Articles by " Andrew Davies"

South Australian defence industry summit

Start lineI was pleased to be invited to speak at the South Australian Government’s Defence Industry Policy Summit (PDF) earlier this week. I was invited in my role as a member of the Defence White Paper Expert Panel, and was asked to help set the scene for the discussion that followed. Here’s what I told the meeting.

Thanks for the opportunity to be here today. My topic is the Defence White Paper process, but I’m not able to say much about that as it’s still very much a work in progress. So let me give you the response I give everyone who asks me how it’s going. ‘It’s everything I expected it to be’.

In terms of this gathering, I’m not sure that the DWP is the most germane document. There are several important pieces of policy work going on in parallel, some of which will have at least as large an impact, particularly the development of a Defence Industry Policy Statement (DIPS), a shipbuilding plan and the First Principles Review of Defence’s organisation. Development of the DIPS is something that I and my Expert Panel colleague Mike Kalms were asked to take on by the Defence Minister here in Adelaide back in June. Read more

As a result of that, we’ve been touring the country to consult with industry groups and making site visits. We’ve heard some clear and consistent messages along the way from industry, and I’ve found some of the visits to be real eye-openers. I’ve been impressed with the industry capability and capacity I’ve seen in various places. That will all help inform the DIPS.

As for the DWP, I think it’s important that the discussion that ensues today takes into account the environment in which work is proceeding. Firstly, the federal government has made it clear that it’s going to continue to make major capability decisions. It has committed something like $20 billion to acquisitions such as the F-35 Joint Strike Fighter, P-8 Poseidon maritime patrol aircraft and the Triton surveillance drones. By doing so it’s avoiding the bottleneck in projects that’s accompanied previous DWPs—sometimes as much as 18 months of deferred decisions, which has a flow-on effect to ADF capability and to industry work-flow.

Second, it’s possible that major decisions will be made about shipbuilding and submarines prior to the DWP release. It’s also possible that they won’t, and I can say with high confidence that no decisions have been made to date. I think it’s fair to say that there are inclinations, but there’s still time for the sort of submission that will come from this meeting to influence the process.

Third, there’s the budget situation. In its first budget the government made good on its promise to increase defence spending. If it sticks to its pledge to reach 2% of GDP by 2023/24, Treasury forecasts suggest that the budget that year will be around $45 billion in today’s terms. That compares to this year’s $29 billion, amounting to an additional $16 billion to invest. That’s a lot of additional capability that can be acquired, and a lot of industry support to be purchased. There should be plenty of opportunity for industry. Not so much in the first few years, as there’s a shortfall from the underfunding of the 2009 DWP that has to be made up, but in the years to come there’ll be lots of new investment.

Finally, but possibly most importantly, there’s a whole-of-government policy environment that has to be taken into account. That’s worth studying for clues about government thinking on industry and innovation. A good place to start is the new Industry Innovation and Competitiveness Agenda (PDF) released last week. It had several policy objectives of relevance here, most notably boosting competitiveness and fostering ‘excellence, not dependence’. It also identified five growth sectors, which represent areas of comparative advantage in the Australian economy. The one of most relevance to defence is ‘advanced manufacturing’,

The Agenda observes that Asian countries are increasingly reducing trade barriers, reducing inefficient public spending, reducing taxes and improving competitiveness. In that environment, Australian industry will have to be innovative, be working at world’s best practice standards, nimble and—in the defence space—provide a capability-edge for the ADF. When putting forward business cases for defence industry investment, they’ll need to be couched in terms of competitive advantage and capability edge, not just ‘net benefit’, however calculated.

Finally, let me swap hats and become an ASPI commentator for a minute. As those of you who read The Strategist—which should be required reading—would know, I was much impressed by the productivity gains I saw in local shipyards recently. The touch labour productivity on the AWD is showing a learning of about 20% between vessels one and two, with a projection of between 10 and 15% from vessels two to three. That’s close to world standard.

Similarly, the Collins availability is much improved, suggesting that the Collins story is more about lack of resources than poor industry performance. Actually, it’s probably a combination of both, but increasing resources has enabled better performance. ASC has some way to go to be world’s best practice standard, but the trend is good.

One final comment. When I read today’s press clippings, I saw the call for a competition for the design and construction of the future submarines. As I’ve said before, that’s the way to do it.

Andrew Davies is senior analyst for defence capability and director of research at ASPI. Image courtesy of Andi Sidwell.

Australia’s latest commitment

A Royal Australian Air Force  F/A-18F Super Hornet loaded with explosive ordnance  in the Middle East.Defence analysis is my core business, so I thought I’d share my views about Australia’s decision about going to war again in the Middle East. As I’ll argue below, I think we’re doing the right thing, but I’m far from sure. Beware of ‘expert predictions’—I’m already on the public record with one observation on this conflict that was at the least premature, and may yet prove to be just plain wrong.

The reasons for my uncertainty are many. When I came to set out my thinking, it became painfully obvious that I’d have to include a long list of caveats: I’m not an expert on the Middle East, I speak none of the relevant languages (something that’s important when trying to understand the motives and thoughts of other people), and I have little understanding for the underlying cultural and religious issues. And I fear that those characteristics are shared by many Western decision makers—which would go a long way towards explaining the litany of policy failures in the region over many years. Read more

At the risk of producing a piece that’s too even-handed to be interesting, let’s start by setting out the pros and cons of Western involvement in this conflict. Starting at the strategic level, preventing ISIS from getting its hands on the full suite of capabilities provided by a reasonably modern industrialised state is a sensible goal. And having a geographically-significant part of a region already riven with sectarian tensions under ISIS’ control is surely worth avoiding.

Set against that is the difficulty of predicting the consequences of either success or failure against ISIS. I’ve yet to hear anyone articulate a coherent view of what the future looks like for Iraq or Syria, and then there’s the uncertainty of the impact on Turkey, Iran and Saudi Arabia. And can the West avoid being caught up in other nations’ political manoeuvres—such as Turkey’s positioning with respect to the Kurds? I don’t know how to evaluate the strategic cost-benefit balance.

Operationally, things are more clear-cut, at least for now. Adopting a policy of ‘air strikes only’ allows the West to intervene confident of being able to do so without loss and to engage and disengage as it sees fit. Of course, that changes dramatically if Hugh White’s right and there’s an escalation to the deployment of large-scale land forces.

My judgment last week was that air strikes would be able to limit the ability of ISIS to continue to overrun large areas of Iraq and eventually take effective control of the state. I said there’d already been a strategic win in that respect—clearly an error. In any case, I never thought that air power can defeat an opponent who can cease to be a conventional combatant and become an insurgent at will. Even if it can’t wrest control itself, through a mixture of conventional and unconventional attacks ISIS can render Iraq ungovernable (Syria already is.)

Last, but certainly not least, there’s the moral side of the balance sheet. I see two moral arguments for involvement. The first is that we can. If we stand by and watch atrocities of the sort we’ve been seeing lately, then we’ve decided they’re less unacceptable than becoming involved. The argument that the West only intervenes selectively to protect innocents, while true, doesn’t sway me. Stopping some bad things from happening is surely preferable to stopping none. I don’t go as far as subscribing to a ‘responsibility to protect’—that could turn into a heavy burden to bear—but every time we don’t there’s a moral downside. The second moral argument is that we helped create this situation, though I think we can go too far with that argument. Those events are now in the past and present-day decisions should be made only on the basis of our best assessment of potential future outcomes.

The only moral reason against intervention is that we could, again, end up making things worse. Perhaps it’s a failure of my imagination, but I don’t really see how much worse it could be than to have ISIS in control of Syria and Iraq—and potentially other areas later.

Bad things are happening, and regardless of what we do, bad things are likely to continue to happen—though if we act we could make things less bad. Intervening has a cost; the risk of an act of violence in Australia is probably higher than before the commitment, and there’s a clear danger of mission creep, escalation and entrapment. But not intervening has a cost too, and we could face a dreadful future enemy if we do nothing.

I think we’re doing the right thing, but I can’t be sure. These are profoundly difficult issues, and take my pondering for what it’s worth. But I only have to write about it. Decision makers in Canberra, Washington and elsewhere have to make these calls knowing that history will judge them. I don’t envy them.

Andrew Davies is senior analyst for defence capability and director of research at ASPI. Image courtesy of Department of Defence.

Coles up close

HMAS Farncomb arrives at ASC North in Osborne, South Australia, for a full-cycle docking.As members of the Defence Minister’s White Paper Expert Panel, Mike Kalms and I have been asked to help prepare the next Defence Industry Policy Statement (DIPS). An important part of that process has been getting around the country talking to defence industry stakeholders to capture their views. The results of our work will have to wait until government has decided which policy settings it wants to pursue before they’ll see the light of day, but getting out of the office and onto worksites gives a fresh perspective not always available from reading reports. And it gives me a chance to wear a fluoro vest and hard hat.

I thought I’d share some observations I made on a recent trip to the submarine side of ASC. (My impressions from a previous visit to the shipbuilding side of the business are here.) Like all the industry players and associations we’ve approached, ASC were good about making time for a discussion and briefing. It’s natural that industry wants to get their views across, but in this case there was another reason for them to be keen to engage: they have some good news to report. I’ve been a regular visitor to ASC submarines for some years now, but this was my first trip since the Coles Review recommendations really got to the implementation stage. Read more

So, what did I find? The most obvious thing was that the submarine-maintenance facility now looks like the hive of industry you’d expect to see in support of a major national defence capability. That might sound superficial, but it’s a striking difference. Where there was once a submarine ‘on blocks’ in the shed with a few small groups of workers gathered at one or two points because there wasn’t much room to work efficiently, there’s now a significant piece of infrastructure inside the shed that allows access to the submarine at multiple points at various levels. And work can now be disaggregated and proceed in parallel, with heavy machinery lifted out of the hull for work ‘alongside’ while the now-empty spaces inside the hull allow much better access to the internals of the boat.

But as important—if not as obvious to the naked eye—is a major reorganisation of the management arrangements between the various players in the submarine enterprise: Navy as capability manager and end customer, DMO as acquisition authority, ASC as maintainer and the Department of Finance as owner. There was previously a confusing and inefficient overlap of responsibilities between the various entities. It resulted in silliness, like maintenance work being held up because the maintainers had to wait for parts—often from offsite—because the procurement and warehousing side of the operation weren’t on the same page.

Diffuse responsibility also meant that no one was able to be held accountable when things inevitably went wrong. The national submarine enterprise consisted of a group of enmeshed and non-moving parts. That’s why essentially all (nine and a half out of ten) of the key recommendations in the Coles Review related to management arrangements. The reputation of the submarines themselves was collateral damage in what was a failure to properly manage the submarine fleet. The public perception of the Collins remains resolutely one of ‘dud subs‘. In truth, after the initial (and not insubstantial) technical problems were ironed out, for a long time they were good subs under dud collective management.

Today the division of labour and responsibility is much clearer and decision making is being devolved appropriately. The impact of those changes has been dramatic. There’s been a nearly 40% reduction in man-hours required for a full-cycle docking (the two-year-long overhaul and refurbishment process at the end of major duty cycles). That’s a productivity gain that the yard is proud of, and the morale of the workforce seems markedly higher as a result.

The result of those efforts has been a marked improvement in the availability of Collins boats for operations and training—which is ultimately what it has to be about. We don’t have long-term metrics yet and so it’s a little early to hail a renaissance in Australia’s submarine capability, but it’s fair to say that Collins submarine-availability report cards have gone from ‘must do better’ to ‘progressing well’.

It’s a crying shame that a $10 billion national investment in submarines had to fall over before the collective heads were conked together and told to sort it out. No doubt there are further improvements to be made, but so far the results are looking pretty good.

Andrew Davies is senior analyst for defence capability and director of research at ASPI. Image courtesy of ASC Pty Ltd.

Big stakes: Army’s future armoured vehicle fleet

An Australian Light Armoured Vehicle positioned to provide over watch for an infatry platoon patrolling through the "green zone" below.We’re at the Land Forces 2014 conference this week. Even a casual glance around the floor of the associated trade show suggests immediately that there’s some serious positioning going on among vehicle manufacturers to supply the Army’s next generation of protected-mobility vehicles.

The prize, of course, is Project Land 400—described by the DMO as:

 …one of Defence’s most significant capability programs both in terms of acquisition cost and its impact on the Army’s war fighting capability. The program seeks to address the emerging mounted close combat capability gap that exists between the current in-service vehicles and increasingly capable weapons that could be used against Australian forces.

The scope of the project is essentially to replace the entire frontline fleet of Army vehicles. Australian Army’s experience of operations in Afghanistan, during which improvised explosive devices were a constant source of danger, has made it an imperative to provide troops with increased protection. That will involve replacing several different fleets, including the Vietnam-era (but upgraded at much expense) M113 tracked personnel carriers and the ASLAV light-armoured vehicles. Read more

Land 400 is literally a big deal—the Defence Capability Plan gives the cost as ‘more than $10 billion’ but many defence watchers and industry players assess that the total value is likely to be double that. It’s big enough to be included in the DCP-distorting ‘big four’ projects with the F-35 and future submarines and frigates. Because of the big price tag, and the oversubscribed DCP, it’s been expected that Land 400 will be ‘phased’, with fleets being replaced progressively as they age out rather than a wholesale upgrade.

The most urgent replacement is the ASLAV fleet, which reaches life of type around 2021. That might seem a comfortable enough time to find the right vehicle on the world market and buy it ‘military off the shelf’ (MOTS). In fact, it’ll be needed, because it’s most certainly not a MOTS project—a fact we were pleased to hear recognised in today’s presentations.

The days in which soldiers embarked and disembarked from a vehicle with no interaction between man and machine other than transport are long past. The Army’s vehicles have to be able to work seamlessly with all of the land (and increasingly joint) information, communications and battle management systems. In other words, there’s a lot of systems-integration work to do. Given the time-honoured habit of integration-heavy projects taking longer than expected, getting the requests for tender out this year, as is the plan, seems about right.

But at least the ingredients—the vehicles and the embarked systems—are all MOTS, so there’s a fighting chance of getting it right. When the US Army set out to plan its Future Combat System (FCS) vehicles, it simply got too ambitious and complex. Requirements changed and caused turbulence in an already ambitious program. Eventually the Department of Defense cancelled the FCS program in 2009. While there are specific lessons to be learned within the acquisitions community, the ADF leadership must keep those in mind during the critical phases to come.

For example, US Army leadership and FCS program directors became too hung up on unrealistic ‘requirements’ in the midst of an on-going war that increased the demand for protection and survivability. In their initial assessment, US Army leaders saw the need to get to a global crisis fast and so maintained a requirement that the vehicles be C-130 deployable, which meant they had to be light. The wars in Iraq and Afghanistan increased the need for protection and survivability, which posed an ever more difficult engineering problem—compounded by the need to integrate a complex and interoperable system of systems that didn’t exist even in prototype. Adding requirements without revisiting assumptions about the needs of the future added difficulties to an ultimately insurmountable challenge. Our industries can innovate in a competitive environment, but there are limits.

ADF leadership is likely to face similar temptations and they’ve made sure the program comes with the caveat that they can change requirements during the process. With the Army Future Land Warfare Report indicating a future operational environment of crowded megacities coupled with increased lethal capability from enemy forces, Army leaders are going to face a similar tension between protection and survivability on the one hand and deployability and mobility on the other. A highly-protected and survivable vehicle that can’t operate because it’s too heavy for bridges or roads of rundown megacities or the poor infrastructure of nearby regional states is as worthless as a light-skinned vehicle that offers our troops only cursory protection.

Land 400 still has a long way to go but it seems headed in the right direction. A phased project based on proven vehicles and established systems isn’t without risk, but it’s less ambitious and less fraught than it might’ve been. ADF leaders will have to be decisive in finding the unhappy medium and then take the inevitable criticism years down the road. Having a good vehicle, built at reasonable cost and reasonably on time will be far better than having no vehicle at all.

Andrew Davies is senior analyst for defence capability and director of research at ASPI. Lieutenant Colonel Jan K. Gleiman is an active duty US Army officer and a visiting fellow at ASPI from United States Pacific Command. The views expressed in this post are his own. Image courtesy of Department of Defence.

Getting the submarine we want

PEARL HARBOR (Oct. 15, 2013) The Japan Maritime Self-Defense Force (JMSDF) submarine JS Unryu (SS 502) arrives at Joint Base Pearl Harbor-Hickam for a routine port visit. While in port, the submarine crew will conduct various training evolutions and have the opportunity to enjoy the sights and culture of Hawaii.It’s not hard to make sense of reports that a Japanese submarine is now in the forefront of the government’s mind as it turns its attention to replacing the Collins class. After all, looking around the world, the only two conventional submarines that come close to meeting Australia’s demanding requirements are our own Collins and (possibly) the Japanese Soryu class. And of the two, only one of them is still coming off an established production line.

None of the European submarines on the market have the range, endurance and habitability qualities required. While there are good reasons to think that the European design houses could help us with design work on a larger more capable boat, they’d be starting well behind the Japanese in terms of proven delivery.

And newspaper reports have the Soryu option at $20 billion—seemingly a ‘bargain’ compared to ASPI’s estimate of $36 billion for a local program, although that figure was predicated on a 4,000-ton boat designed to meet high performance specs. Defence’s most recent public statement showed that the project now has more realistic performance goals: a submarine with the range, speed and endurance of Collins, but with better sensors and better stealth. The cost and risk profile would decrease commensurately.

Read more

It’s perhaps not surprising that the government is impatient. The whole future submarine project was badly side-tracked by the 2009 Defence White Paper, which aimed for the moon as far as submarine capability was concerned. Years were lost working on high risk options that are now off the table. If a quick decision is a high priority, and revised performance levels are acceptable, then the Soryu starts to look attractive.

But it’s not clear how urgent the decision actually is. When Mark Thomson and I looked at the timelines for a future submarine developmental program, a capability gap seemed inevitable unless the life of the Collins class could be extended well into the second half of next decade. Publicly at least, Defence has said there’s no reason we can’t do that, although we have no idea of the cost.

Collins’ lifetime willing, there’s still a lot to be said for competition. The German firm TKMS—well established as an exporter of submarines—also pitched a price of $20 billion for a 12 boat build (video) at our conference back in April. Likewise, the French and Swedish submarine builders (the latter now in the hands of Saab) would like a crack at the project and might well be able to match or better that figure. Testing the market is the only way to know the actual price for sure.

In any case, a decision to opt for a Japanese submarine should be contingent on three major factors. First, the submarine must have the characteristics we want. At least based on open-source data, that’s not obviously the case. For a start, it’s not clear that the Japanese Navy operates submarines at long range for extended periods—the sort of operations that have been the mainstay of the Royal Australian Navy submarine fleet for decades. So any deal done with Japan to source submarines will need some careful government-to-government, navy-to-navy and engineer-to-engineer work done behind the scenes to ensure that what we’d get would be fit for our purposes.

Second, despite sometimes being characterised that way, this wouldn’t be an ‘off-the-shelf purchase’. We mightn’t want all of the options that Japan prioritises for their mission set, and there’s a long-standing and strong preference for American combat and weapon systems. Those would need to be integrated into the Japanese design—and submarines certainly aren’t ‘plug and play’ systems. Again, due diligence is required. As Gumley’s Laws remind us, MOTS + MOTS ≠ MOTS, and ‘Australianised MOTS’ is an oxymoron.

Third, it’s not all about the acquisition. While it’s not widely appreciated, the main failings exhibited during the Collins-class experience weren’t design or construction issues with the boats—those weren’t better or worse than other major construction projects, and they’d been largely sorted out by the end of the build. Rather, the main problem was the through-life support arrangements. That’s evidenced by the Coles review, in which essentially all of the key findings related to the management arrangements put in place—and which were manifestly inadequate for the task. Regardless of where we source submarines, we’ll have to support them properly. That means that any arrangement with Japan would have to be robust enough to provide parts and engineering support both in Australia and in Japan for decades to come.

For all of those reasons, a decision mightn’t be as imminent as some of the recent reporting suggests. The government is clearly attracted to what looks to be a good fit to our requirements, and which would help deepen an increasingly important security relationship. But it also knows that it has some homework to do before signing up.

Andrew Davies is senior analyst for defence capability and director of research at ASPI. This article is an expanded version of an opinion piece that appeared in the Australian Financial Review on 10 September 2014. Image courtesy of Flickr user US Pacific Fleet.

National security and civil liberties

We need to be able to keep an eye on the spooks.

It’s always good to see debate about civil liberties and security legislation here on The Strategist. With a number of changes to security legislation in the wind, it was appropriate that my colleagues Toby Feakin and Anthony Bergin took the issue up recently.

Anthony’s right that there’ll always be an acceptable level of government intrusiveness on individual privacy in the name of safety and security. For example, I don’t think anyone believes that police curbing dangerous driving on public roads is an unreasonable infringement of civil liberty. But when the nexus between government activity and public safety isn’t so obvious—as with metadata retention—or when government activity is secret, it’s reasonable to demand a higher level of justification, and a robust mechanism for accountability. Toby’s right; it’s not a simple matter of balancing security against privacy.

Neither of my colleagues addressed the core question of oversight. History shows us that, left to their own devices, intelligence and security agencies come up with all sorts of unhelpful notions. Despite various mechanisms, including constitutional protections, being in place to protect American’s privacy, that country’s NSA has overstepped the mark more than once (most recently in their offhand and probably unconstitutional treatment of the Foreign Intelligence Surveillance Court). And even the most ardent supporters of secretive state security should be troubled by recent revelations that CIA employees penetrated the computer systems of Congressional Committees investigating allegations of torture. Read more

Australia isn’t immune to security agency overreach, though we’ve managed to avoid anything as egregious as the recent CIA breach. During the Tampa incident, intelligence material collected and promulgated by DSD would ‘probably have been subject to professional legal privilege’ and did not ‘meet the criteria laid down in the rules for reporting the communications of Australians’. ASIO came under criticism for its handling of aspects of the Mamdouh Habib case.

Lest we forget, before not one but two Hope Royal Commissions, agencies could conceive and execute breathtakingly stupid actions. And ASIO’s past efforts in keeping tabs on politically active Australians could sometimes slip over into political rather than security intelligence. (Although Jack Waterford’s haircut positively invited state intervention.)

The Australian examples above suggest that Hope’s multi-tiered oversight framework for Australian intelligence is working. And if all else fails, having a difficult and antagonistic press that’s not afraid of cutting across the bows of the security establishment can help maintain propriety—if the over-zealous CIA employees mentioned above had applied the New York Times test, they might’ve thought better of their actions.

The likelihood of being caught is an important factor in deterring wrong doing (PDF). Applied to the activities of intelligence agencies, that suggests that any proposed changes to national security legislation should be closely scrutinised to ensure that effective and timely independent oversight can be applied. It seems that the National Security Legislation Amendment Bill (2014) doesn’t yet meet that requirement.

In a good example of our parliamentary and statutory oversight mechanisms working together, the Inspector General of Intelligence and Security (IGIS) made a submission (PDF) to the Parliamentary Joint Committee on Intelligence and Security’s inquiry into the bill. (You can see the other submissions, including those from ASIO and ASIS here.) Quite rightly, the IGIS doesn’t judge the value of the proposed amendments—that’s not her job—but makes some important points about oversight provisions.

It’s an important contribution to the discussion and anyone interested in the topic should read it. If nothing else, it shows how complex things can get when you try to codify them. For example, there are two pages and a supplementary submission (PDF) on the ‘definition of a computer’—and the observation that overseers of intelligence will also need extensive technical knowledge to adequately discharge their responsibilities.

The proposed bill would see an introduction of ‘ASIO affiliates’—staff of other agencies acting under ASIO’s auspices—as well as a reduction in both the need for agencies to seek prior authorisation for activities and to report afterwards. All of those changes would increase the need for oversight. They might also decrease the external visibility of operational activities and blur the boundaries between foreign and domestic spying, making effective oversight more challenging to implement.

The press—the ‘unofficial’ intelligence watchdog—are also concerned about the proposed changes. Their submission to the PJCIS (PDF) is worth reading as well, but notes that the signatories (a wide range of media bodies) are ‘concerned that the Bill includes provisions that erode freedom of communication and freedom of the press’ and could see journalists imprisoned for doing their jobs.

In short, I’m not so much concerned about the powers given to intelligence and security agencies, or the secrecy in which they have to work to be effective, as I am about the ability of the wider polity to ensure that those powers are used properly and responsibly. In the absence of effective oversight, we can’t be sure of that.

Andrew Davies is senior analyst for defence capability and director of research at ASPI. Image courtesy of Flickr user PaULis.

Graph of the week: learning about learning

An important concept in understanding production efficiency of complex items is the notion of a learning curve. The term has passed into common usage, with ‘steep learning curve’ being synonymous with ‘really hard to master’ (in all sorts of contexts). That badly misrepresents its actual technical meaning; a steep learning curve actually means that something is easy to master. In the case of the production of things like warships, it means that the workforce concerned is getting on top of technical difficulties quickly.

A learning curve can be drawn in a number of complementary ways, but for the purposes of analysing defence projects, the most useful form plots the number of person-hours required to produce each item as production rolls along. Here’s an illustrative example. If the first item takes 1,000 labour-hours to produce, and the learning (in its technical meaning) is 90%, then every time the production numbers double the labour required for each item reduces by 10%. I.e. the second one takes 900 hours, the fourth takes 810 hours, the eighth 729 hours etc. By the time the 25th is produced, it’s taking a little over 600 hours. (See the blue curve in Figure 1, click graph to enlarge) As production times come down, unit labour costs fall as well, though the more productive workforce often shares some of the gains in the form of higher wages.

Graph showing example learning curves (cumulative production vs hours of labour).

Read more

Now look at the other two curves, labelled 85% and 80%. These are the result of steeper learning curves, and they show the labour hours (and thus costs) coming down more quickly as production continues.

Those numbers weren’t chosen at random. It turns out that numbers around the 80–90% figure are typical for industrial applications. In the defence sector, shipbuilding has a steeper learning curve than aerospace production. The simple reason is that building ships is much more hands-on than building modern aircraft—people do a lot more of the actual producing of the finished item than machines. People benefit more from practice than machines do.

In fact, we can break down the production of an aircraft or ship into several sub-stages, each of which has its own learning factor. For example, in aircraft production (PDF, slide 11), the largely automated fabrication stage has a learning around 90%, while the more human-intensive assembly phase is closer to 75%. The overall learning curve will depend on the relative time taken in each of those stages.

To show this isn’t just theory, let me finish with a real world example, but with a warning that real life doesn’t produce nice smooth curves like the examples above. Any changes in design or materials used, for example, will tend to kick the curve upward again, until the new design and techniques are mastered.

Here’s a graph (click to enlarge) from a study of the production of ‘liberty ships’ in American shipyards in WWII (PDF). The various yards showed different levels of learning, but all of them managed to reduce the number of labour-hours per ship from 1.2 million or more to less than half the starting point. In this case, making efficient use of a limited labour force and producing ships quickly to replace those being lost at ferocious rates in the Battle of the Atlantic were both strategic necessities. The contracts put in place with shipbuilders provided incentives to do both.

Study of the production of 'liberty ships' in American shipyards in WWII.

As a final comment, learning curves show why it’s often desirable to avoid producing items like ships in small numbers: you get all of the disadvantages of the early unpractised workforce, but not the long-term advantages of the mature learning curve. Building ten Anzac frigates was almost guaranteed to give a better overall result in terms of production efficiency than a run of three AWDs. Eight future frigates, on the other hand, would again allow for the benefits of learning to help amortise the early costs (not to mention one-off costs associated with design, engineering and infrastructure) across a larger fleet.

Andrew Davies is senior analyst for defence capability and director of research at ASPI.

Note for the mathematically inclined:

For 90% learning, the ‘learning parameter’ λ = 0.9. It’s related to a power law for the number of hours $T(n)$ for the nth production item relative to the first: if $T(n)=T(1)n^{\mu}$, then $\lambda=2^{\mu}$, or  $\mu = \frac{ln\lambda}{ln2}$. For $\lambda$ = 0.9, $\mu$ = -0.152.

The British are coming

The British are coming!

Last week, The Australian broke the story of BAE Systems potentially being brought in to fix the troubled Air Warfare Destroyer project. The three-ship build is already well underway in Adelaide, and the project is currently managed through an industrial alliance contract involving government-owned ASC Pty Ltd, Raytheon Australia and Defence.

Approved at a cost of $8.5 billion dollars in 2007, the project has accumulated nearly two years of delays and $300 million in additional costs. A government-initiated review of the project by ex-US Secretary of the Navy Don Winter and former Transfield boss John White recommended a range of measures, including ‘the urgent insertion of an experienced shipbuilding management team into ASC’.

But while the summary of the Winter/White report was announced by government just two months ago, the AWD Alliance had taken steps before that to remedy some of the project’s shortcomings. One of us (Andrew) visited the site at Osborne last week, and it was clear that some good work has been done. Read more

When the Australian National Audit Office reported on the project in March, many of the identified problems related to the transfer of the design from Spain to Australia, and the inexperience of the Australian workforce after more than a decade without a build project. Those factors contributed to a poor start to the project. Low productivity was the inevitable result, due to reworking of both the design and often the hardware. As a result, the first-of-class HMAS Hobart took shape fitfully and inefficiently.

A critical question is whether the existing project management and workforce can retrieve the situation. The answer seems to be a ‘qualified yes’. The Alliance has recruited some experience in production engineering and is seeing positive results. Visits to the yard at about the same stage of progress on ships 1 and 2 revealed a huge difference between the two. The blocks for ship 2 in the yard are more complete (with internal piping, painting, insulation etc) and are constructed to tighter tolerances than the first. For example, when the upper blocks were first lowered into position on ship 1, a substantial gap resulted. On ship 2, the parts fitted neatly together. The already in situ features in the modules mean there’s less need for work in tight and cramped spaces (in particular, the need for difficult overhead work is greatly reduced). The reduction in labour hours required from the first to second of class will be well over 20%, with further improvements expected for ship 3.

The ‘yes’ has to be qualified because there’s still substantial work to be done before a functional warship is delivered. In particular, the capability of the vessels will depend critically on how the Aegis combat system and other sensors and weapons function together. The job of building integrated systems into the hulls is yet to come, although land-based integration is well advanced.

Perhaps because of those improvements, news of BAE’s putative role has taken many by surprise, even though the government showed its cards by appointing a team of corporate lawyers and investment bankers as strategic advisors back in June. By seeking out mergers and acquisition specialists (rather than shipbuilders), the government revealed its inclinations.

So what might we expect if BAE is brought in? At a minimum, BAE could provide individuals with shipbuilding expertise to assist ASC. More likely, BAE would be asked effectively to take charge of the project. It’s even possible that BAE would take an equity stake in ASC’s shipbuilding arm—perhaps contingent on completing the AWD project. (There are good reasons to retain ASC’s submarine maintenance role in government hands, at least for the time being.)

BAE taking charge of the AWD project—if that’s indeed what’s to happen—would bring benefits and risks. On the positive side, BAE could reach back to the UK for help. Of course, BAE’s own problems with module construction for the AWD, and their trials and tribulations with the LHD project, show that they’re capable of overestimating their own abilities. If nothing else, however, BAE would bring the commercial focus that the government-owned ASC lacked. And they’d have every incentive to do so; not only will their reputation be on the line, but success with the AWD would likely secure them the massive eight-vessel Future Frigate Anzac replacement next decade.

On the other hand, the disruption that’s an inevitable result of a change of management would have to be carefully handled. While BAE has experience as a subcontractor, there’d still be a lot for them to learn about the project. The challenge would be to ensure continuity of effort concurrent with the introduction of new blood. As we’ve seen, ASC hasn’t been sitting on its hands. A lot would depend on the attitudes taken by the parties involved and cooperation will be critical. If it happens, let’s hope that the intervention adds more than it subtracts from the project’s likelihood of success.

As ever, the devil’s in the details, and another post will discuss the intricacies of third-party intervention and what the alternatives might look like.

Andrew Davies is senior analyst for defence capability and director of research at ASPI. Mark Thomson is senior analyst for defence economics at ASPI. Image courtesy of Flickr user Kayla Casey.

A realistic future submarine—at last

The Royal Australian Navy Collins Class Submarine HMAS Sheean at sunset during a routine transit and training exercise off Christmas Island.

At the Defence and Industry conference this week, we got an official update on the status of the Future Submarine project (SEA1000) from the project head, RADM Greg Sammut and DMO’s General Manager of Submarines, David Gould. That’s welcome, as multi-billion dollar government projects should be exposed to public scrutiny to the extent possible consistent with commercial and security sensitivities. (I won’t comment on other rumours doing the rounds.)

In a recent ASPI paper, Mark and I summarised the thinking that was on display at ASPI’s conference in April. What we got this week showed encouraging progress in the three months since. I’ve been writing about the project for years now (like the F-35, it’s a gift that keeps giving) and have lamented the apparent lack of coherence in planning. So credit where it’s due; with a couple of exceptions which I’ll come back to later, I think we’ve arrived at a sensible approach. Read more

Firstly, RADM Sammut explained that the Integrated Project Team (IPT) in Adelaide is largely composed of industry representatives working in support of DMO’s project office. This ‘above the line’ industry participation is vitally important if the Commonwealth is to be a smart buyer. Having relevant industry experience in-house will allow Navy/DMO to refine their requirements cognisant of their impact on project costs and risks.

A dramatic illustration of that—and a pretty newsworthy one in my books—is that there’s been a significant stepping back from the 2009 Defence White Paper’s wildly ambitious aims. There’s no conventional submarine in the world with the range and endurance of the Collins class, but the 2009 aim was a ‘significantly greater’ performance. That led me (and others) to describe the projected submarine as a ‘conventionally-powered nuclear submarine’ and to question the feasibility of the project.

This week we learned that the revised capability aims aren’t very different from Collins in terms of range, speed and endurance. Capability enhancements will instead focus on sensor capabilities and stealthiness, both of which will make the subs more effective and survivable in the decades to come.

Another sensible step is to take the existing Collins combat system (a highly modified derivative of the USN’s Virginia class system) and weapons into the new class, at least in the first instance. That will allow for a spiral development path, in which the new hulls, sensors and propulsion systems can be worked out without the concomitant risks of developing a new combat system. We tried that with Collins and it caused more grief than it was worth, so full marks there.

Of course, putting new sensors into the future boats will require them to be integrated into the combat system. Planning for that eventuality, we were told that Australian software developers were being contracted for ‘out of cycle’ software-development work (ie not in the USN development cycle). Getting Australian industry into the high value-add end of systems integration, where competing in global markets is entirely possible, is also a welcome development.

David Gould described the next step of finding an industry design partner that’ll take the design brief provided by the IPT and produce a detailed design. It wasn’t 100% clear if the design partner would automatically become the build partner, but that would make sense, allowing for the transition from design to production engineering to flow with lower risk of things being ‘lost in translation’. Mark and I had a fair bit to say about how this might work in the recent paper, so I won’t labour the point here, other than to note that such an approach might have avoided some of the problems the AWD project had.

Lest I seem uncharacteristically charitable, let me point out a couple of things that didn’t sound quite right. As Manager of Submarines, David Gould has to worry about the existing fleet as well as the future one, and he noted that both types would be in service together for a considerable time. He explained that his preference was to have a single support contract to cover both. Presumably the thinking is that having in-house understanding of both designs would allow for a single support arrangement. I’m not totally convinced; unless the same design house is behind both (in practice meaning a Swedish choice), it seems to add complication in managing intellectual property—a significant problem in the past.

Finally, there were hints in the presentation about the possibility of offshore builds, but no discussion of how (or where) that might happen. It’s fair enough to be looking at foreign builds, as the costs and benefits of all options should be diligently explored. But then we were told that a local build (including the first of class) was important for knowledge transfer needed for future support. The mixed signals had some of the industry reps scratching their heads.

All in all, there were more steps forwards than backwards. It might be several years later than would’ve been optimal, but real progress is being made.

Andrew Davies is senior analyst for defence capability and director of research at ASPI. Image courtesy of Department of Defence.

Correction: an earlier version of this post described the Collins combat system as a derivative of the Virginia class system. In fact, the control system fits that description, but the interfaces with sensors and other systems are quite different. The author thanks a keen-eyed reader for pointing this out.

Civilian aviation remains a target

New surface-to-air missile 9M317 of 9K317 Buk-M2E at 2007 MAKS Airshow

The downing of MH17 is another reminder of the vulnerability of civilian aircraft to military weapons. When fired upon by a sophisticated missile system, airliners don’t stand much chance. Weather and collision-avoidance radars won’t give much, if any, warning of an incoming missile (and aren’t designed to) and there aren’t any onboard systems that would allow the aircraft to respond in any case. If the aircraft is in the missile’s engagement envelope—the ‘box’ of airspace the missile’s fuel and manoeuvrability allows it to reach—the outcome isn’t likely to be a happy one.

In short, the only way to keep airliners safe from missiles is to keep them away. For larger surface-to-air missiles (SAMs) of the type likely to have been involved in the recent atrocity, that means keeping a wide berth. A Russian SA-11 (likely last week’s culprit) can reach almost 46,000 ft, which is well above the cruising height of airliners around 30,000 ft. Read more

As the week’s events demonstrated, airliners and tense environments populated by military systems aren’t a good mix. During Cold War tensions, the Soviet air force shot down a Korean airliner in 1983 (and damaged another in 1978) and in 1988 a United States Navy warship shot down an Iranian civilian Airbus on a routine flightpath following a skirmish between surface vessels. If nothing else, the MH17 event might lead to a tightening of the protocols for civilian air traffic over conflict zones—though working against that will be the economics of fuel consumption and ticket prices.

Keeping the aircraft away from the threat by avoiding war zones (or even military exercise areas) is one thing, but a look through the list of historical airliner shoot-down events reveals there’s a risk that the threat comes to the aircraft instead. A number of civilian aircraft have been shot down, and others damaged, by man-portable air defence systems (MANPADS) fired near airfields by irregular groups of militants. Those shoulder-launched missile systems are designed for battlefield use against helicopters and low-flying aircraft and are smaller and more easily concealed than the large SAM systems involved in the incidents described above.

Because their size limits their range and altitude to about 5 km and 10,000 ft respectively, MANPADS don’t pose a threat to commercial aircraft at their cruising altitude. But they represent a real threat to aircraft operating at lower levels, especially at take-off or landing and, in principle, pretty much any airport in the world is vulnerable to attack from these systems. The footprint from which one can be fired against an airliner operating into or out of an airport covers about 800 square kilometers—an impossibly large area to secure. While civilian aircraft have survived hits from MANPADS (such as in this near disaster (video)), a hit on vital systems close to the ground gives the crew little time to respond.

For a terrorist group, those weapons represent an opportunity to prosecute an attack against one of their most preferred targets. The list of attacks shows that they have been used most often by insurgent groups in the Middle East and Africa, and on at least one occasion as part of a coordinated terrorist attack against Israeli civilians in Africa.

The threat to civil aviation from those systems has long been recognised. International efforts to limit their proliferation gained momentum last decade, with the development of the Wassenaar Arrangement for export controls on MANPADS in 2003 and increased regulation and reporting of MANPADS deals. These controls have helped restrict the spread of these weapons, though not before some found their way into the hands of groups such as al-Qaeda (PDF). (It’s not clear that the weapons in the hands of such groups are functional). The Wassenaar Arrangement was designed to keep the weapons safely in state inventories.

But, as Peter Jennings points out, we’re entering a period of history where some states are breaking down and groups of non-state actors such as militant Islamists in Syria and Iraq are perilously close to getting their hands onto the military and industrial inventories of nation states. With MANPADS being in the armouries of over 100 countries around the world, including many of the shakier ones (such as Libya), the possibility of them getting into the hands of extremist groups suddenly looks much more likely.

It’s entirely understandable that Western countries don’t want to get involved in the recent events in places like Syria and Iraq after the experiences of the past decade. But that mightn’t be the right call—the combination of returning fighters and looser control of weapons technologies with the potential to cause significant harm to Western interests and populations requires much greater vigilance. As far as MANPADS go, Australia has the advantage of having no land borders, which takes away the easiest way of smuggling such weapons, but it’s no time for complacency.

Andrew Davies is senior analyst for defence capability and director of research at ASPI. Image courtesy of Wikipedia.

14,000 sounds like a lot—down to the docks again

All four blocks have been lifted onto the LHD01 hull at the BAE Systems Williamstown Dockyard.

Earlier this week, The Australian ran a story about delays in the construction of the Navy’s new amphibious ships. At first blush, it looked like a familiar story of poor shipyard performance, with 14,000 defects found in the HMAS Canberra, the first of the two new LHDs, resulting in a delivery delay of seven months. As the newspaper pointed out, the problems come at a bad time for the Australian shipbuilding industry, after a critical report on the air warfare destroyer project and a government decision to outsource the construction of two new replenishment ships to overseas companies:

They come at a time when the Abbott government appears to be paralysed with indecision about how to proceed with the country’s largest def­ence project, the $36bn construction of up to 12 submarines in ­Adelaide.

The delay in the completion of HMAS Canberra at Melbourne’s Williamstown shipyards has disappointed Defence, which says low productivity, poor skills and a shortage of trained supervisors has combined to delay the delivery of the ship until later this year.

Read more

To be sure, the government is running hard on shipyard productivity. As the Defence Minister recently told the ABC in the context of the AWD program:

If we can’t get the program back on the rails then it will be very, very difficult for me or anyone else to advocate a long-term naval shipbuilding enterprise in Australia.

That’s the context for the Oz piece—another poorly performing naval shipbuilding program imperilling the future of the local industry.  But I think a closer look at the (limited) data in the story tells another tale—or at least allows another interpretation. It’s routine for large industrial projects to have a long list of defects identified for remediation. Collectively those defects form what’s called a ‘punch list‘, a list of items to be completed before delivery can be accepted. Some of the defects can be substantial—the article mentions electrical failures, leaking seals, unaligned pods and corrosion in propellers—or nothing more than a rivet that needs replacing or minor repainting.

For a project delivering multiple items, one measure of contractor performance is how many of the punch list entries are avoided in subsequent builds. To assess that, at the bottom of the story we find:

The second LHD ship, HMAS Adelaide, which arrived in Melbourne in February, has far fewer problems and Defence says it is ahead of schedule to be delivered to the navy in 2016.

The fact that the first ship is late while the second ahead of schedule suggests that the shipyard workers are actually working on a pretty steep learning curve, which—despite common usage to the contrary—is a good thing because it means that mistakes are learnt from efficiently. If there were systemic problems we’d expect both ships to be running behind schedule.

That suggests the initial estimates of how long it should take to do the first of class weren’t accurate. In other words, the preliminary project work and due diligence on the behalf of the Commonwealth and/or the estimates from the contractors involved (in this case Navantia in Spain as suppliers of the hulls and the drawings for the fit out and BAE as the local shipyard) were overly optimistic.

That wouldn’t be a new story either. In the AWD project, the difficulties of bringing together a first-time exporter of a warship design and a start-up shipyard were significantly underestimated. In that case, the problem was probably exacerbated by shipyard productivity falling short as well—but perhaps not as far short as a cursory glance at the outcomes might suggest.

In the case of a two-ship build, as is the case here, it’s almost academic where the fault lies, because the outcome is the same—delayed delivery of capability. (In this case the problems don’t result in a direct cost to the Commonwealth because of a fixed price contract.) But it matters much more if the government is going to commission further builds of larger numbers of ships (such as eight future frigates or 12 submarines). There’s a big difference between an underestimation of the start-up difficulties and costs (which get amortised over the whole fleet and add proportionally less in unit cost the larger the production run) and a systemically poorly performing yard, which potentially adds costs and delays to all of the hulls. It’s important to be able to tell the difference—not least because it also tells you where effort has to be applied to get better results the next time.

Andrew Davies is senior analyst for defence capability and director of research at ASPI. Image courtesy of Defence Materiel Organisation.

Allies, airpower and history

An RAAF P-40 Kittyhawk in PNG in WW2. By the time the strip at Nadzab was available in 1943 the USAAF was making use of more advanced aircraft.

In discussions about the future of ANZUS last week, I introduced a discussion of Australia–US cooperation in air combat and strike. Because of recent force structuring decisions, I think Australia’s well set up to make substantial contributions to coalition air-power operations in the future, but it’s worth thinking through how we might best do that.

History provides some valuable lessons. Australia’s first air operations with the United States were during WWII’s Pacific campaign. Australia started the war with equipment that wasn’t up to speed, and relied heavily on imports from the US and UK. Both of those nations had their own priorities and it took the RAAF some time to catch up. Read more

As a result, the Australian contribution to the allied air campaign wasn’t always especially helpful. As aviation historian Michael Claringbould observes, turning up for coalition operations and bringing along outmoded equipment can be counterproductive:

The formation of 10 Operational Group in late 1943 hindered the [US] 5th Air Force… [T]he limited contribution the RAAF would make at Nadzab [PNG] was at the expense of valuable apron space, badly needed by advanced US types. The fact that No. 10 showed up with obsolete or superfluous types frustrated the Americans, [who] were forced to allow the RAAF to operate from Nadzab for reasons of political compromise, rather than contribution to the war effort.

Flightpath, February–April 2012

So it’s possible for an ally—especially the junior one—to be a nuisance rather than an asset if its forces aren’t what the local commander needs. Balanced against that is the political payoff in having allied support, and it can be a net positive if mere presence is sufficiently valuable. That wasn’t the case in WWII because it was a war of survival, but in operations of a more discretionary nature the calculus can be different. Many of the smaller contributions to coalition operations in Iraq and Afghanistan can be filed under ‘little or no operational benefit but politically valuable’.

Of course, it’s better to make a valuable contribution to the politics and the operation, as Australia managed to do in subsequent wars. By 1945 the RAAF was the world’s fourth largest air force (admittedly after some of the former high-rung position holders were displaced with extreme prejudice), with an inventory of capable aircraft and experienced personnel. The Navy was assembling a capable air arm, and the RAAF and RAN were significant front-line contributors over Korea. The RAAF later took its Canberra strike bombers to Vietnam and worked successfully with American tactical air units, flying over 11,000 missions.

But those successful exercises in alliance air power haven’t been repeated. Australian defence spending fell dramatically after announcement of the Guam doctrine and the end of the Vietnam War, and by the 1990s, Australia’s defence forces were suffering from a scarcity of resources and the inevitable ‘hollowing out’ of capabilities. The ADF played no direct combat role in the 1991 Gulf War and when an American request for Australian F-111s was made for Operation Desert Fox in 1998, the aircraft weren’t fit for purpose, lacking critical electronic warfare equipment. Similarly, the Australian contribution to the air campaign in the 2003 Iraq War came only after Iraq’s air defence system had been effectively eliminated by US forces.

Together, those examples show that Australia has been a valuable contributor to air operations when it had capable and well-maintained equipment which allowed the RAAF to operate effectively alongside American forces. At other times we’ve been an ineffective but tolerated flag flyer (and in Desert Fox not even that).

Today, after a decade of investment into its air-combat capability and with more to come, the ADF is well-placed to be a real contributor to allied air operations, should the government of the day choose. Super Hornets put it on a par with the USN air combat and strike capability, and early next decade the F-35 will move Australia further up the American capability curve.

But tactical strike fighters mightn’t be the most valuable contribution we could make. The US won’t lack those—it will have 600+ Super Hornets and more F-35s by early next decade. Turning up with ‘more of the same’ could be useful, as it was in the widespread wars in Korea and Vietnam, but in a more focused campaign it could complicate American command and control while adding little extra combat capability. To avoid that we could instead contribute capabilities that are almost always oversubscribed in modern air warfare: electronic-warfare support, airborne early warning and control and air-to-air refueling. Again the RAAF is well placed, with small but capable fleets of all those types.

It remains to be seen whether the government will opt for further investment in alliance-specific capabilities. But if it does, the less glamorous but always valuable air-combat support assets would be a good place to start.

Andrew Davies is senior analyst for defence capability and director of research at ASPI. Image courtesy of Wikipedia.