Two experiments were conducted in which subjects in a simulated living room environment judged the annoyance of sessions of airplane noise which contained different noise levels and numbers of flyovers. In the first experiment one, two, or four high noise level flyovers occurred at the beginning, middle, or end of 30-min test sessions each of which contained a total of eight flyovers. In the second experiment, one, four or 16 flyover noises occurred in 15, 30, or 60-min test sessions. The time-of-occurrence of the high noise level flyovers in the sessions did not significantly affect annoyance, but annoyance increased with the number of such flyovers. Annoyance decreased with test session duration but increased with the total number of flyovers in the test sessions. These results support an average energy model better than total energy, annoyance decay, or peak noise level models.