Passenger disembarkation takes place in the confined space of the aircraft cabin. Boarding can be regulated to a certain extent, but this does not apply to disembarking at the end of a flight. COVID-19 constraints require that cabin procedures not only be operationally efficient but also effectively reduce the risk of virus transmission to passengers. We have developed a new mathematical model that accounts for these conflicting goals. We used an already improved seat assignment for passenger groups (e.g., families or couples) and implemented a genetic algorithm that generates improved disembarkation sequences. Our use cases show a significant 40% reduction in disembarkation time when physical spacing between passenger groups is required to comply with pandemic regulations. To inform passenger groups about the disembarkation sequence, we propose to activate the cabin lights at the seats in a dedicated way. Thus, our developed methodology could already be applied to actual airline operations.