Drone Losses Debate

I got an interesting response on my post last week about the Air Force’s 2010 OCO request for $216 million to buy 12 MQ-9 Reaper drones.

Air Force officials say the money is needed to replace lost or damaged Reapers from combat operations.

In my post, I noted that at least the intention was demonstrated in the request that the USAF would rather lose almost a squadron of MQ-9s in combat rather than one F-15 and its crew.

Well, my good friend and longtime Pentagon budget watchdog Winslow Wheeler pinged me with this rejoinder:

I think you are missing an important point in your comment about the 12 drone losses replacing aircraft/pilot losses.  I strongly suspect drone accident rates per 1,000 flying hours are well above, multiples, of aircraft accident loss rates per 1,000 hours. As for hostile losses, drones are so ridiculously easy for any modern (I.e. non-Taliban) air defense to deal with that I suspect, if ever we meet same, they will be quickly attrited.  Are there any drone losses to hostile fire in Afghanistan?  If there is any such number, it compares to zero (I believe) for aircraft.  Wheeler out.

 Well, I did a little research on the first argument, and here’s what I found. Wheeler has a point in that the lifetime Class A ($1 million in damage or death) mishap rate for the Predator/Reaper — as of December 2009 — was “multiples” above that of, say, the F-15 fleet. It takes a little finessing, but combining the lifetime totals of flight hours for the RQ-1 Predator (which begins in 1997) and the MQ-9 Reaper (which starts in 2004), we get a Class A mishap rate of 10.2 per 1,000 flight hours. [CLARIFICATION: The services’ safety centers canlculate mishap “rates” per 100,000 flight hours, typically. But I made my calculations based on Winslow’s 1,000 hour benchmark. Running the numbers, the Predator/Reaper official mishap rate would be 9.7 per 100K flight hours — still very high] The Air Force says it lost a total of 57 Predators since 1997 and seven Reapers. Both aircraft have flown a total of nearly 655,000 flight hours.

Looking at the F-15 rate, USAF stats show over the lifetime (since 1972), the F-15 platform has a Class A mishap rate of 2.42, with 140 aircraft damaged. It’s lifetime destroyed rate is 2.04 with 118 aircraft lost — and that’s over a lifetime total of almost 6 million flight hours. But the stat that 43 pilots have died behind the stick of an F-15 and two of those were killed in fiscal 2009, speaks volumes to the family and loved ones of the fallen. Despite the high mishap rate of the MQ-9, no pilots are dead because of it.

But, yes, the Predator/Reaper mishap rate is more than five times that of the F-15. 

Now on the shoot down issue, I just can’t weigh in. I’ll look into how many of those purported losses in 2009 were from shoot downs or malfunctions. But I don’t think it’s “ridiculously easy” to shoot down a Predator/Reaper. Small target, very high and relatively quiet when they’re up there…But I just don’t have any info on that yet.

I’ll post Winslow’s response when I get it.

— Christian

  • Todd855

    I think that is the whole point of drones/robots. To lose dozens if not hundreds and thousands, without any political backlash. US soldiers/airman and time is never on the US side in conflicts, so tech is only real advantage we have. If we did somehow face a modern opponent, the loss rates for would be substantial for manned and unmanned systems, but unmanned mean nothing politically. Only production rates are the issue, and that is something we know how to master. Imagine relentless waves of air and ground drones.

    I would rather have 10k Reapers instead of 2k F35s, and so would Gates.

  • Mike SNow

    Another way to compare (at least the machines, not the humans of course), is how much did the machines cost. Even at the higher loss rate are the drones in fact cheaper to use than the manned fighters?

    On the loss rate against a sophisticated air defense system, aren’t we going to eliminate that system before we let any large numbers of our machines into the space (exp. Iraq?).

    One last point, we’ve been flying manned fighters on a large scale since WWII (70+ years), drones for 10-15 years. I would think we will get a lot better at using the technology over the next 10-20 years.

  • Marcase

    The RQ-1/MQ-9 isn’t exactly stealthy, and can be detected by modern radar. And as a Russian Fulcrum near Georgia once proved, UAVs are very easy to shoot down.

  • Falcon

    What he said was “drones are so ridiculously easy for any modern (I.e. non-Taliban) air defense to deal with”.

    He’s not talking about losses in Afghanistan with towel-headed Arabs using Kentucky windage, he’s saying that a modern air defense system could shoot down drones all day long.

    • Boogel

      Yeah, I’m sure that’s what he was saying. The drones are only effective against an adversary without an air force. Any nation that has, say the industrial level of the west in the 1920s would be able to neutralize the drones.
      What I would like to see is an example of a nation using these against a more equal enemy. I’m not sure it’s possible. But I would like to at least see someone try it.

  • ohwilleke

    The morality rate for F-15 crashes is laudably low as a percentage of crashes.
    The rate of fatal F-15 crashes is still about 1 per 2000 flight hours, however. Obviously, one doesn’t fly 40 flight hours a week and an F-15 wasn’t designed was safety as the exclusive design consideration.

    Still, if driving were that dangerous per hour of operation, the lion’s share of the nation’s truck drivers would be dead in a few years.

    • TMB

      What were the loss rates for the F-15 and F-16 during their first few years of operation? We’re comparing flight hours of a plane that’s been around for 30 years vs one that’s been around for 10. I seem to recall a fair number of F-16s falling out of the sky during its first few years due to wiring malfunctions and other teething issues.

  • BEW

    I seem to remember reading that the UAV’s are provided with and without an Automated Landing System. The Air Force insists on buying them without the ALS; the Marines and Army insists on buying them with the ALS. I suspect the CIA buys them with ALS, but who knows. The story went on to say that the AF was having a higher crash rate than the other services. It would be interesting to find out how the services compare.

    • Boogel

      Would the ALS change how the Air Force categorizes the aircraft? What I mean is if a AF pilot was flying a drone that had ALS would he get the same flight hours credit as if he were flying one without. I’ve read that there is a cultural resistance to the drones within the AF which has made it harder to recruit pilots for the program. Just curious if this is related.

  • CJ-

    The UAV figure is misleading because the Predator can't land for crap. Aren't the vast majority of Predator losses because of bad landings? It's a design weakness. How many F-15s where lost on landing? The focus should be operational losses (shoot downs, engine failures, structural failures in flight, etc). Crashing because the Predator can't handle a crosswind on landing is only a transient issue. How many Reaper's have crashed on landing?

    If the Predator landing losses are removed from the data and only operational losses are taken into consideration, the UAVs would likely come on top in a big way.

  • kim

    Towelheaded Arabs live in The Middle East, not in Afghanistan.

  • Mack

    I think something that is being criticially overlooked in this debate is the cost and further risk of life that would be associated with the necessary CSAR operations that would be conducted if the drones lost were manned aircraft instead. I haven’t even attempted to crunch the number of man hours, dollars, and combat action that would result if the 12 Reapers lost in action were instead manned platforms which would necessitate atleast a minimum CSAR effort.

    • Rescueman

      It doesn’t matter to us, we get Soldiers, Marines, SOF, allied military, even USAID workers during COIN ops like those in OEF. . . in addition to the Predator parts we pick up when they smack in. During conventional ops, we get those downed aircrew from the manned aircraft currently required to penetrate high threats (if we can, without CSAR-X, that will be alot more difficult).

  • Darthamerica

    It’s not easy to shot down drones. Against a more capable opponent drones won’t operate in isolation. Like manned aircraft, C-130’s for instance, drones will operated in sanitized airspace or in conjunction with other assets that are designed to make sure they can operate with acceptable loss rates. Also, RQ-170 shows that the DoD/CIA are well aware of survivability issues and almost certainly have means to operate unmanned aircraft in contested airspace.


  • Leroy Hurt

    Not to mention: as unmanned systems proliferate, there’ll be the accompanying tightening of accountability as commanders start to pay more attention to their budgets. That will mean a corresponding increased focus on accident-free operations.

  • DarthAmerica

    Also consider that industry is marketing much more high end UCAVs for introduction in the next decade. THings similar to X-45/47 and even proposals to have unmanned versions of aircraft like F-35. Heck, the DoD already operates QF variants of F-4 and F-16 ect with very high reliability. UAV technologies will not stay stagnant. Moreover, you have to consider what Reaper and Predator are designed to do. Cost is a huge factor and primary benefit. They don’t have the same level of systems redundancy and reliability built in as manned aircraft do ON PURPOSE. Consider much more sophisticated UAVs like Global Hawk which have much lower loss rates.


  • Dientboy

    A few things to consider:
    – AF requires manual landings vice automated; bricks Predators/Reapers on landing at three times the rate of the Army (or so I have heard).

    – It is not a disaster to crash a UAV, so we use them differently. In extreme situations (UAV has eyes on a high-value target; no other birds are around), we have flown UAVs to beyond bingo fuel and crash-landed them because we believed the value of the target exceeded the value of the UAV. This would never happen for a manned aircraft.

    – We devote far more maintenance hours per flight hour to manned aircraft, build in more redundancy, etc (along the lines of DarthAmerica).

    – @MCQknight on Predators crashing into mountains: I’ve never flown a Predator, but the USMC’s Raven B (which I have flown) — a throw-away UAV with 50 minute endurance — can be programmed to avoid mountains even when it doesn’t have comms with the base station. I really doubt that’s the issue.

    Overall, think the stats are a combination of (a) technology and employment TTPs are relatively immature and (b) we don’t care as much about losing UAVs, so we take more risks with them.

  • grumpy

    The loss rate has not changed significantly since the 1980’s and I really don’t see it changing much in the outyears.

  • CreasePanther17

    In the latest Popular Mechanics, it discusses the problem of landing a Reaper and how they are going about to fix it. Most of the Reaper losses are from malfunctions and landing gone wrong from oscillation.

  • Carl

    Are you comparing an F-15 mission to a UAV? I’d rather have 10 F-15s to fight with than 100 UAVs.

    • Rescueman

      Really? In OEF, with 100 Preds, I could support nearly every battalion CC and sometimes down to individual companies engaged in combat. You’d want 10 F-15s rather than that? I assume you mean F-15E’s cause C models would just be able to make noise. 10 F-15E’s could generate about 12 two-ship orbits of 4 hrs each over 24 hour period, with trip turning the jets. 100 Preds, could like produce at least 90 18-24-hour orbits over the same 24 hour period. Each F-15E could carry alot more bombs which likely aren’t needed, since one is usually plenty in a COIN or CT scenario.

      Come on dude, I agree that the OSD led destruction of AF manned conventional air power isn’t good for the US, but we need to be able to do both, not one or the other – when the situation suits each.

  • Valcan

    The predator and reaper drones are in there earliest stages. Like other posters have said the technology is still maturing. and its used differently from other aircraft.

    I think in someways the UACV’s the navy is working on are more of what we may see in the future. I figure well see not robot army vs ronbot army but what we have today. UCAV’s and unmanned vehicles supporting human troops and vehicles tripling there firepower and survivability.

  • Chris

    Consider that what eventually crippled the Luftwaffe and Japanese naval air was not a lack of machines, but the attrition of good pilots.

  • Drake1

    The Predator doesn’t have an auto landing system like the Army’s Warrior. I’m more interested to see how those Warrior accident rates pan out in a couple years.

    Near peers would likely be able to shoot down many UASs, but those systems liklely wouldn’t be the first in. F22s are the new F117 that soften up the defenses so that legacy aircraft and UASs can operate later.

  • Mike

    Why not rig them so they can be destroyed by the operator if they are lost for any reason to prevent valuable technology from the enemy?

  • DarthAmerica

    The facts are that comparing UAS lose rates to manned platforms is like apples and oranges.

    a. Different design philosophy
    b. Different TTPs/CONOPS
    c. From the system level perspective immature technology just getting out of dev
    d. designed with post cold war threat matrix in mind

  • guarddog

    So much discussion from so few facts and no first hand knowledge of the subject.

  • Jon

    Some maths for ohwilleke:

    The F-15 loss rate over life is one every 50,847 hours and 24 minutes.
    The fatal loss rate is one every 139,534 flying hours.

    The Predator/Reaper loss rate is one every 10,234 hours.

    UAVs have a role to play, but it's a niche role. Sensor limitations mean that they will never have the situational awareness you get with a man in the cockpit, even when other weaknesses (slow speed, bandwidth consumption, etc.) have been solved.

    • The UAV’s are still fairly new. Technology gets more advanced everyday. Just wait and see. Soon UAV’s will out perform the best manned aircraft, and be 10x cheaper after you add up the aircraft, pilot training, etc. And we wouldn’t have to worry about sending them into harms way when theirs no one on board the craft.

  • nicolas

    “It’s lifetime destroyed rate is 2.04 with 118 aircraft lost — and that’s over a lifetime total of almost 6 million flight hours.”

    I dont understand : are you talking of a rate (loss/time unit), or of actual losses ?
    If you talk about rate, why do you mention its over “6million flight hours”…. since its a rate, it’s not supposed to go up with the length.

    the whole thing seems so superficial, who has time for this ?

  • CK757

    Remote controlled aircraft have had and logically will have a higher loss rate than manned aircraft. At least that's the history.

    Today's (whatever you call them) remote controlled aircraft are still fairly primitive by manned military aircraft standards. As “computing power” is developed/refined, and available to meet military requirements for reliability and environment, things will change.

    A pilot in a manned aircraft is still better able to respond to “unplanned” or “unexpected” events than is a remote controlled aircraft.

    It's still cheaper to replace an unmanned aircraft than a manned aircraft and it's pilot/crew. Not to mention the replacement pilot/crew time to train, etc.

    While the military has been “fooling around” with remote controlled aircraft since WWII, it's only in recent times that serious effort and resources have been utilized. We are a the position with remote controlled aircraft that in some ways is similar to that of jet aircraft in the early 1950's.

    The past reluctance of the air force and military to commit major efforts towards development and common use of remote controlled aircraft was mainly based upon a preference for manned aircraft. The loss of Joe Kennedy's son in a experimental remote control aircraft program towards the end of WWII was an early setback that had far reaching and long lasting effects.

    It's interesting to note that, outside of the military, and government sponsored R&D efforts, etc., use of remote controlled aircraft beyond the operator's “line of sight” is against the law in the US. Further restrictions include a weight limit of 55 pounds without government permit/license.

  • This website certainly has all of the information and facts I
    needed about this subject and didn’t know who to ask.

  • What’s up, I check your blog on a regular
    basis. Your writing style is awesome, keep doing what you’re

  • Howdy! I understand this is somewhat off-topic however
    I needed to ask. Does managing a well-established blog like yours require a
    lot of work? I am brand new to blogging but I do write
    in my diary daily. I’d like to start a blog so I can share
    my experience and views online. Please let me know if you have any kind of recommendations or
    tips for new aspiring blog owners. Appreciate it!

  • I’m not that much of a internet reader to be honest but
    your blogs really nice, keep it up! I’ll go ahead and bookmark your site to come
    back down the road. Many thanks

  • Thank you a bunch for sharing this with all people you actually know what you are talking about!
    Bookmarked. Please also talk over with my site =).
    We may have a link change agreement among us

  • Wow, amazing blog layout! How long have you been blogging
    for? you make blogging look easy. The overall look of your site is wonderful,
    as well as the content!

  • What a stuff of un-ambiguity and preserveness of valuable know-how concerning unpredicted feelings.