Board Member’s Note: I was given the opportunity to read this article because of my history with naval aviation. As I read it, I was reminded of the CNO’s “Read, Write, Fight” guidance (June 2016 Proceedings). I’m aware the Commandant has given similar guidance to his Marines. I don’t know all the ins and outs with respect to Major Lippert’s suggestions, but we all should respect him for writing. I will be following the discussion it will generate.
—Admiral Tim Keating, U.S. Navy (Retired)
Mishaps cost the Navy and Marine Corps 40 F/A-18s between 2006 and 2016, resulting in 15 fatalities and $2.6 billion in losses. Fourteen of these events were because of controlled flight into terrain (CFIT) or were the result of mid-air collisions. While CFIT and mid-air events accounted for only 35 percent of the incidents, they resulted in 50 percent of the cost, and 73 percent of the loss of life. In 2015 and 2016, five F/A-18s were destroyed and three pilots killed because of these types of mishaps.1 Aviators plan meticulously to avoid both CFIT and mid-airs, and numerous systems have been fielded to decrease their likelihood, yet they continue to be the leading causes of the loss of both life and aircraft in naval aviation.
CFIT mishaps are “those accidents in which an aircraft, under control of the crew, is flown into terrain (or water) with no prior awareness on the part of the crew of the impending disaster.”2 This definition includes cases where pilot response was initiated, but was too late to avoid impact. According to both the Federal Aviation Administration (FAA) and the International Air Transport Association, CFIT is the second leading cause of civilian aviation mishaps, both in the United States and world-wide. The leading cause is loss of control in flight, which has only recently overtaken CFIT.3 Knowledge of the dangers of CFIT is not new. According to the Defense Department’s (DoD) 2004 Aviation Safety Improvement Task Force, the “…historical record of DoD aviation assets demonstrates that CFIT is the leading cause for loss of lives, lost combat capability and dollar cost.”4 CFIT-related mishaps almost always result in the loss of an aircraft, and are typically fatal for the aircrew.
Mid-air collisions are also typically catastrophic. Unlike CFIT, mid-airs, as a cause of personnel and aircraft loss, are a uniquely military problem, given the nature and requirements of military aviation. In 1998, the FAA reported rates of civilian mid-air collisions that were approximately 1/10th the rate in military aviation.5 In 2004, mid-airs were the fifth leading cause of losses throughout the fighter community, and as much as 28 percent of pilots and 22 percent of all aircraft losses across the entire fighter force of the U.S. Air Force, Navy, and Marine Corps combined were the result of CFIT and mid-airs combined. 6
Why Do CFIT and Mid-airs Still Happen?
Significant research has been devoted to the root causes of CFIT and means to mitigate the risk. Rooted in the limitations of the human pilot, CFIT is commonly a by-product of spatial or geographic disorientation,7 typically brought about by channelized attention which results in a catastrophic breakdown in a pilot’s situational awareness (SA).8 Channelized attention can be experienced by one, several or all aircrew on an airplane, and is an oft-cited factor in mid-air collisions.9 In response, aircraft systems have been developed to alert aircrew of an impending threat, with the intent to refocus their attention and rapidly rebuild their SA, enabling the pilot to initiate the appropriate recovery maneuver.
Multiple aircraft systems have been developed over the decades to mitigate the threat of impending CFIT by using visual indications such as caution and/or warning lights or heads-up display (HUD) information, such as a “pull-up” cue or upward-pointing arrow in the HUD. These visual cues often are accompanied by warning tones or aural cues telling the pilot to “pull up!” or warning of “terrain!” Such systems are integrated on modern military aircraft and have names such as low-altitude warning (LAW), ground-proximity warning system (GPWS), terrain avoidance warning system (TAWS), and ground collision avoidance system (GCAS). Historically, simple systems like LAW have relied on a radar altimeter with a pilot selectable altitude at which the warning will sound, indicating the passage of a threshold altitude, at which time the pilot would initiate the necessary recovery to avoid terrain. More complex systems, such as GPWS or TAWS, rely on an aircraft’s inertial navigation system coupled with a global positioning system and either radar -derived height-above-terrain or highly detailed digital terrain elevation data programmed into the aircraft’s mission computer. These systems provide a wide range of different cues, ranging from those as simple as the LAW, to HUD indications that provide precise guidance to the pilot on when to initiate a recovery. Some of these systems still require pilot input to set altitude thresholds for the system to function.
As with CFIT, similar efforts have been made to warn aircrew of the threat of mid-air collision because of nearby traffic, though such systems are not nearly as prolific as those that address CFIT. The industry standard is the Traffic Collision Avoidance System (TCAS), which provides timely and accurate range, azimuth, and altitude information of nearby traffic to the aircrew. Aircraft equipped with TCAS have an altitude reporting transponder, and the system gives various audio and visual indications to the aircrew. While TCAS is required for civilian applications, there is no similar requirement on U.S. Navy and Marine Corps tactical aircraft.
With such technology available, or already built into our aircraft, why do losses from CFIT and mid-airs still occur? There is little doubt these systems have decreased the loss rates to CFIT and mid-airs. For the case of TAWS in the F/A-18, the years following its fielding showed a marked decrease in CFIT incidents—none for four years.10 Yet incidents still occur often enough to be the leading cause of military aviation mishaps that result in losses of vital airframes and highly trained personnel. CFIT and mid-air mishaps still occur because the prevention systems currently fielded rely on pilot response to a system indication of a requirement for action.
For a pilot to respond effectively, three steps are required. First, the pilot must recognize the indication and conceive a correct response (reaction). Next, the pilot must translate the correct response into motive force (motor response). Finally, the motive force must be translated to the appropriate control input (such as a stick, yoke, or throttle). Research has shown the reaction step takes on average 230 to 250 milliseconds.11 In this first step, one or more cues are responded to and perceived in the brain. The second and third steps together vary in duration based on the complexity of the motion required and the amount of dead band in the aircraft’s control system. Dead band is the range of motion of the control where any movement does not result in an effect on the aircraft’s control surfaces, much like the first inch or so of the clutch in most cars. The total time from pilot recognition to aircraft response/motion is additive, and pilot response time in safe-escape assumptions for low-altitude training is 750 milliseconds, allowing 500 milliseconds for motor response to traverse the stick through any control dead band and initiate aircraft motion. Pilot response time could be further increased by the hazards of channelized attention or loss of SA because of complexities of aircraft operation and the high workload of tactical flying.
In that one-quarter of a second required for a pilot to recognize a recovery indication, an aircraft traveling at 480 knots in a 30-degree dive (as in a typical dive-bombing pattern) loses more than 100 feet of altitude. More than 300 feet are lost by the time the aircraft has begun maneuvering. For two fighters flying head-on at 450 knots (typical at the start of a dogfight within the visual arena), one-quarter of a second collapses the range of the two fighters by more than 350 feet, and in three-quarters of a second, nearly a quarter mile.
CFIT and Mid-Air Collisions
Class A mishaps are those which cost the U.S. Government a minimum of $2,000,000, or result in loss of life or permanent total disability.12 According to the U.S. Naval Safety Center, since 1985, the Navy and Marine Corps have experienced 1,172 class A mishaps. Of those class A mishaps, 193 occurred in the F/A-18 community across all variants, resulting in the loss of 185 aircraft and 86 aircrew in 31 years. CFIT caused the loss of 34 aircraft and 26 fatalities, while midairs claimed 38 aircraft and 28 lives. In all, 40 percent of aircraft lost and 63 percent of lives lost were because of either CFIT or mid-air, which is well above the 2004 DoD-wide averages of 22 percent and 28 percent, respectively.13
Mishap rates are often represented in terms of number of losses or events for a given number of hours flown by the entire fleet of aircraft. For legacy-model F/A-18s (A through D models), the average annual loss rates for CFIT and mid-air were approximately 1 aircraft per 160,000 flight hours, and 1 aircraft per 230,000 flight hours. The combined rate is 1 aircraft lost per 95,000 flight hours. If the current trend continues for the legacy model Hornets, and the fleet flies an average of 30,000 flight hours a year for 7 years, at least two aircraft will be lost to CFIT and/or mid-air.
In the Super Hornet family (E and F models), the average annual loss rates for CFIT were far lower, with only one CFIT reported in the program’s history, after more than 1.2 million flight hours. The mid-air rate, however, was a staggering 1 aircraft for every 128,000 flight hours—for a community that has flown 129,000 hours per year or more for the past 6 years. If the Super Hornet fleet continues to fly the same number of hours annually, at least one aircraft will be lost per year, on average, due to a mid-air.
Further consideration for all variants of Hornets rests in the number of physiological episodes which have resulted in loss of aircraft, such as g-induced loss of consciousness (G-LOC), hypoxia, and spatial disorientation. Though such events do not meet the criteria for CFIT, they can result in the loss of an aircraft because of pilot incapacitation. Since 1985, there have been 14 F/A-18 mishaps classified as the result of physiological episodes, all of which have been fatal. Though not technically CFIT, two incidents that were reported as clear events of G-LOC, where pilot incapacitation resulted in aircraft and personnel loss, were included in the above numbers.
Automated Collision Avoidance Systems
Automated recovery systems have been in development for more than three decades. In the United States, they were tested first by NASA, and later by the Air Force, and have taken a variety of forms in their development. In an effort to address CFIT, the Air Force developed a system from a NASA test program that became what is now known as Automated Ground Collision Avoidance System (AGCAS, or Auto-GCAS). Automated recovery systems currently are fielded in various forms in the F-16 and F-22 and are planned for future integration in the F-35. AGCAS uses similar terrain elevation data as previously mentioned, but instead of waiting for pilot input, it provides an automatic recovery to a wings-level, safe-flying condition.
This automatic recovery is programmed to happen at an altitude far lower than a pilot’s typical planned pull-out altitude, thus reducing the potential for false pull-ups. Other variations of the automated dive recovery have used a “line in the sky” system, that initiates recovery based on a pilot-programmed barometric altitude. Systems such as the line in the sky are useful for minimizing CFIT, but could also mitigate the risk of mid-airs in congested airspace where aircraft are assigned altitude blocks.
In the Air Force, additional development efforts are under way to implement an Automated Airborne Collision Avoidance System (AACAS or Auto-ACAS) which relies on own-ship and traffic position information, transmitted via data-link, to command an automatic avoidance response (typically a push-over or pull-up). Each of these systems is designed to be activated only at the last possible instant, well after a pilot would plan to recover. Eventually, Auto-ACAS is expected to be integrated with the currently fielded Auto-GCAS, resulting in an Automated Integrated Collision Avoidance System (Auto-ICAS).14
Cost-Benefit Analysis
In 2003, Defense Secretary Donald Rumsfeld set a goal to reduce the DoD mishap rate by 50 percent.15/16 As a result of research motivated by that goal, in 2006 the DoD Safety Oversight Counsel released the “Fighter/Attack Automatic Collision Avoidance Systems Business Case,” which outlined a detailed cost-benefit analysis for bringing automated collision avoidance systems to the four fly-by-wire fighters of the time, the F-22, F-35, F-16, and F/A-18 based on historical mishap rates and forecast lifetime flight hours remaining. For the F-16, the forecast return on investment (ROI), or ratio of dollars saved to dollars spent, was only 1.7 to 1 for an integrated CAS system over 14 years.17 For Auto-GCAS alone, Air Force Research Labs reported a current return on investment of 3 to 1, based on $60 million in development and fielding costs and four recorded “saves” of aircraft and pilots since the system was fielded in 2014 (assuming an F-16 value of $45 million).18 For the F/A-18, the forecast ROI for Auto-GCAS alone was 7 to 1 based on a development and fielding cost of $91.8 million.19 At the time of that report’s writing (2006), TAWS was being fielded in the F/A-18 community. Since 2014, five F/A-18s have been destroyed by CFIT (four legacy Hornets and one Super Hornet), resulting in the death of five aviators. This high rate of CFIT does not mean TAWS did not have an impact.
According to Naval Safety Center data, CFIT events dropped in the years immediately following the fielding of TAWS. Unfortunately, the fleet ten-year average loss rate since the system has been fielded is still 0.42 aircraft per 100,000 flight hours, down slightly from the CFIT loss rate of 0.54 aircraft per 100,000 flight hours for the entire lifespan of the F/A-18. These rates are based on the assumption of CFIT or very clear G-LOC cases as well, and there is great potential that more mishaps could have been prevented by an A-GCAS system.
The 2006 DoD study did not include ACAS cost analysis for the F/A-18. However, assuming the same $500 million as was forecast to field an Auto-ICAS system in the F-16, an 80 percent effective system would yield an overall ROI of 1.1 to 1 in seven years at the current mishap rate based on loss of aircraft alone. Auto-ICAS should save ten aircraft in seven years, based on 30,000 and 130,000 flight hours per year for the Hornet and Super Hornet, respectively, at a cost of $35.1 and $77 million dollars per aircraft. These values for aircraft are conservative and do not account for inflation. Furthermore, the cost of trained aircrew, valued at $1.1 million dollars by the Naval Safety Center since the 1980s, can typically be as much as $9 million or more, per aircrew.20 As this estimate also does not consider costs of other damages, recovery of aircraft, or investigations, the Navy and Marine Corps likely would see a still greater ROI.
Preserving Navy and Marine Corps Assets
The Sea Services must make sound investments to preserve limited assets and highly trained personnel. For the F/A-18 community, the threats of CFIT and mid-air collisions could be mitigated by leveraging current technologies that are fielded in civilian and Air Force aircraft. Installing Auto-ICAS across the F/A-18 fleet would save lives, aircraft, and money. Furthermore, the proliferation of unmanned aircraft systems throughout the battlespace and their ongoing integration with manned aircraft beg the development and deployment of Auto-ACAS on each unmanned aircraft to further mitigate the risk of mid-air collisions.
With the potential to pay for itself in less than a decade, the development of Auto-ICAS for the F/A-18 should be pursued in the interest of asset and force preservation. More importantly it would help preserve our nation’s greatest asset: the lives of our fighting men and women.
1. Naval Safety Center Data, Released 28 September 2016
2. Previc, F. H., Ercoline, W. R., Spatial Disorientation in Aviation, AIAA Publishing, Reston, VA, 2004; 3.
3. International Air Transport Association, Controlled Flight into Terrain Accident Analysis Report, IATA, Montreal, 2015, 6.
4. Defense Safety Oversight Council Aviation Safety Improvements Task Force Safety Technology Working Group, Fighter/Attack Automatic Collision Avoidance Systems Business Case, 2006, 3.
5. Taneja, N., and Wiegmann, D., Analysis of Mid-Air Collisions in Civil Aviation, Proceedings of the 45th Annual Meeting of the Human Factors and Ergonomics Society. Santa Monica, CA, 2001, 1.
6. Defense Safety Oversight Council, 6.
7. Previc and Ercoline, 4.
8. Moroze, M. L., and Snow M.P., Causes and Remedies of Controlled Flight into Terrain in Military and Civil Aviation, Air Force Research Laboratory, 1999, 6.
9. Taneja and Wiegmann, 3.
10. Naval Safety Center Data, Released 28 September 2016
11. Kohl, P, et al., A Comparitive Study of Visual Performance in Jet Fighter Pilots and Non-Pilots, Journal of Behavioral Optometry, Vol 2., 1991, Number 5, 124-125.
12. Naval Safety Center Data, Released 28 September 2016
13. Defense Safety Oversight Council, 6.
14. Personnel Risk Reduction, Office of the Executive Director for Force Resiliency, Fighter/Attack Automatic Collision Avoidance Systems Business Case, 2016, 1.
15. Return on Investment data are estimates based on historical data and assumptions from other programs of record and are not the result of any current program data. There are no current or recent programs of record relevant to the F/A-18 to provide actual cost-benefit analysis, thus these numbers are estimates and have been published without approval from PMA-265.
16. Defense Safety Oversight Council, 3
17. Ibid., 20
18. Air Force Research Lab Data, released 30 September 2016
19. Defense Safety Oversight Council, 20
20. GAO Report, Actions Needed to Better Define Pilot Requirements and Promote Retention, GAO/NSIAD-99-211, 1999, 18
Author’s Note: The Author would like to acknowledge: LT Amanda Lippert, LCDR Eric Zilberman, and CDR Kevin Sproge, U.S. Navy, Major Matthew DeCoursey, U.S. Marine Corps, and Mr. Mike Wallace for their contributions to this article.