Modeling passenger comfort in turboprop aircraft using objective measures.

Only 36 datasets from two flights on one aircraft type were available, and four of the 40 jackets failed to record. The authors say this constrained dataset limits accuracy. Noise and vibration in each passenger's own immediate surroundings could not be measured, so those factors are absent from the models even though the authors say the literature treats them as important. The aircraft used had a relatively large 34-inch seat pitch, which the authors say may have changed how much other measurements such as height and popliteal height mattered. The findings may not carry over to aircraft with less legroom. The data were collected on real flights rather than in a controlled setting, which the authors say added noise to the measurements. Features that stayed constant during the flight, such as seat width, were not put into the model. Passengers chose their own seats from among the 20 jacket seats, so who sat where was not assigned by the researchers. The comfort and discomfort scores came from questionnaires and were rescaled before analysis, so the models predict changes in scores rather than absolute comfort levels.

The authors declare no conflicts of interest. The work was funded by the European Union's Horizon 2020 COMFDEMO project, grant 831992.

Do not treat these factor rankings as things you could change for a patient. The study measured 36 datasets on two flights and reports which measurements moved with questionnaire comfort scores. Nothing was altered to test whether changing hip width, humidity or air quality would make passengers more comfortable, and no health outcome was measured.