On the numbers

“We’re putting in our industrial infrastructure around the world and we’re building new factories, all based on a production model that’s hypothesised by nine countries and potentially some foreign military sales as well,” Burbage told Australian Aviation during a recent visit to Australia. “But every one of those countries is moving their airplanes around in the queue, and while all of them are still intending today to buy the airplanes they said they were going to buy, they’re not on the schedule they originally said they were going to buy them on. ” So the problem for Burbage is, the JSF program is ramping up to be a high rate fighter production program, bigger than any program before it, and the aircraft costing models have all been based upon this high rate. But if the partner nations or any of the US services defer their orders from the early years, the economies of scale become distorted and the costs of the early aircraft start to rise. This in turn potentially prompts more deferrals, and thus further cost increases. “The partners generally are pretty solid except in the 2014-16 timeframe there’s some wavering, and you could argue that the near years are always going to be the uncertain ones because the money in the out years isn’t real money yet,” he explained.

“Pressures on defence budgets mean partners are saying ‘we won’t buy 10, we’ll buy eight’ or ‘we won’t buy six, we’ll buy two’, and those kinds of things. ” So while such numbers may seem relatively insignificant compared to the overall program requirement of more than 3500 and potentially as many as 4500 aircraft with exports to non partner nations, even small delays in aircraft procurement by a partner nation have flow-on effects to that country’s planned infrastructure and sometimes, its industrial returns, and thus may weaken its government’s ‘business case’, and hence its resolve for the number of aircraft it had initially planned to buy.

Opposing political forces and other interest groups often exploit such instances, drawing tenuous links to program flight test delays or the US budget processes that bear little resemblance to what is actually happening within the program itself, or take the opportunity to stir the pot by alluding to program “death spirals” and using other alarmist terminology.

Although no commitments have been made, Burbage says fortunately there is some interest from outside the partner nations for early build F-35s, which, if converted, would take some of the pressure off the early aircraft cost concerns. Israel has been identified as being close to an agreement for 20 F-35As with the likelihood of a further 50, with the most recent reports indicating the Israeli defence minister was expected to announce a deal sometime before September.

In addition, and despite ongoing budget woes, there is still interest and scope within the USAF’s and Navy’s planned buys to bring forward some of their aircraft into production slots left vacant by partners shifting back, as all US services are suffering from dwindling fleets of ‘legacy’ combat aircraft that will need replacing sooner rather than later.

In the meantime, Lockheed Martin is close to finalising a deal with the Pentagon for the fourth batch of low rate initial production (LRIP 4) F-35s that will comprise 32 aircraft, covering all three variants of the JSF, to be built in the 2012-13 timeframe. Burbage says the company had hoped to have LRIP 4 signed in March, but the program’s February restructure by US Defence Secretary Robert Gates, and its Nunn-McCurdy breach and subsequent re-certification, have seen this schedule slip.

“We’re behind where we wanted to be on LRIP 4, but the dynamics of it have changed substantially,” he explained.

“We initially submitted it as a cost-plus proposal, but now we are going to sign it as a fixed price. We didn’t change anything in the proposal other than agree to a fixed price, and we’ve been negotiating with the government since to try to get some reasonable assessment of the risks. ” The decision to go with a fixed price plus incentive LRIP 4 contract was made by Lockheed Martin CEO Bob Stevens earlier this year as a demonstration of the company’s faith in its pricing models and in an effort to add some stability to the program and to counter the commentary surrounding it. The company had originally intended to begin offering fixed price contracts from LRIP 6.

Interestingly, while the contract numbers have not been revealed, Burbage says the company started its negotiations for the 32 aircraft LRIP 4 contract more than 20 per cent below the unit price estimated by the Pentagon’s Cost Assessment and Program Evaluation (CAPE) group estimate of about US$130m (A$153m) earlier this year. Even more interestingly, despite basing its budget projections on the CAPE estimate, the Pentagon started its negotiating from a position a further 20 per cent below that of Lockheed Martin’s starting point, which perhaps demonstrates a more realistic indication of the true cost position of the aircraft.

Indeed, the CAPE estimates have been a source of great conjecture and Lockheed officials have been frustrated that these US specific cost estimates – which includes the cost of establishing about 15 training centres, basing infrastructure and airfield improvements (commonly called MILCON costs) at dozens of bases, initial sustainment and spares, and which also take into account all development costs for the aircraft – have been taken by many as indicative per unit costs for international partner JSFs as well.

“None of those costs apply to international partners, so when you read the numbers the CAPE uses, if you’re not a cost expert – which none of us are – it’s going to be really hard to figure out what the real number is,” Burbage said. “They throw it around like it’s the cost of the airplane, but it’s not the cost of the airplane. Those are not costs that are going to get passed onto partners, those are financial estimates of the total cost of the program, some of which is already sunk and some of which is forward. There’s the sticker price of the airplane which, in our case will be about US$60m (A$70. 6m) in today’s dollars, and for almost all of the international partners including Australia, on average they’re going to pay even less than that because their buy is skewed towards the far side of the [build program]. ”The CAPE modelling to determine its cost predictions have also been based upon the development and manufacturing programs of earlier fighters such as the F-22 and F/A-18E/F, both of which were designed in the late 1980s and early 1990s and are built with far less automation, and neither of which have or will realise even a quarter of the projected number of planned JSFs. Lockheed Martin says all F-35s built to date have consistently come in well below the ultra-conservative CAPE cost estimates which the Pentagon continues to use for its budgeting projections, and even below the more realistic and legally mandated Select Acquisition Report (SAR) estimates. Further, the company says this gap will continue to widen as production numbers rise in later LRIP builds and into multi-year production, and further efficiencies are realised.

Other factors such as the USAF’s decision to base just 57 instead of the originally planned 107 F-35s at the Eglin AFB training centre and the subsequent need to open a second training centre, possibly at Luke AFB in Arizona, have also increased costs, as has a decision to acquire additional mission simulators and to continue funding the GE/Rolls-Royce F136 engine, the costs for which are all included in the CAPE numbers.

“I think it’s been very poorly managed by the [US] Department of Defense in terms of the lack of clarification of what the CAPE numbers mean, because it’s been picked up in all the media as that’s the new price of the airplane – it’s not the new price of the airplane,” said a frustrated Burbage.

In the Australian context, much has been said and written of the $3. 22 billion budgeted for the RAAF’s first 14 initial F-35As expected to enter service from 2014. But these early, low rate build aircraft are expected to account for only about one-third of this number themselves, while the balance will include extensive new basing infrastructure at Williamtown and Tindal, including new squadron headquarters and maintenance facilities, ground support equipment, ordnance loading areas, and expected modifications to runways and ramps. It also includes pilot and maintenance training facilities, material and courseware, an initial basic spares holding, NACC project office costs, and the establishment of training and support infrastructure at Eglin AFB where the initial cadre of RAAF pilots will be trained.

Most of these costs are non-recurring and won’t apply to the RAAF’s follow on buys of F-35s.

“You’ll get all of these other additional costs that are included in that $3. 2 billion that don’t really apply to the airplane,” Burbage said. “And that’s what is happening in the US as well, they’re using that figure as the headline figure. ”

If able to be met, the projected US$60 million unit cost is very competitive with that currently seen for fourth generation US fighters such as the F/A‑18E/F and late model F-16s.

Of course, the US$60 million figure being talked about by Burbage is an indicative average unit cost of all three JSF models across the entire duration of the program in 2010 dollars, and won’t apply to early low rate build or the first multi year production batch. Just at what point the US$60 million aircraft appears on the production curve and subsequent aircraft start costing less is uncertain, as this will largely depend on the rate of the production ramp up in the early stages, but recent media reports have suggested this is likely to occur between 1200 and 1500 aircraft, or somewhere around a third the way into the planned build run.

But if further delays to the production ramp up are forced on the program, this average unit cost will rise and the average cost aircraft threshold will be crossed further down the production curve.

Lockheed Martin says it believes it can continue to stay well under the CAPE estimate and thus the amount budgeted for each aircraft, and hopes there will be an opportunity to ‘backfill’ the difference between the actual and the budgeted amounts with additional aircraft due to ‘buy to budget’ provisions within February’s program restructure language. Indeed, despite the restructure taking as many as 122 aircraft out of early LRIP batches, funding for the infrastructure and manning required to support those 122 aircraft has so far been retained in forward budgeting estimates.

Other conditions imposed upon the program in the February restructure and the subsequent Nunn-McCurdy recertification include the shifting of US$2. 8bn (A$3. 3bn) from procurement funding to development over the next five years, additional software development engineers and resources added, an additional F-35C carrier variant development aircraft added to the flight test program, the provision to use LRIP 3 production aircraft to support the development program if required, and the extension of the development phase by 13 months to mid 2014.

“The test community is used to planning a test program around the way it’s always been done,” explained Burbage, himself a former test pilot, “So you have your basic number of test flights and a basic number of test points, and then you assume you’re going to lose so many flights with bad weather, and so many flights for ground aborts, and you’re going to have to go back and do regression testing when you find problems, so [the assumption is] the planned 3000 flights becomes 5000 flights. ” But Burbage added that because of the high degree of software development and integration work already completed before flight test in laboratories and in flying test beds such as Lockheed Martin’s 737 based CATBird, and the reliability of the small but growing F-35 test fleet to date, the flight test program is one more of validation than of discovery, and is hitting key test points earlier in flight test than legacy programs. He added that the “test community” has continued to mandate that the flight test program be conducted with “more of a legacy type approach”, and as a result 6000 additional test points had been added to the program.

“To complete that additional scope, it pushes the back end of the test program out,” he said. “But the airplanes are turning out to be more reliable than legacy – right now we’re seeing the CTOL airplanes are much more reliable, but the STOVL airplanes aren’t quite there at this point.

There’s nothing wrong with the STOVL lift system or anything [technical] like that, but just maintenance and things. If we get that corrected and they get back on track, we could well start pulling that end back in if we don’t need the regression time and the re-fly time. We’ve got a long way ahead of us, but if we can be reliable and effective and we can start pulling that in, we could well finish at the time we had originally predicted. The conservative estimate now has us pushed out about 13 months, but that 13 months is also misunderstood in some circles as ‘everything’s been delayed 13 months’. No, the completion of tests has been extended by 13 months, but that doesn’t affect the production of the airplanes, and it doesn’t affect delivering airplanes to Australia. ” If met, the projected US$60 million unit cost is very competitive with that currently seen for fourth generation US fighters such as the F/A-18E/F and late model F-16s. “And with the F-35, you get all your essentials in the package which you often have to buy separately with legacy airplanes,” explained Burbage. “Historically on airplanes that carry things externally, it’s called ancillary mission equipment or AME. What we’re looking at is your infrared is carried in a pod, you have externalfuel tanks, and EW suites often have to be paid for separately too – it’s a separate colour of money. But on F-35 everything is wired into the airplane, it comes with the airplane and the cost includes all those extras. The $60 million is a full up airplane in Block 3 configuration, all the sensors, engine… everything. ” And on top of all the budgetary hoops the program has been forced to jump through in recent time, the JSF has also been drawn into the 2011 US defence budget negotiation process on Capitol Hill, especially as funding for the GE/Rolls F136 second engine continues to be at stake. As part of the HR5136 defence bill, the Senate agreed to the program’s plan to complete 394 test flights this year and to meet a planned number of key test points, but the House side has added additional test points and milestones that need to be met by the program by January 15 next year in order for 100 per cent of the planned contract funding in FY011 to be released, ostensibly as leverage to ensure continued funding for the F136.

“There are four defence committees, two authorisation and two appropriation – in theory the authorisers authorise what programs get funded, and the appropriators decide how much money to spend for this purpose,” Burbage explained. “But this isn’t always how it works. In the language that the committees put in, they always do their mathematical adjustments of the budget, but then they put in controlling language. What it implies is, before this money goes under contract, you have to complete (a number of test points). We don’t see any problem with any of them, but then on the House authorisation side they added a couple more ‘zingers’ in there because they really want the second engine funded, so they said that the program office is not authorised to raise more than 75 per cent of the funding for the [ JSF] program until all the funding for the second engine is released. ” Because both sides of the House must pass identical legislation for it to become law, the seemingly cumbersome funding issue is yet to be resolved. Representatives of the two authorisation committees are due to meet in ‘conference’ where the second engine and other issues will be thrashed out and the final language of a common bill will eventually be agreed upon. With mid term elections due later this year, it is anticipated this process will be completed by mid to late September so the member senators and representatives can get back to their home districts to campaign.

Doubts have also been cast upon the eventual numbers of F-35s the US services will require, both from the most recent US Quadrennial Defense Review (QDR) process which predicted the USAF in particular may only ever take about two-thirds of its stated requirement for 1763 F-35As, but also from extrapolating the logic of Lockheed Martin’s own arguments that the F-35 will be several times better than each of the numerous aircraft types it will replace, and thus questioning whether it will be necessary to replace F-16s, A-10s and AV-8Bs on a one-for-one basis. But Burbage is confident that, not only is the USAF number solid, but he even has hopes the total US requirement may actually rise.

“On the one-for-one issue, the 1763 was based on [the USAF getting] about 650 F-22s, but they’re only going to get about 150 combat coded F-22s, and only 187 altogether,” he explained. “So if you look at the way the Air Force force structure mans up fighter wings, there’s a shortfall in there that seems to underscore the need for 1763 JSFs. I think most people thought if the USAF got all the F-22s they needed, the number of F-35s may have someday come down to about 1400. But they’re 400 or so short on F-22s so it sort of balances that out. I think that, unless there’s a future QDR that physically changes the strategic number of Air Force fighter wings or the number of Navy carriers, the numbers are pretty tightly wound to those requirements right now. ” Burbage added that anecdotal evidence suggests the Navy and Marine Corps requirements may be understated as they were originally based on plans to rationalise Marine Corps units, which “doesn’t seem to be happening. ” Certainly, despite the F-35’s apparent superiority over legacy fighters, in order for the US to maintain a ‘global force’ capable of fighting two major conflicts concurrently while maintaining sufficient resources for the defence of US airspace, it will require a minimum number of fighter aircraft, squadrons and wings in order to achieve this, and it would require a dramatic reversal of US policy for overall fighter numbers to be substantially reduced.

Despite the current US focus on fighting the counter insurgency in Afghanistan and its apparent near term strategy to developcapabilities more suited to such asymmetric warfare, longer term strategists suggest this may be the last such conflict for some time, with the next war likely to be a higher intensity one against adversaries with more traditional air and anti-air capabilities.

“It will be a different scenario and it’s not just because of the air-to-air threat,” Burbage acknowledged. “It’s a different scenario because all of those [possible adversaries] have very sophisticated surfaceto- air threats too. This is the real reason the F-35 is there. ” For Australia, the US theory of critical mass also applies, albeit on a much smaller scale. Some observers have asked whether fewer fighters than the planned 100 would be sufficient for Australia’s defence requirements.

But if the RAAF wants to maintain two or three operational F-35 bases – Williamtown, Tindal and later, Amberley – allowing for maintenance and training requirements, certainly 50 and perhaps 75 F-35s would be spread too thinly across three locations.

Senior RAAF planners are adamant that 100 F-35s is the minimum number required to be able to maintain a workable critical mass of aircraft and pilots, and have even raised the prospect of possibly retaining 12 or even 24 of the by-then mature Super Hornets/Growlers long after the final F-35s enter service in the early to mid 2020s.

Testing on the F-35’s survivability has also progressed, with the aircraft’s radar cross section (RCS) predictions and measurement processes validated in testing recently, first in a specially built anechoic chamber in Fort Worth and then at Lockheed Martin’s secluded outdoor RCS test facility at Helendale in the desert north of Los Angeles. In the outdoor tests, a full scale production representative model of the aircraft was mounted upside down on a pole and moved in pitch, yaw, bank and other axes while radar waves in multiple spectrums and frequencies were fired at it from a large antenna located almost a kilometre away.

Criticisms of the F-35’s lack of full spectrum low observability (LO) have been constantly levelled at the program from some quarters, especially when compared to the stealth performance of the F-22 Raptor and the emergence of new Russian systems such as the S-300/400 SAMs and the new PAK-FA development aircraft.

“There are some people in the world that have this analytical description of stealth, and they claim that we really knew how to do it when we did the F-117, but sometime, somehow between the F-117 and the F-35, we forgot everything we knew,” said Burbage. “And all of the stealth experts of the world are suddenly with the Russians and the Chinese now, none of them are left here – it’s crazy stuff! We’ve got far more experience – anybody can take a picture of a stealthy airplane and try and recreate that shape, but to get all of that technical integration of high gain antennas and low signature returns, there’s only one company in the world that can do that and that’s the [Lockheed Martin] Skunk Works. Find anybody else that’s done it, nobody else has ever made a real success of it, so to think that all of that capability has (sprung up overnight) and to think that you can now produce it effectively at any kind of production rate is naive until you go and get some real assessment. It’s taken us more than a billion dollars to go from F-16 and F-22 production to the F-35 generation with our investment in our facility in Fort Worth in automation and precision and motion based production systems, so that we could get the tolerance control down to allow the airplane to be built as a stealthy airplane.

It’s not easy you know.

“And even though some airplanes claim that they’ve done some LO treatments, as long as they carry their weapons externally they just won’t have the survivability that we have. We also purposely damage the [full scale model] to test to see at what point we start picking up the degradation of stealth based on normal wear and tear or damage to the airplane. ” In a broader sense Burbage’s role is about survivability too – promoting the JSF message to partners and politicians to shore up and formalise their commitments to the program to ensure that Lockheed Martin has the production volumes it needs so that the low per unit cost affordability targets survive.