Community Research and Development Information Service - CORDIS


VERRES Report Summary

Project ID: 10968
Funded under: FP5-GROWTH
Country: United Kingdom

Acquire knowledge and experimental data from a double deck VLTA cabin section

This result concerns the analysis of passenger behaviour during a series of large-scale evacuation trials conducted in a VLTA mock-up. The primary analysis undertaken by the FSEG of the University of Greenwich concerned passenger use of the stairs and passenger exit hesitation time analysis for upper deck exits. The full report will be made available on the FSEG web site at

A series of evacuation trials were undertaken at Cranfield University designed to explore passengers' use of stairs and also to gain passenger exit hesitation delays from upper deck exits.

Unfortunately, the trials did not proceed in the controlled manner that was originally planned and so the analysis did not yield the detailed information that was originally hoped. The main difficulties associated with these trials were:

1) CC did not behave as originally intended.
2) Camera angles not ideal to record required data.
3) As these trials were conducted with great caution, the majority of crew behaviour at the upper deck exits can be described as extremely non-assertive. This crew behaviour significantly biases the behaviour and hence performance of the passengers. It is thus not clear if the resultant passenger behaviour is a result of the sill height and slide length or the lack of assertiveness of the crew.

While the trials did not proceed in the controlled manner that was originally planned, much has been learned from theses trials.

It is clear from these trials that crew can exert an influence on the performance of passenger stair usage. Passenger behaviour in utilising the staircase is both rich and complex and warrants further investigation. These trials support the view that for crew to consistently make appropriate or optimal redirection command decisions that include the possibility of using the stairs as part of the evacuation route, they must have sufficient situational awareness. Equally, passengers can only make appropriate or optimal redirection decisions if they too have sufficient situational awareness. This situational awareness may need to extend between decks.

Passengers were also noted to make heavy use of the central handrail while both descending and ascending the stairs. The presence of the central HR effectively created two staircases. By effectively separating the crowding on the stairs, reducing passenger-passenger conflicts and providing as additional means of passenger stability, it is postulated that the stair flow rates may be positively influence through the presence of the central HR.

Passenger flow rates were measured for both the UPWARDS and DOWNWARDS flow directions. Flow rates in the UPWARDS direction were found to be greater than flow rates in the DOWNWARDS direction. This was thought to be due to the packing densities on the stairs, which is a function of the motivation of the passengers, the travel speeds of the passengers and the feed, and discharge characteristics of the staircase and surrounding geometry. Clearly, most of the parameters can be influenced by both crew procedures and cabin layout.

Concerning the passenger exit hesitation times for the higher sill height, the trials produced inconclusive results. While the exit flow rates are lower and the passenger exit delay times are longer than would be expected for a normal Type-A exit, it is clear that the extreme unassertiveness of the cabin crew positioned at the exits and the lack of motivation of the passengers exerted a strong influence on the data produced. The reaction of the passengers in these trials was to be expected as the trials were not performed under competitive conditions and the reaction of the cabin crew could also be understood, as safety concerns were paramount given that these were the first trials of their type to be conducted at Cranfield.

Finally, due to the small number of data points provided by these trials, there is insufficient data upon which to claim statistical significance for any of the observations.

Clearly, much more work is required in order to generate essential data to improve our understanding of passenger performance, passenger-crew interaction and passenger-structure interaction within VLTA configurations.

Reported by

University of Greenwich
School of Computing and Mathematical Sciences, University of Greenwich, Old Royal Naval College, 30 Park Row, Greenwich
SE10 9LS London
United Kingdom
See on map
Follow us on: RSS Facebook Twitter YouTube Managed by the EU Publications Office Top