Predictions of Sleep Disturbance for Different Nighttime Airport Operation Strategies Using a New Markov State Transition Sleep Model
Invited paper
Division of Sleep and Chronobiology, Department of Psychiatry
Tuesday 2 june, 2015, 14:40 - 15:00
0.4 Brussels (189)
Abstract:
To balance benefits and costs of potential airport operation changes such as
noise curfews, changes in flight schedules, or flight paths, models are needed
which can predict the time varying nature of the effects of aircraft noise on
sleep. While a Markov transition model has been developed which predicts the
transitions between 6 sleep stages throughout the night (Wake, S1, S2, S3, S4,
and REM), it has two limitations. The Markov model was developed based on data
from a laboratory study in which a greater probability of aircraft noise-
induced awakenings was found compared to field studies. In addition, the model
predicts the same probability of awakening for all aircraft events, regardless
of the noise level. To overcome these two limitations, a new Markov transition
model was developed using data from a total of 483 nights from 63 subjects who
participated in a polysomnographic field study that was conducted around
Cologne-Bonn Airport. Similar to the previous Markov model, transition
probabilities between sleep stages were calculated using 1st-order
autoregressive multinomial logistic regression models. However, in addition to
elapsed sleep time, the maximum noise level has been added to the model as an
explanatory variable. This new Markov model was used to predict the number of
awakenings and the time spent in each sleep stage for different nighttime
noise mitigation strategies. The development of the model and differences and
similarities in the predictions for the different operation scenarios will be
discussed.