University of Maryland Fire Protection Engineering Department Issues White Paper Entitled 'Review of Foam Fire Suppression System Discharges in Aircraft Hangars'

WASHINGTON, March 11 -- The University of Maryland Fire Protection Engineering Department has issued a 22-page white paper on Feb. 22, 2021, entitled: "Review of Foam Fire Suppression System Discharges in Aircraft Hangars".

The white paper was written by James Milke, Nicholas Brondum, Steven Dunne and Colleen Whitcomb.

* * *

Executive Summary

This report is a product of a research study on the impacts of low-expansion foam, high-expansion foam and deluge systems in aircraft hangars. A survey was conducted to determine the circumstances and losses associated with incidents involving foam system discharges experienced by commercial airlines and the US Department of Defense (US DoD). A review of fuel spills in the database compiled by the US Coast Guard (USCG) was conducted to determine how many fuel spills occurred in hangars. Finally, a scan of data from fire incidents from 2009-2018 included in the National Fire Incident Reporting System (NFIRS) was conducted to identify fires in hangars.

The review of the NFIRS data resulted in the identification of fourteen incidents that involved an aircraft in a "parking structure." However, no data was available for these incidents on whether an automatic extinguishing system (AES) was present, what type of AES was installed, and if the AES operated, so no further analysis could be done relative to the performance of installed foam systems.

In the review of 5 years of data in the USCG database, 851 incidents were found to involve a fuel spill, though only 5 spills occurred inside a hangar, with a resulting annual rate of 1 incident per year. The 5 spills in hangars represents 0.6% of the total number of spills. The USCG database does not identify if any fires occurred in any of the spill incidents.

From the survey of incidents compiled by the research team, a total of 245 incidents were reported from the commercial airlines and US DoD. A total of three incidents involving fire were included in the incidents reported back to the 1960's. Of the 245 incidents reported, 233 were included in a 17-year window (one database included an incident from the 1960's). In that 17-year window, only one incident where the foam system discharged was in response to a fire while 201 incidents were accidental in nature, with a known cause and no fire was present to motivate the activation (the cause of the activation was unknown in 31 of the incidents). Overall, an average of 11.8 accidental foam discharges occurred per year. The trend of the frequency of accidental hangar foam system discharges is increasing by almost one incident each year. The most common cause of the accidental discharge was a failure of the suppression system of unspecified cause.

1. Background

This report supplements data included a previous report (Milke, et al., 2019). The previous report outlined code requirements for fixed foam fire suppression systems in Group II/1 aircraft hangars and analyzed the performance of the fixed foam fire suppression systems in those applications. The research team requested incident reports of discharges of foam fire suppression systems from several insurance companies and Fixed Base Operators (FBOs) who provide coverage for either the aircraft and/or aircraft hangar. Damage estimates for aircraft and the building/building systems were requested in the form, along with cause of the discharge and cause of the fire.

In the current project, the survey was expanded to include incidents experienced by commercial airlines and the US DoD. In addition, a search of fire incidents in aircraft hangars was conducted in the National Fire Incident Reporting System (NFIRS) and a review was conducted of data on fuel spills compiled by the U.S. Coast Guard (USCG).

2. Survey Methodology to Collect Foam System Field Data

The research team requested incident reports of discharges of foam fire suppression systems from commercial airlines, US DoD and insurance companies. A form to facilitate data reporting, The Data Collection Form, Foam Suppression System Discharge Analysis, developed by the University of Maryland (UMD) was provided to each of the participating organizations. The data form is included in the Appendix. Damage estimates for aircraft and the building/building systems were requested in the form, along with cause of the discharge and cause of the fire.

3. Data Analysis

3.1 NFIRS Data

An analysis of the National Fire Incident Reporting System data from the years 2009 through 2018 was conducted. First, this data was sorted for incidents occurring in other vehicle storage, including airplane and boat hangars and excluding parking garages. In order to narrow this field down to incidents occurring in aircraft hangars, a second sort was conducted for incidents involving an aircraft. In the ten years examined, fourteen incidents occurred under the circumstances outlined above. However, there was no available data for these incidents on whether an automatic extinguishing system (AES) was present, what type of AES was installed, and if the AES operated, so no further analysis could be done relative to the performance of installed foam systems.

Figure omitted:

3.2 U.S. Coast Guard Data

United States Coast Guard collects data on all hazardous liquid spills, including fuel spills, in the U.S. USCG data from January 2016 to November 2020 was reviewed to assess the frequency and location of fuel spills in hangars (as compared to all other locations) (USCG, 2020). The distribution of the cause of the fuel spill by year is presented in Table 1 and Figure 2. Of the 851 incidents observed during almost five full years, only 5 spills have been found to have occurred inside a hangar, resulting in an annual rate of approximately 1 incident per year. The 5 fuel spills in hangars represents 0.6% of the total number of fuel spills. The USCG database does not identify if any fires occurred in any of the incidents involving fuel spills.

Table omitted:

As indicated in Table 1 and Figure 2 the leading cause of all 856 jet fuel spills in the U.S. over the last 5 years is equipment failure. The term "equipment failure" refers to a variety of events such as leakage from oil tanks, aircraft fuel tank malfunction, and failure of internal fueling lines.

The descriptions of the five incidents of fuel spills which occurred in hangars are:

* 3 of the incidents occurred when maintenance was being conducted

* 1 incident occurred when an aircraft valve was opened (the reason for the valve being opened was not reported)

* 1 incident occurred when the fuel valve was rotated (the reason for the valve being rotated was not reported)

The next most frequent causes of spills are those caused by activities at a fuel island or fuel truck and then operator error. It is also noteworthy that 18.6% of incidents have an unknown cause.

Figure omitted:

3.3 UMD Survey

Analysis of the provided incident report data began with a review to determine if multiple reports were received from two sources for the same incident. In a limited number of cases, using the date and location of the incident, the research team recognized that two incident reports had been filed related to the same incident. These duplicate incidents were thus combined into a single incident.

A total of 245 incidents were reported from the commercial airlines and US Department of Defense (DoD). The DoD compiled information for incidents dating to about 1960. A total of 229 incidents were included in the DoD database. A total of 16 incidents were reported in facilities operated by commercial airlines. The causes for the 245 incidents are summarized as follows:

* 3 incidents include a discharge in response to a fire (all at DoD facilities)

* 214 incidents include a discharge with no fire present, i.e. an accidental discharge

* 31 incidents include a discharge due to an unknown cause/2

Having received reports of 217 total incidents with a known cause, the 3 incidents that occurred in response to a fire represent 1.4% of the total number of reported incidents.

Reviewing the dates of the incidents included in the DoD database, incidents appear to be reported sporadically from the 1960's until 2004. In 2004 to 2020, there are more regular entries of foam discharges in hangars suggesting that more attention was being given to entering any incident involving a foam discharge in the DoD database. As such, the remainder of the analysis in this report will focus on the 17-year period of 2004 to 2020.

Twelve of the incidents reported among the 245 total number of incidents reported occurred prior to 2004. Eleven of the twelve incidents occurred in DoD facilities, while one was in a hangar operated by commercial airline. A summary of the incident reports of foam system discharges from 2004 to 2020 is included in Table 2. Considering that only one of the 205 incidents with a known cause involved a discharge in response to a fire. That one incident represents only 0.5% of the incidents.

Table omitted:

The annual average number of accidental discharges with a known cause from 2004 to 2020 is 11.8 discharges per year. The trend of the frequency of accidental hangar foam system discharges is increasing by almost one incident each year. The most common cause of the accidental discharge was a failure of the suppression system./3

A distribution of the year in which incidents involving accidental discharges occurred and the trend line are included in Figure 3. This graph depicts 201 incidents for all 17 years and does not include any incidents of an unknown cause or an unknown year. The trendline provided in Figure 3 has a slope of approximately 0.895 incidents per year, meaning that the frequency of an accidental hangar foam system discharge is increasing by almost one incident each year.

Of the examined incidents, only 89 incidents reported a monetary damage value. The value of damage reported appears to be only the damage to the contents and the hangar itself. In no case was the cost of clean-up included in any of the incident reports provided via the UMD survey. Consequently, the damage estimates presented throughout this report only represent a portion of the loss, rather than the total cost of the incident. Therefore, the total cost is expected to be much greater than the presented damage value in this report, given that neither clean-up nor environmental impact were included. For those incidents that did not report a monetary damage value, it is likely that a cost was at least associated with the cleanup.

Figure omitted:

The number of incidents per year involving accidental discharges in DoD versus commercial facilities is included in Table 3 and Figure 4. No information is available to explain the variation in the number of incidents by year in the two sets of facilities. The frequency of the incidents in DoD hangars is increasing at a greater rate than those in hangars for commercial airlines.

Table and Figure omitted:

The distribution of the damage incurred in incidents involving an accidental discharge is shown in Figure 5. The 89 incidents are divided into six categories of monetary loss. The reason for the lower number of incidents recorded in this graph, as compared to the overall total number of incidents is due to a lack of reporting. Most of the incidents reports did not reported the associated cost of damage. The cost incurred by each incident is divided into increments of 0.25 million dollars (USD). The increment with the greatest number of incidents is the 0-0.25-million-dollar category with 67 incidents. The total of all damage estimates for incidents with accidental foam discharges is $9.56 M, for an average loss of $0.107 M per incident.

Figure Omitted:

Results of an analysis of the trend in annual total damage in accidental foam discharges is provided in Figure 6. Data from only 2013 through 2020 are included in the graph as no incidents had damage reported from 2004 to 2012. The data from a total of 88 incidents are shown in Figure 5 along with a trendline. The slope of the trendline is -$0.15M/yr slope, which is contrary to the findings from the previous University of Maryland study [Milke, et al., 2019]. In a review of the data, the losses in the incidents that occurred in 2015 to 2017 are significantly larger than those in other years, while losses in 2019 are unusually low.

A comparison of annual monetary damage in DoD versus commercial hangars is presented in Figure 7. The reason for the significant difference in trendlines for the two sets of hangars is unknown.

figure omitted:

An additional explanation for the negative slope trendline in Figures 6 and 7 is presented in Figure 8. The graph in Figure 8 is created by dividing the total amount of damage accumulated by the number of incidents per year that reported damage. While this graph still contains a trendline with a negative slope, the slope is much less than that in Figure 6 and 7.

While this graph is in better agreement to the figure presented in the previous report, it still presents a trendline with a negative slope of -$1720/yr. The graph indicates a low amount of damage in 2019 and a much greater value for 2017. The 2017 data can be explained by a larger number of incidents which reported damage. An average of 44.6% of incidents reported data per year, but the years of 2015, 2016, and 2017 all exceeded this average. The reports of damage from 2013 and 2020 were significantly less than this average. A further depiction of this data is presented in Table 4 which includes the total amount of reported incidents, the number of incidents that reported damage, and the percentage of incidents that reported damage. The reason for the variation in the frequency in reporting monetary loss from year to year is unknown.

Figure and Table omitted:

There was a single fatality reported and a minimum of 21 injuries reported through these incidents. A graph of these incidents is presented in Figure 9. Very few of the reported incidents reported any injuries. Furthermore, the injuries and fatalities recorded in 2006, 2014, and 2016 were all the result of one incident in each year. The 2019 and 2020 data are best-case scenarios for each of these years, as 2019 and 2020 years had at least two injuries during each year. Each year had two reported incidents that stated that there were injuries in two different incidents. However, these incidents did not state how many injuries occurred during these incidents.

Figure omitted:

The causes of the accidental discharges are organized into seven categories:

* false detector activation

* suppression system failure/4

* intentional or malicious activation or human error

* error made during inspection, testing, and maintenance

* weather-related causes

* unknown cause, or

* multiple causes.

The causes of the accidental discharges were organized into these categories by the research team based on information provided on the survey form. The cause of accidental foam system discharges was reported in most responses. The distribution of causes is presented in Figures 10 and 11 and Table 5.

Comparing the causes in the DoD incidents versus those with commercial airlines, "suppression system failure" was the most frequently cited cause for the DoD facilities at 40% of the incidents, while that cause was noted only in 15% of the incidents for commercial airlines. For incidents reported by commercial airlines, the leading cause of a discharge was for "false detector activation" at 23% of the incidents, while this cause was noted in only 4% of the DoD incidents.

Figure and Table omitted:

The types of systems that were activated is presented in Figures 12 and 13 and Table 6. The type of system was not reported in all incidents, as indicated by the large proportion of "unknown" cases. Furthermore, not all the activations released foam concentrate.

Activations that only released water were also listed in this analysis as an AFFF or High Expansion Foam release. The large proportion of incidents involving AFFF for the incidents in DoD hangars is likely attributed to the policies of the DoD agency which led to the predominance of AFFF systems in those hangars.

Figure omitted:

4. Summary

Requirements for foam fire suppression systems in NFPA 409 were initially justified to provide protection from fires involving fuel spills. However, the occurrence of a fuel spill in a hangar in the U.S. is rare and fires involving such spills even less common. Information in the USCG database indicate only 1 fuel spill occurs each year in the U.S. (no information is available whether any of these spills resulted in a fire). In a survey conducted by the research team to commercial airlines and the US DoD, 242 of 245 incidents of foam discharges that occurred predominantly over a 17 year period occurred despite no fire being present. The discharges without a fire, i.e. "accidental discharges" comprised 98.8% of the total number of discharges. In one of the three incidents that included a foam discharge due to a fire, only one is confirmed to include a fuel spill.

References

Milke, J.A., Behera, S., Lee, Kelliann and Slingluff, Caroline, "Review of Foam Fire Suppression System Discharges in Aircraft Hangars," for National Air Transportation Association, Department of Fire Protection Engineering, University of Maryland, November 2019.

NFPA (2016). Standard on Aircraft Hangars. NFPA 409. Quincy, MA: National Fire Protection Association.

USCG (2020). National Response Center, United States Coast Guard. Retrieved from: http://nrc.uscg.mil/.

About the Authors

James Milke, Ph.D., P.E., is the Chair of the Department of Fire Protection Engineering at the University of Maryland. Nicholas Brondum is a graduate student and Colleen Whitcomb and Steven Dunne are undergraduate students, all majoring in Fire Protection Engineering at the University of Maryland. Questions about this report may be directed to Jim Milke at [email protected].

Acknowledgement

Support for this project was provided by Poole Fire Protection, Inc.

Signatures

James Milke, Ph.D., P.E., University of Maryland, Department of Fire Protection Engineering, Professor and Chair; Jack Poole, P.E., FSFPE, Poole Fire Protection, Inc., Principal.

* * *

Footnotes:

1/ A Group II aircraft hangar is classified in NFPA 409 (NFPA 2016) as a hangar with an aircraft access door height of 28 ft or less and a single fire area limited by the type of construction. More

details of the definition of Group II hangars are provided in Section 1.1 of this report.

2/ While the presence of a fire would likely have been noteworthy, no information on the cause of these activations is available. Hence, the reason for discharge, i.e. whether in response to a fire or accidental, cannot be stated definitively.

3/ The cause or nature of the suppression system failures was not identified.

4/ The cause or nature of the failure of a suppression system was not always noted.

* * *

The full white paper, including tables and figures, can be viewed at: https://www.nata.aero/assets/Site_18/files/NFPA%20409/UMD-Poole%20Hangar%20Research%20Report.pdf

TARGETED NEWS SERVICE (founded 2004) features non-partisan 'edited journalism' news briefs and information for news organizations, public policy groups and individuals; as well as 'gathered' public policy information, including news releases, reports, speeches. For more information contact MYRON STRUCK, editor, [email protected], Springfield, Virginia; 703/304-1897; https://targetednews.com