As you could all imagine I am sure, the title of this lecture embraces a subject which is very close to the airline operator's heart. One of the prime objectives of any airline is to operate on time all the time.

The word ‘delay’ of course can encompass a wide range of problems. As the second oldest airline in the world—after KLM—and the operator of the world's longest average passenger journey length we do not mind admitting that we have had our share of delays over the years. I know this is a very serious subject but I could not resist having a brief look through our records recently to pluck out a few of the more extreme reasons for delays that you run into in the airline business.

Barnes has said ‘simulation is like a meal in a restaurant —more depends on the way it is prepared and served than on the basic ingredients'. We know a good deal about the ingredients, the hardware, of the simulator meal but we are not yet particularly good cooks. The lore of cooking has largely been obtained by ad hoc experiments, and at its highest expression is an art; the same is still true, in the main, of simulation techniques. Simulation should come of age. The precocious youngster grew up relatively undisciplined using the sledge-hammer of technology in an endeavour to produce a ground-borne aeroplane, with little science used in identifying the requirements.

The purpose of this paper is to compare air passenger transportation in North America with ground passenger transportation as it is today and as it will probably be in the next few years. A precise statistical comparison is practically impossible; air figures include both domestic and international travel, scheduled and non-scheduled; rail figures include inter-city traffic and commuting; bus and car figures are for inter-city traffic only. This choice was arbitrary and represents a trade-off between highlighting selected comparisons and strict consistency for its own sake.

In 1971 The Royal Aeronautical Society's Handley Page Air Safety Award funded a preliminary study of the complex aerodynamic phenomena associated with fully deployed parachutes and results were sufficiently encouraging for the Ministry of Defence to sponsor a full research programme in the hope that a more rational approach to canopy design might evolve.

Preliminary investigations revealed that the essentially quasi-static two dimensional concepts of conventional aerodynamics as applied to aircraft were of limited value since a parachute is very much a three dimensional device which has a tendency to undergo large oscillations at relatively low descent airspeeds. The project therefore involved rethinking of basic aerodynamic principles and a return to hydrodynamic theory as applied to airships since they also have pronounced three dimensional anatomical characteristics and experience significant changes in relative air flow in their natural mode of operation. The objective of this paper is therefore to show how this same theoretical approach can be used as a basis for developing computer techniques capable of effectively simulating full scale parachute descent behaviour from basic wind tunnel test data.