Dropdown items
My Academies

Personal Library

Account settings

Peer Review of the Federal Aviation Administration's Study of the Effects of Passenger Seat Width and Pitch on Airplane Evacuation Performance (2025)

Chapter: Appendix C: Preliminary Analysis of Reported Evacuation Times

Visit NAP.edu/10766 to get more information about this book, to buy it in print, or to download it as a free PDF.

Previous chapter
Page of 98
Search this publication

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

Page 91 Bookmark

Suggested Citation: "Appendix C: Preliminary Analysis of Reported Evacuation Times." National Academies of Sciences, Engineering, and Medicine. 2025. Peer Review of the Federal Aviation Administration's Study of the Effects of Passenger Seat Width and Pitch on Airplane Evacuation Performance. Washington, DC: The National Academies Press. doi: 10.17226/29070.

C

Preliminary Analysis of Reported Evacuation Times

The total evacuation times for each of the repeat trials are reported in Tables 24, 25, and 26 of the CAMI report¹ and reproduced here with highlighted entries in Tables C-1, C-2,² and C-3. Given the potential for learning effects, the analysis in this appendix focuses on the times from the first trial only, or the second trial only, or from the combined first and second trials for each cohort. For each seat pitch/width trial, Tables C-2 and C-3 identify the data from Tables 25 and 26, respectively, that correspond to the first trial for a cohort (highlighted in yellow) and the second trial for a cohort (highlighted in orange).

___________________

¹ Weed, D. B., et al. (2021). Effects of Airplane Cabin Interiors on Egress I: Assessment of Anthropometrics, Seat Pitch, and Seat Width on Egress. https://www.faa.gov/sites/faa.gov/files/2022-04/Effects_of_Airplane_Cabin_Interiors_on_Egress_I.pdf.

² There appears to be an error in either Table 24 or Table 25 of the CAMI report. The evacuation times for day 7 do not match and are different in the second decimal point. Only the time for the second trial of day 7 (39.87) matches in both tables.

Page 92 Bookmark

TABLE C-1 Adjusted Group Egress Times (in seconds) by Test Day and Egress Trial Order

Test Days 1–12 (N = 54)
Run	1	2	3	4	5	6	7	8	9	10	11	12
1	43.77	39.42	42.52	43.03	39.90	35.32	35.95	38.67	37.70	38.07	36.08	38.38
2	39.92	39.03	39.45	36.65	33.98	34.82	39.87	33.53	30.77	38.22	32.67	33.03
3	39.10	37.18	35.83	37.82	37.75	34.23	32.83	39.00	32.48	38.95	33.65	34.05
4	37.95	38.03	37.25	36.52	32.00	30.63	37.02	33.68	34.17	33.75	33.73	34.13

NOTES: Total evacuation times for each of the 48 trials over 12 days. Yellow highlight indicates the first run or trial for a cohort. Orange highlight indicates the second run for a cohort.
SOURCE: Table 24 from Weed et al. (2021). https://www.faa.gov/sites/faa.gov/files/2022-04/Effects_of_Airplane_Cabin_Interiors_on_Egress_I.pdf.

TABLE C-2 First Testing Matrix Groups by Seat Pitch and Seat Width Combinations

Test Days 1–8 (N = 54)									Basic Descriptive Statistics
Condition	1	2	3	4	5	6	7	8	Min (Fast)	Mean	Max (Slow)
32/18	43.77	37.18	39.45	36.52	33.98	30.63	35.97	39.00	30.63	37.06	43.77
32/16	39.10	39.42	37.25	36.65	32.00	34.82	32.85	38.67	32.00	36.34	39.42
28/18	37.95	39.03	35.83	43.03	37.75	35.32	37.03	33.53	33.53	37.43	43.03
28/16	39.92	38.03	42.52	37.82	39.90	34.23	39.87	33.68	33.68	38.25	42.52

NOTES: Total evacuation times for test days 1–8 by seat pitch and seat width combinations. Yellow highlight indicates the first run or trial for a cohort. Orange highlight indicates the second run for a cohort.
SOURCE: Table 25 from Weed et al. (2021). https://www.faa.gov/sites/faa.gov/files/2022-04/Effects_of_Airplane_Cabin_Interiors_on_Egress_I.pdf.

Page 93 Bookmark

TABLE C-3 Second Testing Matrix Groups by Seat Pitch and Seat Width Combinations

Test Days 9–12 (N = 54)					Basic Descriptive Statistics
Condition	9	10	11	12	Min (Fast)	Mean	Max (Slow)
32/18	37.7	38.95	32.67	34.13	32.67	35.86	38.95
32/16	32.48	38.07	33.73	33.03	32.48	34.33	38.07
34/18	34.17	38.22	33.65	38.38	33.65	36.1	38.38
34/16	30.77	33.75	36.08	34.05	30.77	33.66	36.08

NOTES: Total evacuation times for test days 9–12 by seat pitch and seat width combinations. Yellow highlight indicates the first run or trial for a cohort. Orange highlight indicates the second run for a cohort.

SOURCE: Table 26 from Weed et al. (2021). https://www.faa.gov/sites/faa.gov/files/2022-04/Effects_of_Airplane_Cabin_Interiors_on_Egress_I.pdf.

Page 94 Bookmark

Using these recorded times, Table C-4 shows the total evacuation times for the first and second trials for each cohort as a function of seat pitch and width. Table C-5 shows the average total evacuation times for the first and second trials combined.

Albeit preliminary, the following observations and resultant questions may deserve further consideration by the CAMI research team.

The results from the first trial for each cohort are the most reliable as they represent the results from completely naive participants, with no learning effects. Thus, to eliminate the potential impact of learning effects, only the first trial for each can be considered. Note that the mean evacuation times for the second trial for a given seat configuration are consistently faster than the first trial (see Table C-4). With the exception of the first and second trials on day 7 and

TABLE C-4 Evacuation Times for the First and Second Trials for Each Cohort with Mean Evacuation Times and Mean Evacuation Times for a Given Seat Pitch

1st or 2nd Trial for a Given Cohort	Pitch/Seat Condition	Evacuation Times from Days 1–12 (in Seconds)			Mean Time (in Seconds)	Mean Time for the Pitch (in Seconds)
1st or 2nd Trial for a Given Cohort	Pitch/Seat Condition	(Contains Cohort/Test Day Number)			Mean Time (in Seconds)	Mean Time for the Pitch (in Seconds)
1	28/16	42.52 [3]	39.90 [5]		41.21	40.2
1	28/18	43.03 [4]	35.32 [6]		39.18	40.2
1	32/16	39.42 [2]	38.67 [8]	38.07 [10]	38.72	38.93
1	32/18	43.77 [1]	35.96^a [7]	37.70 [9]	39.14	38.93
1	34/16	36.08 [11]			36.08	37.08
1	34/18	38.38 [12]			38.38	37.08

2	28/16	39.92 [1]	39.87 [7]		39.9	38.09
2	28/18	39.03 [2]	33.53 [8]		36.28	38.09
2	32/16	36.65 [4]	34.82 [6]	33.03 [12]	34.83	35.1
2	32/18	39.45 [3]	33.98 [5]	32.67 [11]	35.37	35.1
2	34/16	30.77 [9]			30.77	34.5
2	34/18	38.22 [10]			38.22	34.5

^a Day 7 data point where average data from the two conflicting data entries are used. SOURCE: Times and cohort numbers are from Tables C-2 and C-3 above.

Page 95 Bookmark

TABLE C-5 Mean Evacuation Times Derived by Combining Data for the First and Second Trials for Each Cohort with Mean Evacuation Times and Mean Evacuation Times for a Given Seat Pitch

Pitch/Seat Condition	Mean Time (in Seconds)	Mean Time for the Pitch (in Seconds)
28/16	40.55	39.14
28/18	37.73	39.14
32/16	36.78	37.02
32/18	37.26	37.02
34/16	33.43	35.87
34/18	38.3

day 10, the same is also true for the individual seat configuration trials for each day. On day 7 and day 10, the second trial is slower than the first trial.
With the exception of days 7 and 10 (italicized below), the second trials for each cohort are quicker than the first (see Table C-1, Table 24, from the CAMI report). For the 10 trial days excluding these two anomalous days, the second trials are between 1% and 18.4% faster than the first trials. However, because the trial conditions for the second trials are not the same as for the first trials, there could be other factors (such as seat configurations or participant body sizes) that contribute to the second trials being faster than the first.
1. Day 1: first trial 32/18 (mid pitch, large width), second trial 28/16 (small pitch/width), second trial 8.8% faster.
2. Day 2: first trial 32/16 (mid pitch, small width), second trial 28/18 (small pitch, large width), second trial 1.0% faster.
3. Day 3: first trial 28/16 (small pitch/width), second trial 32/18 (mid pitch, large width, second trial 7.2% faster.
4. Day 4: first trial 28/18 (small pitch, large width), second trial 32/16 (mid pitch, small width), second trial 14.8% faster.
5. Day 5: first trial 28/16 (small pitch/width), second trial 32/18 (mid pitch, large width), second trial 14.8% faster.
6. Day 6: first trial 28/18 (small pitch, large width), second trial 32/16 (mid pitch, small width), second trial 1.4% faster.
7. Day 7: first trial 32/18 (mid pitch, large width), second trial 28/16 (small pitch/width), second trial 10.9% slower.

Page 96 Bookmark

Day 8: first trial 32/16 (mid pitch, small width), second trial 28/18 (small pitch, large width), second trial 13.3% faster.
Day 9: first trial 32/18 (mid pitch, large width), second trial 34/16 (large pitch, small width), second trial 18.4% faster.
Day 10: first trial 32/16 (mid pitch, small width), second trial 34/18 (large pitch/width), second trial 0.4% slower.
Day 11: first trial 34/16 (large pitch, small width), second trial 32/18 (mid pitch, large width), second trial 9.4% faster.
Day 12: first trial 34/18 (large pitch/width), second trial 32/16 (mid pitch, small width), second trial 15% faster.

Questions deserving considerations from the results:

Are these differences statistically significant?
Can the results be explained by learning effects alone? What explains the anomalous results of day 7 (see g) and day 10 (see j)?
Could seat configurations, body size, and seating location of participants with large body size have contributed to these differences?
On a given day, as the same people participated in each trial, could seating location of the larger participants have contributed to the faster performance in the second trials?
On day 4 and day 6, two different cohorts were tested with the same seating configuration for Trials 1 and 2, but day 4 is 14.8% faster and day 6 is 1.4% faster. Also, the first and second trials on day 4 are consistently slower than the first and second trials on day 6. Is there a difference in the number of larger sized participants on these trial days or the seating location of the larger participants?
Repeat question (v) for day 3 and day 5 when same seating configurations was tested on both days, but trial 1 on day 3 is 7.2% faster than trial 2 while on day 5 the trials are 14.8% faster.

For both the first and second (i.e., repeat) trial for each cohort, there are at most only three repeat data points for a given seating configuration and this is only for the 32/16 and 32/18 configuration (i.e., three cohorts produced first trial data for the 32/16 configuration, and three other cohorts produced first trial data for the 32/18 configuration, and similarly for the second trial data for these configurations; see Table C-4). Given the number of repeat data points, the averages for this seat configuration are probably the most reliable. However, there is only one data point for each

Page 97 Bookmark

of the 34/16 and 34/18 configurations for the first trial and one data point for each configuration from the second trial (i.e., only two cohorts experienced these configurations as a first or second trial). Given the small number of repeat data, the data for these two seat configurations are likely to be least reliable. Furthermore, it is counterintuitive that for the second trial for the two cohorts who experienced the 34/16 and 34/18 configurations, the evacuation time for the 34/18 trial (38.22 s) is almost 25% larger than for the 34/16 trial (30.77 s), suggesting that these data may be of questionable reliability. Consider that not only is the difference large (25%), but one would also expect that if seat width played a role, the wider seat would produce a smaller evacuation time. (NOTE: For the first trial for the two cohorts, the 34/18 trial [38.38 s] is only 6% larger than for the 34/16 trial [36.08 s]; thus while still larger, the difference is much smaller.)
It could be helpful to analyze each of these trials carefully, at least for the first trials of each cohort, and potentially the second trials for each cohort as well. There are some large differences between evacuation times for a given seat/pitch configuration (e.g., for the first trials of a cohort at 32/18—there is a 22% difference between the 35.96 s and 43.77 s evacuation times). Is this the natural variation that would be expected due to a repeat evacuation with different cohorts, or is it to do with the body size of the participants in each cohort—for example, were there more large-body-sized participants in the cohort producing the 43.77 s trial or were the large participants seated in critical locations? Another example that stands out is the difference in the second trial results for cohorts in the 34/16 and 34/18 trials, producing evacuation times of 30.71 s and 38.22 s, a difference of almost 25%.
For a given seat width, mean evacuation time decreases with seat pitch for the first trial (completely naive participants) and for the second trial (potentially trained participants?) except for the 18-inch seat as shown in Table C-4.
1. For the first trial results for the 16-inch width, the decrease in evacuation time with increasing seat pitch is 12.4%.
2. For the first trial results for the 18-inch width, the decrease in evacuation time with increasing seat pitch is 2.0%.
3. For the second trial results for the 16-inch width, the decrease in evacuation time with increasing seat pitch is 22.9%.
4. For the second trial results for the 18-inch width, the increase in evacuation time with increasing seat pitch is 5.3%.

Page 98 Bookmark

If the results for the seat widths are combined for a given seat pitch, evacuation time decreases with seat pitch for both the first and second trials—see Table C-4.
1. For the first trial results with combined seat widths, the decrease in evacuation time with increasing seat pitch is 7.8%.
2. For the second trial results with combined seat widths, the decrease in evacuation time with increasing seat pitch is 9.4%.
If the results for the first and second trials for each cohort are combined, for a given seat width, mean evacuation time decreases/increases with seat pitch—see Table C-5.
1. For the 16-inch width, the decrease in evacuation time with increasing seat pitch is 17.6%.
2. For the 18-inch width, the increase in evacuation time with increasing seat pitch is 1.5%.
Combining the results for first and second trials for each cohort and the results for different seat widths, the mean evacuation time decreases with seat pitch—see Table C-5.
1. The decrease in evacuation time with increasing seat pitch is 8.4%.
For a given seat pitch, the mean evacuation time for the 16-inch seat width is less than the mean time for the 18-inch seat width, except for the 28-inch seat pitch for both the first and second trials—see Table C-4.
1. For the first trial results, the mean evacuation time across all seat pitches for the 16-inch seat width is 38.67 s, while for the 18-inch seat width it is 38.90 s.
2. For the second trial results, the mean evacuation time across all seat pitches for the 16-inch seat width is 35.17 s, while for the 18-inch seat width it is 36.62 s.
3. Combining the results for the first and second trials, the mean evacuation time across all seat pitches for the 16-inch seat width is 37.17 s, while for the 18-inch seat width it is 37.76 s.
4. The differences in mean evacuation times as a function of seat width across all seat pitches are small and unlikely to be significant.

These preliminary results appear to suggest that seat pitch has a greater influence on evacuation time than seat width. It merits emphasizing, however, that the analyses presented here are offered to suggest some additional lines of inquiry for CAMI and the results were not tested for significance.