Risk Assessment of Proposed ARFF Standards (2011)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

43 CHAPTER 5 ESTIMATING THE REDUCTIONS IN FATALITIES AND SERIOUS INJURIES ORIGINAL APPROACH In reviewing the accident records, the ACRP oversight panel specifically directed the research team to look for any information in the reports concerning the following six elements: ARFF staffing, ARFF mission and response strategies, ARFF response times, ARFF response area, ARFF equipment requirements, and ARFF training requirements. The reasoning was that if any of these elements were positively or negatively mentioned in the report, the information would be helpful in comparing the FAA, ICAO, and NFPA standards. After reviewing the reports (and in some cases the dockets) for the 81 “accidents of interest,” the research team concluded that the necessary data to perform the analysis that was envisioned by the panel are not available. In some accident reports, it was noted that there was an ARFF response to the accident but did not state what ARFF standards that the airport operator used. The following observations are offered on each of the six areas after reviewing the preceding reports: 1. ARFF staffing: The accident reports often stated the number of firefighters that responded to the accident. They sometimes also state the number of mutual aid responders. The reports sometimes even document the typical number of firefighters on a shift. However, only one report recommended that there should be more firefighters in the initial response. However, in the same report it was noted that additional firefighters would probably not have made a difference in the number of fatalities. 2. ARFF mission and response strategies: Often the accident reports stated that everyone or the vast majority of persons on board the aircraft had evacuated the aircraft by the time ARFF arrived on the scene. Chapter 4 provides some details on the nine accidents where ARFF entered the aircraft and extricated people from the aircraft—sometimes fire was involved and sometimes it was not. 3. ARFF response times: The response times set forth in the three standards (FAA, ICAO, and NFPA) are for demonstrations held under optimum conditions. Since accidents rarely occur in optimum conditions, these “standard” times are somewhat theoretical for actual accidents. In responding to an actual accident the practice is to get there as quickly and safely as possible. In many cases the actual response time of at least the initial vehicle is recorded or estimated in the accident report. However, it is not unusual for the various clocks used in deriving time stamps in an accident to be unsynchronized. Chapter 4 identifies some of the issues that have been mentioned in accident reports regarding response times. These include weather (fog, snow, etc.) and communications (vague information on location of accident or, in case of a runway collision, that two aircraft were involved.) 4. ARFF response area: In the standards, a specific area or location is linked to the ARFF response time. i.e., ARFF needs to demonstrate that it can reach a point in

44 a specified area in a specified time during a demonstration under optimum conditions. In reality, the response area of an airport ARFF unit for an aircraft accident is anywhere on the airport and, in some cases, depending upon mutual aid agreements, may also include areas off the airport. Many of the accidents that were reviewed involved ARFF units responding to accidents on aprons, taxiways, and in safety areas, as well as on runways. 5. ARFF equipment requirements: There were not any accidents where the investigators cited inadequate equipment when an ARFF response was initiated. Sometimes problems that occurred with the equipment during the emergency are mentioned. 6. ARFF training requirements: There has been no mention in any of the reports that the responding firefighters were not adequately trained. ALTERNATIVE APPROACH Since the data necessary for the original approach were not available, the research team proposed to review the aspects of each accident as a team and develop a range of reduction for fatalities and serious injuries where it believed that different ARFF standards could have resulted in such reductions. In making its determination, the research team considered carefully what the accident investigating agency said about the survivability of each accident. Although this approach was somewhat subjective, the members of the research team believed that they had the background to make objective reviews and determinations on these accidents. The team’s decision to determine a range for the reductions, e.g. 0 to 2 serious injuries provided a means to capture uncertainty introduced by the subjective nature of the determination. The research team prepared two case studies using accidents that occurred at Little Rock, AR (6/1/99, KLIT), and Quincy, IL (11/19/95, KUIN), to illustrate this alternative approach and the underlying logic. See the two shadow boxes. While the Little Rock and Quincy accidents were rich with data, it turned out that most of the other accidents were not. This was particularly true for accidents that occurred outside the United States. Some accidents were relatively easy to analyze. For example, many of the fatalities were the result of trauma from impact or where the fire occurred so quickly the accident was considered non-survivable. In those cases, a change in ARFF standards would not have affected the number of fatalities. However, in other accidents, there would be a statement that “all the fatalities were due to impact or were fire related.” In those accidents, the research team could not determine if a change in ARFF standards would have made any difference in the accident outcome. There were two accidents in Indonesia (9/5/05, WIMM and 3/7/07 WARJ) where ARFF at the airports was not in accordance with the applicable ICAO standards. Among the problems noted in these two accidents was an uncoordinated response to the accident with the surrounding communities. The research team concluded that fatalities most likely would have been reduced if the airports were in compliance with the applicable standards.

45 Little Rock, AR, Accident June 1, 1999 A MD-82 aircraft with 6 crew members and 139 passengers crashed on landing after it overran the end of runway 4R during landing at Little Rock, AR. The airplane struck the left edge of the instrument landing system localizer array, passed through a chain link fence and over a rock embankment to a flood plain, located about 15 feet below the runway elevation; and collided with the structure supporting the runway 22L approach lighting system. The captain and 10 passengers were killed; the first officer, the flight attendants, and 105 passengers received serious or minor injuries; and 24 passengers were not injured. The airplane was destroyed by impact forces and a post-crash fire. The controller called the ARFF units on the crash phone after several attempts to contact the flight crew after the airplane landed. The controller indicated the possibility of an accident at the end of runway 4R but did not specify which end of the runway. The ARFF units proceeded to the approach end of runway 4R, but the airplane was off the departure end. As a result, the ARFF units had to travel back to the taxiway at which they entered the runway and then proceed to the other end of the runway. The ARFF units located the airplane about 11 minutes after the initial call from the controller. However, they did not arrive on scene until 5 minutes later (16 minutes after the initial notification), because they had to travel in the opposite direction to an access road, turn onto a perimeter road back in the direction of the accident site, stop to manually unlock a perimeter security gate, and then continue on the perimeter road to the accident site. According to the Pulaski County Coroner, the captain and 5 passengers died as a result of traumatic injuries, and 5 died from smoke and soot inhalation and/or thermal injuries. Because the accident was potentially survivable for the passengers in seats 27E and 28D, the Safety Board considered whether a shorter ARFF response time could have prevented the fatalities but determined that the passengers’ lives would not have been saved if emergency responders had arrived on scene earlier. Even with the shortest possible response time, the passenger in seat 28D would have already received the second- and third-degree burns to over half of her body and the severe inhalation injury from which she later died. The passenger in seat 27E remained on the airplane and therefore needed to be rescued from the wreckage. The Safety Board could not determine whether the passenger in seat 27E would have survived if sufficient ARFF personnel had been available to perform a rescue. The research team believes that the ARFF response time, given the circumstances, was appropriate. Even if the ARFF facility had been closer to the runway to meet NFPA standard of 2 ½ minutes to any point in the Rapid Response Area, the delay would have been experienced as the visibility and heavy downpour would probably have prevented them from seeing the accident site upon leaving the station. Additional ARFF personnel may have allowed earlier entry into the aircraft, but the ARFF vehicular response would be delayed by the need to avoid the passengers who already exited the aircraft, which may have hampered immediate interior rescue attempts. The research team does not believe that a change in ARFF standards would have resulted in a reduction in fatalities in this accident.

46 Quincy, IL, Accident November 19, 1996 To meet its obligations under Part 139 the Quincy Municipal Airport had a 500 gallon ARFF vehicle that was staffed by fire fighters from the Quincy Fire Department when there were air carrier operations with more than 30 passenger seats. This vehicle was stored in a building by the terminal. Since this operation involved a Beech 1900 aircraft with 19 passenger seats, the ARFF vehicle was not staffed at the time of the accident. Since the accident, FAA's legal authority has been changed that allows the agency to certificate and require ARFF service for scheduled air carrier operations with aircraft having more than 9 passenger seats. The coroner found that 10 aircraft occupants died from "carbon monoxide intoxication from inhalation of smoke and soot" and the remaining four occupants died from the "inhalation of products of combustion." The research team believes that any of the current FAA standards, ICAO standards, or NFPA standards would have saved some, if not all, of these lives. The Beechcraft 1900 is 57 feet 10 inches in length. Under the current standards the ARFF service required would be Index A for FAA and Category 4 for ICAO and NFPA. -Response time would not have been a factor since the fire station was located approximately 1800 feet from the accident location. The NTSB determined that the ARFF truck should have been able to reach the accident site in no more than 1 minute. (The crew of the Beech 1900 was scheduled to be relieved at Quincy. The pilot from the relief crew made it on foot to the accident wreckage before the fire started.) -Each of the three standards (FAA, ICAO, NFPA) requires one vehicle for this operation. -The NFPA standard requires three firefighters for this type of aircraft operation. Neither FAA nor ICAO specifies a staffing number; however, it is very likely that under either of these standards only one person would have been assigned to the ARFF duties. -The three standards vary in the amount of agent required with the NFPA requiring the most and FAA requiring the least. It took the Quincy Fire Department about 14 minutes to arrive on the airport. By that time, both aircraft were engulfed in flames. It took them approximately 10 minutes to control the fire. It is impossible to estimate how much agent would have been necessary to control the fire if the initial response had been made from an on-airport ARFF unit. The research team believes that the additional two firefighters and agent that the NFPA standard would have required could have resulted in a reduction of 3 to 14 fatalities. This staffing level would have allowed one firefighter to apply agent to the Beech 1900, another one to take a handline and apply agent to the King Air, and the third firefighter to attempt to extricate the passengers from the Beech 1900. Since the FAA and the ICAO standards would have only one fire fighter responding, that firefighter would most likely have to choose between applying agent to the Beech 1900 with 12 people or the King Air with two people. Any efforts to extricate the passengers would have had to wait until mutual aid assistance arrived. Assuming that the firefighter would choose the Beech 1900, the research team believes that the FAA or ICAO standard would have resulted in a reduction of 1 to 12 fatalities.

47 In the case of Quincy (11/19/96, KUIN) where there was no requirement at the time of the accident for an on-airport ARFF response, the research team concluded any one of the three standards being compared in this study would have resulted in a reduced number of fatalities if it had been in effect. For many of the accidents, the data included in the accident reports were not sufficient to allow the research team to conclude that a change in ARFF standards would have changed the accident outcome in terms of fatalities and/or serious injuries. Based on the data the research team was uncomfortable even with providing a range of estimates for fatalities and serious injuries. If additional data were available, it is possible that the research team may have reached a different conclusion for some of these accidents. Problems with the data are discussed in the remainder of this chapter under the following headings:  Fatality and Injury Information  Standards in Effect at Time of the Accident  Response Time The results of the research team’s analysis and determination for each accident are included in Appendix A. Fatality and Injury Information The research team encountered several problems in trying to use the data from the accident reports to make meaningful determinations. For these determinations it was important to have information on the timing and cause of the fatality/serious injury. Some of these problems included:  While some reports explicitly stated that X persons died on impact and Y persons died from thermal injuries or smoke inhalation, other reports stated that there was only a handful of autopsies and summarily stated that the remaining fatalities were all due to injuries that were either impact-related or fire-related.  For accidents that involved several fatalities and serious injuries, the accident reports often focused on the fatalities and provided very little information on the serious injuries as to how they happened or when they occurred.  As discussed in Chapter 4 there were 38 accidents where all the serious injuries from the accident occurred during the evacuation of the aircraft. Most of these accidents involved only one or two people being seriously injured. The accident report often provided specifics on how the injury occurred such as passenger broke ankle from jumping off the wing or passenger broke wrist falling off the slide. Most of these injuries fell under the second criterion of the definition for serious injury, i.e., results in a fracture of any bone (except simple fractures of fingers, toes, or nose), and did not appear to be life-threatening. Some of the serious injuries in the remaining 43 accidents also occurred during the evacuation of the aircraft but those accidents also had fatalities/serious injuries due to impact.

48  The NFPA, in NFPA 402, Guide for Aircraft Rescue and Fire Fighting Operations, includes a provision that states “If time and conditions permit, ARFF personnel should assist in the off-loading of evacuees at the base of the evacuation slides to minimize injuries.” However, there is no specific guidance on how the firefighters should prevent these injuries. In view of the generality of the provision, it appears to be an action that should be considered by the on-site incident commander but with the current wording certainly could not be considered a requirement. Standards in Effect at the Time of the Accident For some accidents at non-U.S. airports the accident report did not specifically state the ARFF standards applicable to the airport. In these accidents it was assumed that the airport operator complied with the ICAO Annex 14 standards. There were two accidents at non-U.S. airports where the investigating agency specifically stated that the airport operator was not in full compliance with all aspects of the Annex 14 standards. The research team did not find any accidents where it was stated in the accident records that the ARFF standards set forth by the NFPA were in effect at the airport. Response Times It would seem logical that if a standard requires a quicker demonstrated response time, the result would be a quicker response to an actual accident. However, as was shown in the discussion in Chapter 4, real world conditions involving communications, weather, and access to the accident site often make demonstrated response times meaningless during an actual accident. None of the accident reports indicated that any of the airports involved complied with the NFPA demonstrated response times. Trying to locate additional stations on the airfield to comply with current demonstrated response times for a historical accident is difficult. Furthermore, attempting to determine actual response times to those historical accident sites from hypothetical fire stations is not a meaningful effort. CHAPTER SUMMARY The inability to determine the cause of the fatalities and serious injuries, i.e., trauma vs. fire/smoke related, and the timing of these occurrences from the information in the accident reports hinders the ability to make a direct link to how many lives would be saved if changes were made to current ARFF standards. For some accidents that did not involve a fire, the ARFF response to the accident may not even be mentioned or may be discussed in one or two sentences in the accident write-up. For accidents occurring outside the United States, some reports do not even mention what ARFF standards the airport uses. There does not seem to be a correlation between demonstrated response times required by standards and the response time to an actual accident. This can be attributed to the fact that demonstrated response times are achieved in ideal conditions such as daytime, dry pavement, good visibility, and understandable communications. One or more of these ideal conditions are not normally present during a response to an actual accident. In view of the above, the research team found that there was no conclusive information in the accident reports that indicated a change in ARFF standards would have resulted in a reduction in fatalities or serious injuries.