable lavatory facilities to be located west of the FlexSim for human comfort. As this project has progressed to the point it will be performed during the winter months, this project will require additional materials including propane heaters and tent side panels for subject comfort between trial runs.

Transportation will be provided to transport participants from the MMAC visitor center to the CAMI building. This project will use numbered vests for participant identification. This project will use calibrated scales to gather participant weight and height, tailor’s measuring tapes to measure participant’s girth around the waist, an anthropometer to measure

participants shoulder width, sitting buttocks-to-knee length, and sitting hip width, and a meter stick to measure participant sitting floor to knee length. This project will use an aircraft-seating mock-up in room 127 for evaluation of research question #1.

Study Population

This project will use 12 groups of 60 naive participants each. Each participant group will be comprised of approximately half females and half males, and range in age from 18 to 60 years, and no more than 40% of the participants in any given group shall fall into a single decade category. For example, using age groups 18–30, 30–40, 40–50, and 50–60, no decade group shall contain more than 40% of the total subjects. The contractor will screen the participants to ensure they are in good physical and mental condition, and have no other conditions that would endanger themselves or others by their participation in the project and that they are representative of the flying public. CAMI personnel will screen participants the day of the trials to verify participant naiveté toward evacuation research and that participants are in a proper condition to participate in the day’s trials. Participants will be required to read, speak, and understand written and spoken English. They will arrive at CAMI wearing long sleeved shirts, long pants, and shoes that are firmly affixed to their feet (no sandals or slip-on shoes or high-heeled shoes). Participants will be informed by the contractor before arriving at the MMAC that they must not bring any extraneous items with them on the day of testing, including cell phones, purses, knives, bags, books, or anything other than a billfold containing their identification and the keys to their transportation. The participants will be provided by a contractor.

Method

On day of testing, participants will arrive at the MMAC visitor center. The participants will pass through the security screening checkpoint to ensure they are not bringing unnecessary items with them. The contractor providing participants for this study will check in participants. Upon arrival, participants will receive a clipboard with a copy of the initial participant information form (Appendix C) and a copy of the informed consent form (Appendix B), with instructions to fill out the first form and wait until the briefing to fill out the informed consent. The participants will receive their initial briefing (Appendix A: Initial Participant Briefing) of the study and read the informed consent form (Appendix B: Informed Consent) as a group at the visitor center. Upon turning the initial participant information form in, and being accepted as suitable for participation in the project, the

participant will receive a numbered vest with their participant number to be tied on by a same gendered member of the research staff. This number will be how each participant is tracked throughout the study. Participants will then be transported in groups to CAMI. Upon arrival at CAMI, participants will be taken to room 127. In room 127, each participant will be taken to one of several visually isolated stations within the room for anthropomorphic measurement by a same gendered member of the research staff (Appendix D: Participant Anthropometrics Worksheet). Participants will then be taken to the simulated aircraft seating arrangement station for research question 1. Participants will give the research staffer at the seating arrangement station their demographics form and be asked to sit in the seats if they are able. After completing the first test, participants will be taken to room 117 to sit at one of the tables in that room. Participants will then be given an experimental seating post-test form (Appendix E: Experimental Seating Post-Test Questionnaire) to describe their subjective experience with the simulated seating. Once all participants have turned in this form, they will be briefed (Appendix F: Pre-Test Briefing) as a group about the specifics of the trials. After this briefing, the participants will be taken to the simulator, handed their boarding cards with specific trial seating placement, and board the simulator. After seating, research staff will collect the briefing cards and the PI will give the pre-trial briefing (Appendix G: Pre-Trial Briefing) and leave before sounding the buzzer that signals the trial start. After the evacuation, participants will be gathered at the participant gathering tents and given the between trials survey (Appendix H: Between Trials Survey). Upon completion, they will take the survey to a research staff table to receive their boarding pass for the next trial. The participants will then repeat the boarding and evacuation process three more times. Participants will then be taken back to room 117 to fill out a post-test survey (Appendix I: Post-Test Survey) and receive a group debriefing. Participants will then coordinate with the contractor for their payment for participation plus any bonus, before turning in their vests and being taken in groups back to the visitor’s center for final payment processing by the contractor and dismissal. This procedure is expected to take no more than 5 hours per group.

STAFF FAMILIARIZATION AND SITE RISK ASSESSMENT

The complexity and size of this research project and the associated research staff will require at least two “dry-run” trials before any trials involving paid human subjects, for staff familiarization and identification of any weaknesses of the methodology above that may need to be addressed. These staff familiarization runs will require 20–30 volunteers from around the MMAC campus to play the role of test subjects with the pre-test day

understanding that these trials are to allow the research staff to practice the procedures for an upcoming research project before the actual trials begin. These volunteers will sign and receive a copy of the familiarization run informed consent (Appendix J: Informed Consent for Project Familiarization Groups) and otherwise receive the same briefings as actual subjects, with the understanding that they will not be competing for a bonus during any evacuation, and will only be participating in two mock evacuations with no flight attendants present. This will allow research staff to identify the flow and better estimate time requirements for checking in subjects, assigning numbers, measuring demographics, and changing over the simulator seat width between runs. All data sheets filled out during these familiarization runs will be destroyed before the participants leave the area; no data about the participants in these familiarization runs will be retained.

After a final set-up of the experimental area and before any pre-test rehearsal runs for staff familiarization, the PI will request a site visit by AMP-100 to help identify any obvious hazards that subjects or staff may encounter during the test runs. Any findings of this review will be reviewed by the research staff and IRB for determination of which, if any, identified hazards are correctable or inherent to an evacuation research project. All correctable hazards will be mitigated before data collection with research subjects begins.

Data Collection, Analysis, and Confidentiality

Data collection will be pen and paper for demographics, anthropomorphic, and pre- and post-test surveys. Video data will be collected with the assistance of the CAMI iZone team recording participants sitting in the simulated aircraft seating, and all activities in the simulator at 30 frames per second in high definition. Additional recording in and around the simulator will be accomplished using the FlexSim built-in video data collection system, consisting of 16 Go-Pro cameras recording at high definition and 30 frames per second. Breakdown of the video and all data entry will take place in rooms 108 and 110. Microsoft Office Excel and IBM SPSS 23 will be used for all data entry and analysis.

Participant confidentiality will be maintained by ensuring that the participant names are only retained on the informed consent documents held by the PI in a secured storage container. Any reference to any participant in the data or subsequent reports will be made by referring to the participant’s number only. Any publications using pictures generated in this project will have participant’s faces obstructed by opaque coloring or other digital editing means.

Design

This project will consist of three series of 2 by 2 between-subjects tests. The test matrix is illustrated in Table 1. This design was selected as the best compromise between collecting the data required to answer the research question, the defensibility and understandability of the subsequent data, and the limitations of the research facility being used. The number of groups (12) was selected to allow for proper counter balancing of the run order within each series (Table 2). The group size (60) was selected to best simulate the crowded conditions that may occur during an airplane evacuation, that is, to provide the most realistic evacuation experience possible to draw the data from. Pre-test analysis shows that 30 data points per cell would provide enough statistical power to identify a moderate effect. This design will provide 237 points of data per cell for individual egress times.

Table 1: Proposed Test Matrix

OBSERVERS EXTERNAL TO STUDY

This research project has received an inordinate amount of outside attention, resulting in multiple parties requesting invitations and permission to view the evacuation trials themselves. As the PI, I have been refusing any observers who are not federal employees of the U.S. Department of Transportation (DOT) or of the FAA and only accepting those federal employees who would be required to view the evacuation research project trials as part of their job function (such as the study sponsor or the DOT Office of Inspector General, which is currently conducting an audit of the FAA’s Cabin Safety program as a whole), or those who have contributed to the success of the research project thus far and may be called upon as research staff volunteers (select Cabin Safety Inspectors from the Flight Standards branch, for example).

There is, however, one group which is not exclusively made up of federal employees, which has a specific function that can be (and apparently has been) interpreted to require that they be allowed to observe some of the research trials in person (though the value of direct observation vs. being

Table 2: Group Run Order

able to observe the recordings of the trials in their entirety is debatable). The FAA Evacuation Aviation Rulemaking Committee (ARC) was formed in accordance with Section 337 of the FAA Reauthorization Act of 2018. The Evacuation ARC, and upper management within the FAA, are requiring that this group be allowed to observe, and, as the PI, in consultation with the IRB, I am amending this IRB narrative to detail the observation scenario, restrictions, and requirements.

The Evacuation ARC has scheduled a meeting to be hosted by CAMI, December 11–13. This will overlap one planned test day (December 12) for this research project. To be allowed to observe live test subjects, ARC members will be required to attend a briefing by the IRB Chair about the ethics involved in the use of human subjects for research on December 12 at 08:00. After this briefing, ARC members wishing to observe the tests will

be required to sign an agreement document (Attachment K: Evacuation ARC Test Observer Agreement) including a section on the non-disclosure of information relating to research subjects. The intent of the agreement is to bind those outside observers to similar limitations as federal employees regarding disclosure of personally identifiable information of human test subjects, as well as an agreement to not interact with the test subjects or attempt to interfere in the course of the evacuation trials they will be observing or the research project as a whole. They will also agree, if they wish to observe the trials, to leave any cell phone or any other device with recording capability within their meeting room or other secure location (such as a vehicle) and not to bring any device capable of making recordings with them to the designated observation building. This prohibition is specifically including any objects capable of taking notes (pens/pencils/paper) as well as electronic recording devices.

The day of the evacuation trials, the Evacuation ARC will be taken to the designated observation building, the blue barn just south of the FlexSim at approximately 11:45 AM, before any subjects are taken out to the FlexSim to begin their evacuations. The designated observation building has been equipped with a monitor which will have mirrored camera feeds that are going into the Research Observation Station (ROS) within the FlexSim. These will be six camera feeds, three each from forward and aft of the cabin, which will show (1) a down-the-aisle view, (2) an internal exit view, and (3) an external exit view of the exits being used for the trials. As these are live feeds going from the FlexSim to the monitor, they will not have the capability to be manipulated (pause, rewind, etc.). Once within the designated observation building, the Evacuation ARC members will not be allowed to leave until there are no subjects within or around the FlexSim (either between trials 2 and 3, or after the conclusion of trial 4). After subjects are out of the area, Evacuation ARC members will be escorted back to their meeting room to resume their scheduled agenda.

If the designated observation building is unable to be used for whatever reason, the research team will coordinate the use of an alternate location, either with direct observation ability or a location capable of accepting the ROS view feeds.

REFERENCES

McLean, G. A., and Corbett, C. L. (2004). Access-to-Egress III: Repeated Measurement of Factors that Control the Emergency Evacuation of Passenger Through the Transport Airplane Type-III Overwing Exit. (DOT/FAA/AM-04/2).

McLean, G. A. (2001). Access-to-Egress: A Meta-Analysis of the Factors that Control Emergency Evacuation Through the Transport Airplane Type-III Overwing Exit. (DOT/FAA/AM-01/2).

McLean, G. A., Corbett, C. L., Larcher, K. G., McDown, J. R., Palmerton, D. A., Porter, K. A., Shaffstall, R. M., and Odom, R. S. (2002). Access-to-Egress I: Interactive Effects of Factors that Control the Emergency Evacuation of Naive Passengers Through the Transport Airplane Type-III Overwing Exit. (DOT/FAA/AM-02/16).

McLean, G. A., George, M. H., Funkhouser, G. E., and Chittum, C. B. (1996). Aircraft Evacuations onto Escape Slides and Platforms I: Effects of Passenger Motivation. (DOT/FAA/AM-96/18).

McLean, G. A., George, M. H., Funkhouser, G. E., and Chittum, C. B. (1999). Aircraft Evacuations onto Escape Slides and Platforms II: Effects of Exit Size. (DOT/FAA/AM-99/10).

Muir, H., and Cobbett, A. M. (1995). Influence of Cabin Crew During Emergency Evacuations at Floor Level Exits. (CAA Paper 95006:DOT/FAA/AR-95/52).

Muir, H. C., Marrison, C., and Evans, A. (1989). Aircraft Evacuations: The Effect of Passenger Motivation and Cabin Configuration Adjacent to the Exit. (CAA Paper 89019).

This page intentionally left blank.