So this is how it is... F22 program officially killed...

[Spidey]

Duper…assuming a company is only working on only one product at a time. (don’t forget I’m in business and have a good idea how these things work)

Snoopy…I haven’t over estimated CAD…it’s only another tool that’s all, not the complete system.

Modularity is a good idea, a completely new craft every 10 years is not necessary. As a matter of fact, I was thinking along those lines. It would take the “A B C” versions to the next level.

I do believe the F22's electronics are in fact modular.

Good thinking…I have used modularity in my life many times…my security system (home built from scratch) and game controllers are 2 examples.

Modularity and “Evolution” you don’t have to develop a new fighter every 10 years, just a bigger and badder version, but once you produce a whole slew of them…you are stuck with them. So you build and develop…rinse and repeat.

Krom has a valid point, and would have to be worked on. But since a lot of the cost in these matters are political and whatnot, there could be a solution.

[Drakona]

Also, never underestimate the ability to restart production, of whatever in a time of war.

It doesn't work that way. It's not as though the blueprints get put on a shelf and can be pulled out again at any later date. Any given piece of techology exists within a technological ecosystem that changes over time.

For example, a lot of games I used to be able to play don't run on modern hardware. Even though I have the programs untouched and unchanged, even if I could acquire the source code and recompile them on modern systems, they don't work. They depend on hardware I can't buy anymore, sometimes on software I can't get anymore. They're dead. To resurrect them, the easiest route is either to write an emulator for the hardware or just reverse-engineer the software. It sure isn't to find all the pieces necessary to get the old system working.

Specialized techology has that problem, only on a whole other level.

Technical debt can accrue at an astonishing rate, even on a living, breathing program. Your hardware can go from being cutting edge to being hard to find to being not available, all before you go into full production. Sometime during that process, while you still have all the engineers who understand how everything fits together, you migrate to a new solution. It happens with hardware in the software world. It happens with languages in the software world. It happens with materials and processes and engineering techniques and all manner of stuff.

Now imagine something that's been sitting on the shelf for ten years unmaintained. The computers it used to run on cannot be found anywhere. The system you have doesn't boot right and the one guy who used to know how to make it work (whose fuzzy notes you can't quite follow) has since become a missionary and lives in Papua New Guinea. The software team has been scattered to the four winds; together, ten years ago, they would have found it a challenge to move the software to a new platform. Today you cannot even find software engineers who will work in that language. At all.

The folks who made the parts have gone on to other things. The processes were similarly shelved, and are similarly hard to rehabilitate. Their experts work on other things now, or are gone. The tactical world has been marching along as fast as the technological world (faster, maybe), and the system needs some tweaks to handle the new battlefield environment well. The old specs are full of cryptic requirements, stating tolerences on parts that your current crop of engineers are scratching their heads over. Are they nonsense? Defunct? A result of subtle testing or obsolete standards? The folks who could tell you are long gone. If you're lucky, they left a name on the documents. A manufacturing process can't be performed anymore, at least not cheaply; will a modern equivalent work? The specs were written before it existed, the parts untested. No one can tell you if it'll work.

See, these techologies are old before they're even born. That's the nature of the business. They're always in deep technical debt during production. Trying to rebuild them after shelving for a while is like trying to repair an old car or rehabilitate old software. Only on a massive scale.

That's really what this was about. Early this year was supposed to be when industry needed to order long lead parts for the next batch. If you don't order them, the folks who make them stop. And then it becomes a shelved technology problem. The government bought a little time for the decision by ordering some spare parts and just a couple more . . . but that only buys you a little time. We're at the crucial "buy more or end the program" juncture.

I’m pretty sure the Air Force only ordered 183 of these badboys anyway, so I can’t see the fuss.

That isn't so. The initial order was for over 300, and for a long time folks were saying the minimum number they needed was 381. That number isn't pulled out of thin air.

The Air Force states a need for one squadron of 24 F-22A aircraft for each of the 10 Air Expeditionary Forces, the planned organization of the Air Force aircraft and personnel for operations and deployments. This requirement was established to carry out missions including support in major regional conflicts, home land security, and others. According to the Air Force, this requires a total of 381 F-22As, 240 primary aircraft and 141 aircraft for training, attrition, and to allow for periodic aircraft depot maintenance. The Air Force stated that if all 381 aircraft are acquired, the Air Force could retire about 566 legacy aircraft; if not, several billions of modification dollars will be required to extend their structural life to keep them operational.

I don't know what changed with those numbers. I'm kind of curiously waiting the next QDR to see where they're going with this. On the one hand, it could just be a manifestation of the tendency of the DOD to focus too heavily on the problem at hand--COIN between major wars. And that may not be a bad thing; it avoids wasting money on problems you're not actively solving, but it does have a strategic cost. On the other hand, they've been talking about going to an all- or mostly- drones fleet for a while now; a lot of folks think Raptor will be the last and greatest of the manned fighters. (JSF so totally does not count.) So maybe they know something I don't.

I'm confused. Why do we need a large number of expensive F-22 Raptors when terrorists don't have fighters of their own? Wouldn't the money be better spent on more mobile Special Ops Forces and better intelligence gathering? Last I looked, modern warfare has to deal with a new type of foe that is spread out worldwide and hard to find.

We could conceivably be at war with a full-fledged world power, if things go the wrong way with, say, Iran (or North Korea). Some might not think it's probable, but it's certainly in the realm of possibility.

Indeed. The surest way to achieve peace is to be ready for war. Most of us have grown up in a world with a single non-aggressive superpower, which is naturally peaceful. But you only have to take away a little bit of the 'power' part before war becomes the natural state. I like peace through strength. I think it's a relatively efficient and bloodless strategy.

Raptor's good at it, by the way. It's by far the scariest thing in the sky, an extremely destabilizing force. Nobody, nobody, nobody wants to fight it. Just sitting on the tarmac, it prevents war.

we need to build smaller fleets more regularly.

Say…100 or so every 10 years, keeping the fleets up to date & leading edge

. . . 8 to 10 years seems doable. (I would bet on less)

So every decade you could have a new fighter in production, instead of waiting for the entire fleet to become obsolete, then starting again.

Acura used these techniques to design their new P1...pretty spiffy.

Hmm. Well, a car is not a fighter jet. I know the auto industry is supposed to be the cutting edge place for systems engineering right now, but the problems are just not the same complexity. I mean, Agile folks will tell you that a lot of web technologies are updated on a monthly cycle, and for projects like that, I'm sure it works well. That doesn't mean you can release a new operating system once a month if you follow their methodologies.

It really depends on the scale of the project. For little UAVs, ten years might be realisitc from cradle to grave. But "build lots of crappy aircraft" is the Soviet way. "Build a few invincible platforms" is more the American philosophy. And for something that's not a toy -- some highly integrated avionics project like, say, AWACS -- ten years would be crazy ambitious.

Let me put it to you this way: my favorite program was a little DARPA project--nontrivial, but still something folks called a "go-cart of an aircraft". It was a dream-team of engineers and experts, together with great corporate support and systems enginering, and an extremely aggressive schedule. We were trying to get techology to market in a certain window, and everyone from executive management through the guys on the floor believed in the urgency of it. I really don't think it could have happened faster. Six years in, we were wrapping up prototype A, and midway through the design of prototype B. It'd have been another four or five before we could do full scale production.

A medium system and a dream team would've missed your ten year mark. I'm not saying it couldn't be done in a space race kind of scenario. Just that the pace has less to do with engineers dragging their feet and more to do with the fact that the systems are complex and the problems are hard.

At any rate, you don't need to produce new systems on an arbitrary schedule. Sometimes retrofitting an old craft will do you just fine, and it's often cheaper than designing a new one. (The modularity that enables that is already done to the nth degree, if you ask me--Mil standards, COTS technology, line replaceable units. Or, look at something like JTRS, for a crazy over-the-top example; I think it's too modular to be practical!). So to some degree, we already take that approach where it's sensible, but it's no panacea. The problems are still hard.

I wouldn't muck with arbitrary technical demands for new platforms. Procurement the way it is now, with the exception of political entanglements, works: the proposals are driven by tactical need, folks propose solutions old and new (sometimes a retrofit old system, sometimes a design from scratch), and the solutions are judged on cost and risk and effectiveness. I think that's as it should be. Systems are upgraded when that'll work, and new systems are built when they're the cheapest option that does what you need. The current fifteen to twenty-five year tempo isn't legislated, it's a natural result of looking for solutions to problems. I judge it to be about right given the tradeoff between the complexity of a redesign and the pace at which the techological and tactical worlds march along.

Which kind of comes full circle in this discussion. I don't know whether ending F-22 was a good idea or a bad one. It depends on why they did it and how they plan to fill the role in the future. It's significantly fewer jets than the Air Force had said they wanted, and that concerns me. And I worry that it's a result of folks who are too busy watching counterterrorism stories in Iraq, and not paying enough attention to near peer countries (that being a well-known euphemism for China and sometimes Russia). It's not exactly a reversible decision, and keeping the production line open at a low level isn't that expensive. But then again, the outcry against it has been mostly from politicians who want to keep jobs (a sorry argument for keeping any program), and from pilots who wanted to fly it. Not very credible sources if you ask me, so . . . I don't know. The folks who do that strategic stuff for a living haven't been making a lot of noise, at least not that I see or in public, and I'm not really one to second-guess what's going on there.

[Duper]

The folks who made the parts have gone on to other things. The processes were similarly shelved, and are similarly hard to rehabilitate. Their experts work on other things now, or are gone. The tactical world has been marching along as fast as the technological world (faster, maybe), and the system needs some tweaks to handle the new battlefield environment well. The old specs are full of cryptic requirements, stating tolerences on parts that your current crop of engineers are scratching their heads over. Are they nonsense? Defunct? A result of subtle testing or obsolete standards? The folks who could tell you are long gone. If you're lucky, they left a name on the documents. A manufacturing process can't be performed anymore, at least not cheaply; will a modern equivalent work? The specs were written before it existed, the parts untested. No one can tell you if it'll work.

We run into exactly this problem at work all the time with some of our "legacy" products that are only 8 to 10 years old.

btw, it took several years to write and debug the software that runs the F-22.

HERE is a pretty cool read .. if you can get through it all.

and HERE is a nice article with bib to follow up on for more detail. Drak, you might find some answers there.

Nice post Drak. still digesting it.

Spidey. Cool on the business mngt. how long just outta curiousity. I don't quite get what end you're answering from: "assuming a company is only working on only one product at a time." They will do fine? or They are wasting time and money. ?

You could certainly span ten builds over the course of a year, but Lockheed is a slicker company than that and builds serveral at a time at various stages of construction at any given time.

Where numbers are concerned. In November and December of 2007. 700 F-15's were globally grounded and decommisioned. That's a mighty big hole to fill. The Harrier, which did much of the work in work in the last couple of wars in Iraq is due for retirement soon as well. The F-35 is supposed to help fill this hole. Also the F-117 was also retired last year.

Indeed. The surest way to achieve peace is to be ready for war. Most of us have grown up in a world with a single non-aggressive superpower, which is naturally peaceful. But you only have to take away a little bit of the 'power' part before war becomes the natural state. I like peace through strength. I think it's a relatively efficient and bloodless strategy.

Raptor's good at it, by the way. It's by far the scariest thing in the sky, an extremely destabilizing force. Nobody, nobody, nobody wants to fight it. Just sitting on the tarmac, it prevents war.

Bingo! ... man why do I always seem to dance around what I REALLY want to say only to have You post it later!

[Spidey]

Well Drakona, you are making a lot of points against things I never even had in mind in the first place. I’m not talking about reinventing the wheel every so many years. The idea was just to make smaller production runs, with constant improvement along the way. Retrofitting is fine, but only works up to a point.

I’m not going to nittley pick every point you made, I just don’t have the time right now, or the desire, but I will say…many of your points can be disputed, are not relevant to the way aviation manufactures or are pointed at something I didn’t mean to imply in the first place.

If you think the proper way to go is to produce large numbers of craft and keep them in service over long periods of time is the way to go, that’s fine with me, I just have a different point of view.

[Duper]

The main thrust is that it's not efficient to try to restart a highly complex process back up after mothballing it for \"x\" years.

With any manufacturing process, you normally have a need to \"fill a pipe line\" and after that you can maintain a trickle to keep demands met. The number being built was a bit evasive. I found one figure at a projected 35 planes a year, but I wasn't able to find the actual number.

With 700 planes missing from our arsenal, 35 a year is a modest amount in my estimation. If you figure that 1 F-22 can do the job of about 5 F15's then we would be fine with 183. ... of course how that plays out in REAL life is anyone's guess.

[Spidey]

Well, I’m just glad to have all you experts around to slap down my dumb ideas.

........................................

Timeline…Aircraft Development

Development begins…
6 months in…Pentagon raises specs…4 month delay
Pentagon raises specs again…6 month delay
Congress cuts funding…6 month delay
Pentagon wants different engine…2 year delay
Pentagon raises specs again…6 month delay
Pentagon decides to have a contest to determine who will get to build craft, instead of honoring original contract…2 year delay
Pentagon again raises specs…4 month delay
Congress cuts funding…6 month delay
Prototypes finally ready to fly…bureaucratic BS delays and aggravates contestants…4 month delay
Contest finally starts…2 years to complete
Tests end and production is almost ready…but the Pentagon now wants the craft to land on carriers…2 year delay
Pentagon now wants craft to carry different weapons…6 month delay
Pentagon now wants craft to fly 500ft higher…1 year delay
Pentagon now wants craft to fly faster…6 month delay
Congress again cuts funding…4 month delay
Pentagon wants craft to have 500 mile further range…1 year delay
Congress cuts funding yet again…4 month delay
Pentagon can’t decide what color to paint the prototype………

I have decided to leave this thread with a little humor…but it’s not that far from the truth. So I concede…aircraft development CANNOT be speeded up.

[Drakona]

Heheh.

Apologies if you feel attacked, Spidey. The issue is near and dear to me. I've worked in the relevant industry for a few years now, and have some strong opinions on it. It isn't personal. I'll try to be clearer.

Two things I perceived you to say tweaked me enough to provoke a response.

I perceived you to say that production could be restarted in a time of war, and I wanted to point out that restarting production isn't easy, either in wartime or peace. Crank it up, yes; revive it, no. My experience is that even extremely cutting edge platforms are constantly making use of technology that is either unique to the platform or just generally obsolete. Just maintaining them while they are alive is difficult; I honestly think it'd be easier redesign than resurrect one.

I'm not speaking theoretically about that, by the way. I'm a software engineer who graduated near the beginning of the Java CS 101 age. Lately I've been trying patch bugs in some Fortran77 code. On even older hardware. \"Long development cycle\" and \"uses obscure technology\" seems like such an inadequete description of the reality. And I know exactly how hard this would be for me without help from the greybeards who actually know the system.

I perceived you to say that large platforms should be developed on shorter timescales and in smaller lots, resulting in a more technologically forward approach and perhaps greater agility. I have a couple serious issues with the approach.

First, it's not an inherently efficient approach. A lot of the cost of a given platform is the development, not the production. For example, the Wikipedia article states that for Raptor, the Air Force spent $28 billion on development, and $34 billion buying actual airplanes. You'll hear two different numbers talked about for the cost of the jet; when people say $360 million, they're counting sunk development costs. It's actually $140 million to build a new one at this point. The point is, the development part is extremely expensive, and the cost of that doesn't go down if you build fewer aircraft. If anything, even the production cost goes up in smaller lots. Once they're developed, you should try to get as many as you need and as much use out of them as possible. And it is standard practice to try to get a lot of use out of them, through modernization programs and the like. B-52 is the poster child for that, but in fact lots of programs (and countries) do it. It's just efficient.

Second, I'm not sure it's terribly feasible. Large and complex systems take time to build; that's got to be some sort of engineering conservation law. The history of the software industry in general is one massive landscape of craters left by managers who thought they could change that. I'm not saying efficiency improvements can't happen, just . . . the problem's inherently hard, too. Don't discount that.

Thrid, I perceive the approach to be inherently wasteful because it's pursuing the wrong things for the wrong reasons. Some folks criticize the DoD and the Air Force in particular, saying they develop technology for its own sake independant of how useful it is. Whether or not that's actually what they do, it's a valid criticism. Yet that's exactly what I hear you proposing: build a new platform on a given schedule, develop constantly, pursue technology just to pursue technology. I like the way procurement is now, thank you very much: driven by tactical and strategic need. Ask for a new platform when you have a concrete reason to need one. Ask to upgrade an old one when you have a new mission for it. Don't do it on an arbitrary schedule just to \"be modern.\"

There is a time and place to be pursue technology for its own sake: R&D programs, tech demos, experimental programs. Private industry does it a bit, and DARPA does it a lot. And I'm glad they do, and I think they're good at it. But full-scale production of workhorse platforms is not the place for that kind of thing. Those are for concrete solutions to concrete problems that are clearly a good investments.

I guess what I'm getting at is that the large lots, long development times, and long lifetimes make a lot of sense given the problem space. I think it's a natural approach, and a rhythm that comes about by organically approaching tactical problems from a \"what's cheapest long term\" perspective. I see no reason to impose arbitrary restrictions. There are a few things about the industry I don't like, but that's one of the things that I really think does work.

[Valin Halcyon]

This is all rather amusing. The US HAD to know Obama would try his best to disarm your military.

At any rate, the news media has you Americans so focused on terrorists and terrorism that you forget your traditional adversaries. Remember North Korea? I hope so. They're trying to become the next COMMUNIST nuclear power! Oh, and what about the PRC? They already have nukes, tanks, hundreds of fighters...and not cold war vintage either, such as your F-15 is.

You NEED the F-22, F-35, and whatever else you're not telling the world about. And in that need, you NEED to keep the production lines up, even if at a very low capacity.

What Obama has done, is essentially destroy any chance of the US maintaining air superiority in a traditional war for an extended period of time. Oh, and before you quote the extremely short war that was the first Gulf War...let me point out the SECOND gulf war and the fact that Iraq had next to nothing for an air force to begin with?

Pray China doesn't get a burr up it's collective ass and start shooting at someone.

[Spidey]

How bout you fight the next war instead.

[Duper]

Pray China doesn't get a burr up it's collective ass and start shooting at someone.

Trust me Valin, the general populace hasn't lost sight of this. (most of us thinking types that is) Don't let "our" media (I'm not claiming it ) fool you into thinking that it is OUR opinion. That's why we're jumping up and down waying our arms. Some of us are old enough to remember things like Korea (the First time), Nam, and Panama to realize what Drak (bless her heart) is quite true:

The surest way to achieve peace is to be ready for war. Most of us have grown up in a world with a single non-aggressive superpower, which is naturally peaceful. But you only have to take away a little bit of the 'power' part before war becomes the natural state. I like peace through strength. I think it's a relatively efficient and bloodless strategy.

We ensure that the other big players will think twice before considering a head to head. The 3 engagements I listed were indirect conflicts between the "supers" kinda like Afganistan is.

[Valin Halcyon]

Spidey: I saw enough war for anyone in the former Yugoslavia. Also I doubt you'd give anyone that same advice if you yourself had seen what I have. Go fight the next war indeed...

[Duper]

Valin, I work with a gal that's from Boznia. She told me all about what went down there. o.m.g. Our news never even scratched the surface.
She lost half her family in a single evening when fighting first broke out. She's been here for about 14 years now and is concidering going back home.

[Spidey]

Read between the lines Valin, the comment had to do with you telling us what we should do.

I don’t think 'anybody' should have to fight the next war, but when people assume it should be us…

[Valin Halcyon]

Spidey, heh, what recent big war has not been started by the US? Tell me that. The US government would have the world seeing boogymen under every rock, boogymen that only they can destroy.

Is it any wonder, then, that we expect you to fight your own wars?

Over here in the UK, there's a lot of anti-US sentiment, which is pretty common in Europe in general. Up until the second gulf war, we English got along rather well with the Americans. Now we tar, feather and lynch any politician that even suggests aiding the US in fighting terrorism abroad.


In an attempt to return to the topic...
It's quite simple. Closing down the F-22 manufacturing lines is going to have a seriously adverse effect on American combat ability in the next 20-25 years. The F-15 lines weren't even shut down until a few years back when it was clear the F-22 was to replace them.

It all boils down to Obama and the Democrat way of thinking. If you've studied your US history, EVERY time the Democrats come to power, the following Republican administration has to scramble to repair, restore and modernize the US military. As a programmer, I know a great number of \"shooting yourself in the foot\" jokes. I can come up with more than a few to apply to the American political landscape.

[Spidey]

2 things…

1. You said “Pray China doesn't get a burr up it's collective ass and start shooting at someone.” So I have to assume that we won’t start that one…

2. Define “recent big war” Because I know my history…and it wasn’t the US that started Korea, Vietnam or the Gulf War…(we just finished that one) We didn’t start Afghanistan either…so I will give you Iraq.

Don’t bother to return to the subject…Drakona has already spoken…