Towards a more accurate prediction of aircraft passenger discomfort by integrating temporal analysis of seat-interface pressure with static passenger anthropometric features.

The relationships among passenger anthropometric factors, pressure distribution at aircraft-seat interface, and passenger perceptions on seat discomfort are widely investigated by specialized literature to identify meaningful correlations, however without leveraging these insights for predictive modeling. This study aims to fill this gap to obtain a more accurate prediction of passenger seat discomfort by integrating information gained from the temporal analysis of seat pressure distribution with static passenger anthropometric features. Experimental results show a significant improvement of predictive performance of Time Series Classification, with accuracy gain up to 32%, depending on the algorithm.