248 The UMAP Journal 28.3 (2007) We conclude that boarding strategies starting farthest away from the en- trance or farthest away from the aisles yield shorter boarding times than ran- dom boarding. The most successful methods are combinations of these strate. gies, their detailed implementation depending on the exact layout/size of the aircraft. The method yielding the shortest total boarding time is not neces- sarily the one with shortest average boarding time per passenger. By consid ering standard deviations of total and individual boarding times over many iterations of the simulation, we can derive conclusions regarding the stabil- ity /consistency of the specific boarding strategies and how evenly the waiting time is distributed amongst the passengers. By selecting appropriate strategies, time savings of 2-3 min for st medium aircraft could be achieved. for a custom 800-seat aircraft aisles, more than 6 min could be saved compared to random boarding the pared our results to actual turnaround times quoted by airlines, we believe m to be realistic. The text of this paper appears on pp. 463-478248 The UMAP Journal 28.3 (2007) We conclude that boarding strategies starting farthest away from the en￾trance or farthest away from the aisles yield shorter boarding times than ran￾dom boarding. The most successful methods are combinations of these strate￾gies, their detailed implementation depending on the exact layout/ size of the aircraft. The method yielding the shortest total boarding time is not neces￾sarily the one with shortest average boarding time per passenger. By consid￾ering standard deviations of total and individual boarding times over many iterations of the simulation, we can derive conclusions regarding the stabil￾ity! consistency of the specific boarding strategies and how evenly the waiting time is distributed amongst the passengers. By selecting appropriate strategies, time savings of 2-3 min for small and medium aircraft could be achieved. For a custom 800-seat aircraft with two aisles, more than 6 min could be saved compared to random boarding. Having compared our results to actual turnaround times quoted by airlines, we believe them to be realistic. The text of this paper appears on pp. 463-478