It’s difficult to see a convincing case on ‘national interest’ grounds for moving full-cycle docking of the Collins-class submarines from ASC’s Osborne shipyard to Fremantle in Western Australia. Indeed, doing so may put the navy’s operational capability at risk.

First consider the program and capability risk involved in attempting to relocate such a significant industrial capability. ASC manufactured all six Collins submarines in Adelaide and has warned of the dangers of a move. Maintaining and extending the in-service life of the Collins fleet is now vital to the transition to the Attack class in the 2030s.

The intellectual property, the skilled workforce, the industry base, the infrastructure and the university and international partners for Collins full-cycle docking and the critical life-of-type extension (LOTE) are well established in Adelaide. It’s unlikely that the skilled workforce in Adelaide would willingly endure the personal and family costs of moving to the west. Lucrative jobs are becoming available at Osborne in the Attack-class submarine and Hunter-class frigate programs. Relocating infrastructure would be expensive and potentially pose a significant risk to the availability of submarines.

Next consider strategic geography. The government is yet to decide where in Australia the 12 Attack-class submarines will be based. For sound reasons, it’s anticipated that half will be in Western Australia and the other half in New South Wales. Adelaide is midway between the two fleet bases. Having half of the navy’s submariners located in Sydney strengthens the case for centrally positioned Adelaide and ameliorates to an extent some of the very important personnel arguments advanced by Rowan Moffitt in a recent Strategist post.

Claims from Western Australia and other quarters that the necessary skilled workforce, infrastructure and industry support to simultaneously build the Attack-class and Hunter-class vessels and sustain the Collins submarines is beyond the capability of South Australia first surfaced in 2014–15. The proposed move to WA is yet to be supported by a sound business case.

Delivering the required skills and infrastructure will be challenging for South Australia, but ASC, industry and the state government have clear plans to achieve this. The higher education and skills training capabilities in SA are well proven. The engineering and technical challenge of a Collins full-cycle docking and LOTE have much in common with a new build and require the capabilities that reside in Adelaide with the submarine builder. The requirements for skills, infrastructure and industry support for Collins full-cycle docking and LOTE and for the Attack class are complementary.

WA has industrial challenges of its own. The RAN fleet is home-based near Fremantle at HMAS Stirling, which is already committed to find workers and capacity for mid-cycle Collins sustainment, for maintenance of other ships in the fleet and for a vibrant small ships construction industry. On top of these demands, the state accounts for 46% of Australia’s exports, including 98% of our iron ore, 75% of our gold, 50% of our gas, 61% or our oil and 76% of our condensate. GDP in WA is already $82,000 per capita compared with $58,000 in the rest of the country. Skilled workers and industry in WA enjoy fortunate choices. How is it in the national interest to have Collins full-cycle dockings compete for the workforce and support facilities against these other conflicting demands for labour and resources in WA?

The WA government is keen to support naval shipbuilding, but during the mining boom defence industry leaders had to struggle to gain its attention. Mining booms will come again, and skilled people will be drawn away from defence work to high-paying jobs elsewhere as has occurred in the past. By contrast, SA gives defence 100% of its attention. The skilled workforce at ASC in SA has a proven ability to guarantee Collins submarine availability to the navy. With change comes additional risk.

On top of these considerations, Australia is facing dynamic strategic change in the Indo-Pacific. Defence is reassessing its capability requirements. As Hugh White and many other contributors to The Strategist have reminded us, the remote prospect of Australia needing to defend itself independently from a major aggressor is part of that conversation. Is it wise to locate all the Collins-class and the Attack-class submarines at one port in WA along with deep sustainment, full-cycle docking and LOTE, thus putting all our eggs in one basket?

SA was chosen as a major defence industry hub during World War II because it was secure from attack and offered proven industrial capability. If Collins submarines are to be on the hard stand, dissembled, out of the water and out of action for years at a time over coming decades, they should be in the most secure location.

When we undertook the nation-building Snowy Mountains hydroelectricity scheme, we created an authority with the power to get the job done. In a previous Strategist post, I suggested that we consider creating a national shipbuilding authority to report directly to the prime minister with an enterprise- and industry-focused executive board, a CEO and the authority to coordinate the national effort for both new construction and sustainment. Improving the machinery to deliver the national shipbuilding plan would better inform decision-makers on questions about how best to harness our people and resources to concurrently build and sustain operational capability.

The final decision on Collins full-cycle docking and LOTE will need to be supported by a convincing strategic argument and a sound business case and must be seen to be genuinely in the national interest. The credibility of the ministers who make the decisions and the advice they have received from Defence is about to be tested.

Australia’s next submarines will be the most sophisticated naval vessels under construction in the world and they will be built here in Australia with Australian workers using Australian expertise and Australian steel. The strong desire to maximise Australian content has more to it than mere symbology. In fact, as David Nicholls argued recently, it goes to the level of sovereignty achieved and will be a critical determinant of the operational capability provided to Navy, and define Australia’s control over the design, along with local industry’s involvement in its support. It will also determine whether many of the long-term high-tech jobs are created in France or in Australia.

The announcements by the government and Defence officials are encouraging. But why is sovereignty so important and what level should we aim for?

Deterrence is Australia’s top-level requirements of its submarines, facilitated by capabilities to:

• contribute to sea space denial, guarding our approaches;
• deploy forward, exploiting stealth to gain access to key areas; and,
• gain valuable intelligence and warning or to strike as appropriate.

These requirements, combined with our geography, underpin the arguments for a larger submarine, not currently available on the shelf. It’s necessary then that a special design is undertaken. Since the design will be unique to Australia, we must sustain and maintain it, as well as evolve the knowledge to support it through the 50-year life of the program. Capability, responsiveness, security and probably cost are best served if we do all of that in Australia, just as we have done for Collins-class for the past 16 years.

To achieve this we need a higher level of sovereignty than we seek for off-the-shelf purchases like the Joint Strike Fighter. It should include complete freedom to develop the design, with delivery of all information required to achieve this, unencumbered by any intellectual property inhibitions involving the French design.

The conceptual design for Australia’s next submarine fleet is now underway in France. The phases that follow—the detailed design, production design and the planning for in-service support—should be performed in Australia by an Australian entity with DCNS support, just as was done for the Collins boats. This will ensure the maximum utilisation of an Australian supply chain for the components as the Collins has demonstrated—over 90% of ASC’s support funds for Collins are currently being spent in Australia.

Achieving a successful build under Australia sovereignty is best achieved by advancing ASC’s existing design support capability for Collins and the very successful Submarine Enterprise’s capability that has achieved the rejuvenation of the Collins capability.

Encouragingly, the recent decision to split ASC into 3 government-owned companies opens the possibility for a fourth new company: a partnership between the ASC and DCNS to construct the submarines at facilities leased for this purpose. That wouldn’t be a new experience for DCNS, which is 62% owned by the French government.

Sovereign doesn’t mean solitary; we should be engaged with and supported by our allies. Given the selection of DCNS as the design partner, that will require a new relationship with France and her Navy. It will complement our existing strategic, operational and tactical relationship with the USA and USN.

The proposed Australian design environment will also allow us to maximise protection of  third party IP and gain optimum benefit from Allied support. The mobilisation phase now underway must factor in the level of sovereignty we seek.

There are a number of platform issues arising from our requirements and, since these are the most difficult/expensive characteristics to alter in service, a big effort to resolve them is required during the design process.

Given the lead time and the input into the detailed design process suitable shore-based platform R&D facilities are now urgently required to resolve platform issues including those arising from the long transits, reducing the snort signature, achieving excellent mobility and dived endurance. For example:

• propulsion system developed for the tropical operating environment
• the choice of higher performance lithium batteries involves significant design and safety issues
• Air Independent Propulsion (AIP) will be a key requirement for the challenging operating environment
• automation to support long mission requirements and reduce crew demands

We currently rely on the USN for modification to our Command Management System, which impacts our ability to meet specific RAN requirements for conventional submarines and has severely throttled our very capable SMEs in this field. Sovereign control requires local capacity to support and optimise the software in Australia; a similar capability that we had with the Oberon Submarine Warfare Systems Centre that drove the successful Submarine Weapons Update Program. It provided a peer to work with USN counterparts and served to mobilise, focus and drive Australian SME capabilities.

Sovereignty is a fundamental enabler for maximising Australian interests. Demanding operational requirements and a rolling construction program of 12 submarines enable and justify a high level of sovereignty. Decisions made in the contracts now in place to commence mobilisation and the design process plus those to be signed for platform R&D and construction entity will determine the level of sovereignty achieved. The same logic should be extended to the Command Management component of the combat system.

Sovereignty is more than symbolism. The level of sovereignty achieved will have a major impact on the operational capability boosted by incorporation of French, Australian and US R&D and Australian skills/employment in the high-tech areas of design, engineering and R&D for future evolution and support of the design.

We should establish and nourish an Australian design environment. The current efforts to achieve this reflected in official announcements should be applauded and encouraged.

Last week we wrote about the transition from the Collins class to the future submarine. Among other things, we pointed out that there’s a tight relationship between the delivery tempo, the maintenance cycle of the boats and the industrial effort required to first produce and then sustain them. We described the effects of the problematic transition from the Oberon class submarines to the Collins in our 2012 Mind the Gap paper. In thinking through the future transition, we now understand the cause: failure was effectively planned into the Oberon to Collins transition, and mismanagement then made it worse.

Probably the clearest way to explain that observation is to work backwards from the current arrangements for Collins maintenance. Outlined in the Coles review into the Collins fleet, the so-called ‘10+2’ arrangement has finally produced a sustainable solution for submarine maintenance—albeit 20 years after delivery of HMAS Collins. It entails ten years between two year periods of major overhaul and refurbishment work (Full Cycle Dockings, or FCDs). In between FCDs there’s a one year mid-cycle docking (MCD) and two intermediate maintenance periods of six months each. So in each 10+2 cycle, a submarine is available for operations for eight years.

The 10+2 duty cycle is a thing of beauty (figure 1). It sequences Collins support to ensure that two boats are in maintenance at any given time—one undergoing an FCD (in Adelaide) and another undergoing either an MCD or 6-month intermediate docking (in Western Australia). In this model four boats are continuously available for operations, so Navy has a steady availability of sea days to plan training and operations around. And there are two maintenance teams continuously occupied, with no peaks or troughs in workload.

Figure 1: The Coles 10 +2 duty cycle for the Collins in six month time increments

The 10+2 model is being introduced at present. Each of the vessels is scheduled to be delivered from an FCD to begin their third operational period, commencing with Farncomb in mid-2016, followed by Collins in mid-2018 and the other four at two year intervals thereafter. It’s taking some juggling to get to 10+2 because the historical accidents that brought us to where we are means that the FCD dates of the fleet don’t quite line up. Some boats will have to be withdrawn from their second operating period early (HMAS Collins has been out of the water since 2012) or have their current operational period extended a little beyond 10 years (or be out of commission for a while after the 10 years elapses). But the disruption is worthwhile, because it’ll lead to an orderly and sustainable regime for the fleet, culminating in boats leaving service at two-year intervals; Farncomb in mid-2026, Collins in mid-2028, and so on and so forth, see Table 1.

But when the Collins-class was delivered, the plan was for an ‘8+3’ cycle, which couldn’t have delivered anything like the neat arrangement we’ll soon have, especially given the irregular delivery schedule of the Collins program. Using the delivery dates above, the Collins maintenance schedule would have looked something like figure 2 below (we’ve simplified the intermediate dockings a little for this schematic but the point remains).

Figure 2: Schematic of Collins maintenance requirements on ‘8+3’ cycle after initial delivery

As the figure shows, once the series of FCDs begins, eight years after the delivery of HMAS Collins, at various times there are variously one, two or three boats in full- or mid-cycle maintenance. The resulting peaks and troughs in maintenance requirements are difficult to manage. It’s hard and costly to ramp up and down the skilled labour force required to service submarines, and even harder to arrange for additional facilities to be available during surges. It seems likely that this mismatch was a contributing factor to the extended periods some of the boats spent out of the water around the middle of the 2000s. The capability implications were just as dire; even if everything went as planned, at times half of the nation’s submarine fleet would have been offline for a year or more—which in fact happened more than once. The painful experience of the 1990s capability gap and subsequent availability shortfalls under the shambolic maintenance regimen of the 2000s effectively wasted billions of taxpayer dollars.

The conclusion we reluctantly come to is that the Collins program was so poorly planned for through life support after delivery that a substandard outcome was assured even if everything went according to plan (which it didn’t). Getting it right next time will depend on carefully balancing the delivery interval between boats, their subsequent operational and maintenance periods and managing the remaining life in the Collins fleet. We understand that the process to identify the industrial partner for the future submarine requires the bidders to produce a coordinated delivery and sustainment plan. Let’s hope that it all works out—but there are a lot of moving parts to be choreographed.

The recent announcement that the government would seek a continuous ship-building program in South Australia has attracted endorsement (from Warren King) and criticism (from Nic Stuart and Hugh White among others). But in this post I’d like to say something about how governments make decisions when big pots of money are involved. Anyone who believes that those decisions are guided by a single imperative—to build a submarine, for example—needs to get with the program.

Go back and have a look at the decisions surrounding the Collins-class submarines. The government certainly wanted new submarines. But it didn’t just want that. It also wanted the bulk of the monies to be spent in Australia, solid job outcomes for Australian workers, technology transfer to Australian firms, and that favourable political outcome it believed should be the proper return on good policy-making. In short, it wanted to hit five birds with one stone. Some people even think it did manage to hit four of them—no mean feat, though perhaps an outcome not to the immediate satisfaction of those who wanted good new submarines.

The current wave of naval procurement faces similar pressures: anytime a government’s spending some tens of billions of dollars on something, it will want to achieve multiple goals from that spending. We elect governments precisely in order to take those decisions. Otherwise, we’d just let a group of strategic, defence and naval experts decide what to buy and how to buy it.

Does this mean the government can decide whatever it wants without criticism? Of course not. But it’s not an especially telling blow to criticise a government for wanting to achieve multiple objectives in relation to a major procurement program. Defence issues in general aren’t ‘above politics’. Yes, we want to get the decision ‘right’—in the sense that we want to procure effective military equipment efficiently. But we want similar effective, efficient outcomes in other portfolios—health and education, for example—and none of those is above politics.

One part of getting the decision right when five birds are involved is to make sure all of the birds are worth hitting. If we’re talking about five plump pheasants to grace Australia’s dinner table, that’s one thing. Five sparrows constitute a more meagre repast. So how do we maximise our return in relation to the money spent?

If we just wanted to inject government monies into the economy, we needn’t build ships at all. Governments could just hand out money. But we want ships and submarines to be part of the outcome—indeed, they’re the raison d’être of the program. So the costs we accept to get those ships and submarines (one of the birds) can be higher if they give us four more birds as well. But not grossly higher. Nor should we accept a submarine markedly inferior in its capabilities just to get the other four birds. The question of what’s an acceptable level of variation in cost and capability might well produce multiple answers across the electorate—though probably with less spread than some might imagine. The government ultimately has to defend its decisions before that electorate. Still, I can’t recall any Australian government being voted out of office because of poor defence procurement decisions.

So, is there a model of defence procurement that steers us through the issues to ensure an optimal outcome? Not that I know. Lauren Holland has previously written of three procurement models that turn respectively upon strategic analysis, technological momentum and political support. The five-birds option would seem to fit best in the political support model. And what’s the danger with that? Well, in Holland’s words, the principal danger is that ‘too many sets of interests must be met by a single weapons system’. For a country like Australia, which runs few major procurement programs, that’s a real hazard.

Around the world, Defence Procurement regularly finds itself the stablemate of Political Dealing. And the quality of political decision-making is mixed. The Brits slumped into an orphan Chevaline program for their nuclear forces; the US Congress regularly changes procurement to keep open industrial facilities in key districts; the Kiwis have mixed-and-matched forces to ensure they stay on good terms with a range of countries. But politicians aren’t just responsible for all the poor decisions; they’re responsible for all the good ones as well, including those taken against the advice of their public servants.

Purists might wish those decisions were taken differently. But wishes build no submarines. As Paul Keating once observed (though not in relation to his federal colleagues) it’s dangerous to stand between a politician and a bucket of money. I suspect the Australian government has now passed the point where it’s willing to focus on only one bird. We should do what we can to make sure the five birds—perhaps four birds, since voters might have an interest in responsive government but don’t have a direct interest in the government holding its marginal seats—are as plump as possible. That’s what will determine whether the decision’s a good or bad one.

It’s hard to say which is worse: the Navy’s attempt to hide its performance behind a cloak of secrecy or its abject failure to do so effectively. Back in late 2012, it was possible to reverse engineer a chart in the Coles Review to show the annual number of Unit Ready Days (MRD) achieved by the Collins fleet for three years beyond the last deliberate disclosure in the 2008-09 Defence Annual Report. Well they’ve done it again—this time in an answer to a Senate Committee question-on-notice, providing us monthly data all the way up to February this year.

The chart above shows the results. Note that the last data point is the rolling annual average as at February 2015. The good news is that the remediation of Collins sustainment appears to be delivering improvements commensurate with Navy’s plan to get up to the international benchmark of just under 1,200 days per year from its fleet of six boats. That said, the benchmark appears to have slipped down to about 1,120 in the latest data—we’re not sure what’s going on there.

The uptick in availability shows that we’re now starting to see a maturity in the approach to maintaining the Collins fleet that was sadly lacking over its first decade in operation. The complexity and cost of being a parent nation has finally sunk in, and the current effort doesn’t just look good, it’s delivering. That’s a substantial improvement on what we saw late last decade. To see how far we’ve come in terms of our understanding of the task at hand, it’s worth revisiting what the Navy was telling the Parliament (PDF) in 1999. Back then it expected (or at least expected to be believed) that:

‘…when the mature fleet of six Collins submarines was available, five could be expected to [sic] available for operations at any given time, and, as the Collins were significantly more capable than the Oberons, it would therefore provide a much greater overall submarine capability.’

That prediction is totally at odds with the ‘rule of three’ that forms the usual approach to managing submarine fleets. If, as expected, the Future Submarine fleet grows in numbers, it’ll still be governed by that rule of thumb. The government understands that as well, expecting that:

‘…two submarines are in full cycle docking at any one time, with, in general terms, one and sometimes two in shorter dockings and maintenance. This means Defence can currently plan on having two and sometimes three submarines available to the Fleet Commander for tasking at any one time.’

Let’s be clear: we wouldn’t be putting submarine availability data out in the public domain (or, more accurately, repackaging data that’s already out there) if we thought it was at all likely to jeopardise national security. Any would-be adversary also understands the rule of three and can take an educated guess about the number of Australian submarines that might be out and about at any given time. And even if it isn’t sure about the status of one or two of the boats, prudent planning would require a working assumption of more rather than less.

Finally, we’re reporting on past availability performance, which doesn’t reflect what’s happening today. We think this is pretty clearly a case of “official’s secrets” rather than “official secrets”.