It will be noted from the foregoing table that during the period under consideration there were a total of 931 accidents. Fifteen percent, or 138 accidents, were Fatal Accidents. Of these 138 Fatal Accidents, 89 were Non-Survivable Fatal Accidents and 49 were Survivable Fatal Accidents. Upon individual investigation of each of these 49 Survivable Accidents, it was determined that 13 were not pertinent to this study. These 13 involved deaths as a result of accidents which did not require an emergency evacuation. The 36 Survivable Fatal Accidents remaining are pertinent to the problems investigated in this study. In each case a "crash" landing was made, damaging the aircraft and necessitating passenger evacuation by all possible exits. Details of these accidents are shown in Appendix 2. In addition to the 36 Survivable Fatal Accidents, there are certain Non-Fatal Accidents which are also pertinent. These are included in the overall statistics shown in Table 1. A close examination of Table 1 reveals that during the eleven year period under consideration, there were 931 minus 138, or, 793 Non-Fatal Accidents. An analysis of these indicated that 23 of them involved emergency evacuation. These emergency evacuations took place for two different reasons: four were necessitated by an on-board fire but were not complicated by crash damage to the aircraft. (One was an aborted take-off. The other three were normal landings.) The remaining 19 evacuations took place because the aircraft had been "crash" landed in each case. However, all 23 required emergency evacuation, and details are shown in Appendix 3. The statistics shown in Appendices 2 and 3 were extracted from statistical analyses prepared by the National Transportation Safety Board. More detailed information concerning these 59 events (36 Survivable Fatal plus 23 Non-Fatal Accidents) was obtained by studying the original accident reports when available, and other reports prepared by the National Transportation Safety Board and the International Civil Aviation Organization. These reports varied considerably in the treatment they gave to the individual factors that are important to this investigation, i.e., cause of death, exits used, evacuation details, etc. In general, reports prepared after 1962 treated human factors as a separate problem, with the result that this aspect of the report was thorough and detailed. Prior to that time, this type of information was scattered throughout the report. As a result it was often incomplete and, on occasion, totally lacking. Nevertheless, the information that was available has provided significant data. These data have been recorded on the form shown in Appendix 4 and have been used, in turn, to generate the other tables and data in this report. "Seven were cases in which individuals were struck by propellers, and two were taxi accidents. The four remaining were: a ground crewman run over by a taxiing airplane; an occupant sucked out of a door in flight; clear air turbulence; a man ignited dynamite in the lavatory and blew himself out of the airplane in flight. 74-827 O-72-59 In summary, the general accident picture with which this report is concerned is as follows. c. Personnel Survival When an accident occurs in flight, the personnel on board are confronted with two basic problems: first, they must withstand the impact forces incident to the emergency landing; second, they must evacuate the aircraft before they fall victim to secondary effects of the landing such as fire, smoke, noxious gases or drowning. Although this study is concerned only with the second problem, the forces engendered by the landing impact must be taken into consideration since they can vary considerably and have a wide range of influence on the secondary effects of an emergency landing. The effect of the impact in an emergency landing on the survival of personnel varies from none to being the sole cause of instant death to all on board. The four Non-Fatal Accidents which were not complicated by crash damage (Appendix 3) are examples of emergency landings in which the impact of the landing had no influence on the ultimate survival of personnel. On the other hand, the 89 Non-Survivable Fatal Accidents were crashes in which the forces engendered by the impact of landing were beyond human tolerance and, hence, probably the sole cause of death. The 36 Survivable Fatal Accidents shown in Appendix 2 and the remaining 19 Non-Fatal Accidents in Appendix 3 fall in between these two extremes. An analysis of the causes of death in the 36 Survivable Fatal Accidents is contained in Table 2 below. Here, cause of death is divided into four different categories: Category I - Death from trauma brought on by injuries sustained at the time of impact. - Death from fire. Category II Category III Those cases where death could have been due either to fire or trauma, or both. 15%. It is equally evident that these increases are due solely to an inability to evacuate promptly. Category II represents an even more blatant lack of ability to evacuate aircraft promptly after crashes. In investigating each of these accidents separately, it was determined from the sequence of events that led up to impact with the ground that the aircraft involved in Category II events struck the ground with a glancing blow. As a result, the force of the impact was dissipated over a period of time and the forces engendered in the final stoppage were minimal. This is obvious from the fact that there were no deaths due to trauma in any one of these seven accidents in Category II. Nevertheless, there was a death rate of 39% due to fire in these same seven accidents. The entire death rate can only be explained in terms of an inability to evacuate promptly. Having made these determinations, the following evaluation of deaths due to inability to evacuate may be made. Before any evaluation of this summary is made, it should be understood that determination of the cause of death in the accidents covered in Categories I and II is precise in each case. However, the cause of death in Categories III and IV is not. For example, in Category III, there is no indication in the public record of the number of deaths due to trauma, fire, or a combination of both. The same is true of the four accidents in Category IV. All deaths in these four accidents were reported as being due to drowning. However, an examination of Category IV events reveals that one was a controlled ditching directly into heavy swells, and the other three were uncontrolled crashes into water. In all four cases, the impacts were severe enough to cause death by trauma; hence, it seems almost certain that some of the deaths in Category IV were due to trauma rather than drowning. Nevertheless, there is a positive relationship between Categories I, III, and IV that is meaningful and can serve as a guide in discriminating between deaths due to trauma and deaths due to secondary causes. In comparing Categories I, III, and IV, it is immediately apparent that a post-crash fire appreciably increases the "average" death rate in an accident, and the additional hazard of evacuating into water also appreciably increases the "average" death rate. In analyzing the period 1957-1967, it is evident that fire increased the death rate, on the average, from 46% to 64, or an increase of 18%. A crash into water increased the death rate, on the average from 46% to 61% or an increase of The NTSB report of an accident fixes the cause of death, where possible, based upon the evidence found at the crash site. Accident Investigator's are authorized but not required to call for autopsies to specifically assess the cause of death in individual cases. For proprietary reasons, the results of these autopsies are kept in the NTSB confidential file and are not available for public inspection. a. Category I-None b. Category II-All, or 170 deaths c. Category III-18% of 542, or 98 deaths d. Category IV-15% of 158, or 24 deaths This is a total of 292 deaths out of 828 and means 35% of all fatalities in the Survivable Fatal Accidents pertinent to this study were due to an inability to evacuate. d. Evacuation Difficulties It is obvious that this 35% of the death rate can be reduced. It can be reduced by identifying the difficulties that have prevented prompt evacuation in the past and remedy |