Projects: AWD – facing into the weather | ADM June 2011

Gregor Ferguson | Sydney

The module constructions problems disclosed last year were the tip of a significant iceberg for the Air Warfare Destroyer program. The project is now 12 months behind schedule with the potential for further delays if other problems emerge.

On 26 May the Ministers for Defence and Defence Materiel, Stephen Smith and Jason Clare, unveiled a fundamental reworking of the construction program for the Navy’s Hobart-class Air Warfare Destroyers.

The technical difficulties and project delays exposed by their announcement are no credit to anybody involved.

The consequences are fairly clear: the entire project has shifted 12 months to the right; the first ship, HMAS Hobart, will now be delivered in 2015 – assuming nothing else goes wrong which could impact on the schedule. Without the project changes they announced, the schedule could have slipped by a further 12 months – a two-year delay in a construction program which began only 18 months ago.

The schedule for the construction phase of the AWD Project, Sea 4000, has been slipping gradually since 2009. The Preliminary Design Review (PDR) for the Hobart-class was held on schedule in Adelaide in December 2008 and went well. Design data was delivered on schedule from equipment suppliers and sub-contractors to Navantia to enable detailed platform design to proceed on schedule. And the company delivered the first batch of an eventual 10,000 production drawings to the AWD Alliance in Adelaide on time in early-2009, though detail design of the final module for the first ship wasn’t expected to be completed until this year.

In May 2009 the then-minister for defence, Joel Fitzgibbon, announced that Forges in Newcastle and AIMTEK in Cairns (trading under the long-established NQEA brand name) were the preferred tenders for the estimated $450 million contract to build and fit out 66 hull and superstructure modules (or ‘blocks’ as Navantia calls them) for the three ships. AIMTEK was to manufacture 36 blocks over five years and Forgacs would build 30.

ASC Shipbuilder would construct only the 27 core modules for each ship, including the critical ‘Aegis Tower’ containing the Aegis radar antennas and their wave guides. Contracts were scheduled to be signed later that year.

The announcement of the block sub-contractors was expected much earlier in 2009 and the delay showed tender evaluations had proved harder then expected. The choice of AIMTEK in preference to BAE Systems at Williamstown, given the latter’s experience of building and consolidating blocks for the Anzac frigates, was widely regarded as counter-intuitive (even bizarre). It was no surprise, though still something of a relief to many in Defence and industry, when AIMTEK proved unable to meet its obligations to provide financial guarantees and the work was shifted to Williamstown. In hindsight, AIMTEK may feel it got off lightly.

This extended process put module construction slightly behind the planned program, but reportedly didn’t affect the project schedule – the first steel didn’t need to be cut until September 2009, with the first modules due at Osborne in early 2010. It is supposed to take about 16 months to integrate the modules for each ship, including installation of the Aegis system. The final block for the third ship was scheduled to be delivered in late-2014.

All that has now changed. The first block to arrive at Osborne is due next month – some 18 months behind the original schedule. And it won’t be Block 107, on which so many weld and quality defects were identified last year, but Block 109, the adjacent keel section for the first ship. Meanwhile, BAE Systems’ work share has been reduced. When the problems with that first hull unit, Block 107, were revealed late last year the AWD Alliance re-allocated some of BAE Systems’ work. Of the 12 blocks on each ship which the company was originally to fabricate, three – the bow and two stern sections – will now be fabricated by Forgacs.

However, in their joint statement the two Ministers made it clear Williamstown was under much greater pressure then originally thought: the yard is also fabricating superstructure blocks for the Navy’s two new Canberra-class LHDs and will be responsible for completing these ships when the hulls arrive from Navantia. 

“The Melbourne BAE Systems shipyard remains stretched, working on two major projects at the same time – steel blocks for the Air Warfare Destroyers and the superstructure and integration of the Landing Helicopter Dock (LHD) ships,” the Ministers said. “The Government, the AWD Alliance and BAE Systems take the schedule for both these important projects extremely seriously.

“In February 2011, BAE Systems advised the AWD Alliance of potential schedule delays. Over the last few months, the AWD Alliance and BAE Systems have been working closely to develop options to improve the production program. In March, the Minister for Defence met with Guy Griffiths, the Group Managing Director - International of BAE Systems UK, in London to discuss this project.

“The Minister for Defence Materiel has also met with the CEO of BAE Australia, Jim McDowell, on a number of occasions about this project. Earlier this month BAE Systems presented the AWD Alliance with a plan to adjust its workload on the AWD Project.”

The AWD Alliance advised that if no action was taken to relieve the pressure on the Williamstown yard the first ship would be two years late, approximately 25 per cent over schedule. So construction of up to 13 blocks for the first two ships will be reallocated between Adelaide, Melbourne and Newcastle – seven for advanced fit out and six for construction.

BAE will complete the structural steel and initial outfitting work on the seven steel blocks it is currently working on, as well as all its work on the 14 blocks for the superstructure of the LHDs and the subsequent integration work.

Up to five of the AWD blocks will be reallocated to Navantia in Ferrol. ADM understands these will be keel blocks for the first two ships: these are the most challenging to fabricate, requiring the heavy steel plate of the ship’s bottom to be curved and shaped very precisely. The shaping is critical and the tolerances are very tight, and the technical challenge is compounded by the pipes, ducts and vents for electrical cables, air conditioning and fuel, water and sewage which run through these blocks.

ADM understands the production difficulties have been compounded by data issues with the engineering drawings that Navantia is delivering to the AWD Alliance and the two block builders from which the blocks are fabricated. This isn’t a criticism of Navantia’s design and engineering processes, which are acknowledged to be world class, and it’s understood there’s nothing wrong in themselves with the data in the 10,000 drawings.

The problem is the ‘tacit knowledge’ required to fabricate the blocks and create a functional warship: ADM has been told the drawings don’t incorporate changes to construction methods and the experience, know-how and ‘tricks of the trade’ accumulated by Navantia’s dockyard workforce and supervisors at Ferrol where the company has built four F-100s with a fifth under construction. The AWD Alliance has tackled the issue by getting an independent review team to check the drawings before they are issued. It has also negotiated access to Navantia’s own CAD data so that Australian engineers can conduct early interference and fit checks before cutting metal.

To be fair, much of the required knowledge and data can’t be included easily in a technical drawing. This type of ‘tacit knowledge’ is something that most technology transfer regimes must grapple with, and there is no substitute for personal experience. While Navantia has stationed advisors and experts in Australia to provide support, there may be no shortcuts up the dockyard learning curve in Williamstown, Newcastle and Osborne, and the project schedule seems to have made no allowance for this.

One reason Williamstown was awarded the block work was because the yard had previously done exactly the same sort of work to a very high standard on the 10 Anzac frigates. So what is different about the AWDs?

First, the workforce has had to be reconstituted after a lengthy lay-off following the completion of the Anzac frigate and RNZN Protector patrol boat programs: there is a technical and supervisory skills gap which clearly hasn’t been bridged yet. This is in large part the result of a structural problem right across Australia’s naval industry and won’t be solved until Defence and the Federal government decide it needs to be solved.

Secondly, BAE Systems is working with Navantia through a third party. On the Anzac frigate project the ship designer, Blohm+Voss (now Thyssen Krupp Marine Systems), was in a direct, contractual relationship with Tenix (who then owned the Williamstown yard) through the medium of Blohm+Voss Australia, whose engineers spent a considerable amount of time in Germany adapting the platform design for the Portuguese Meko 200 which was the baseline for the Anzac-class frigates. These engineers then returned to Australia and carried out the detail design to which the ships themselves were built.

Blohm+Voss made sure they were closely involved in the design and construction process -and worried they weren’t close enough. The company had carried out similar technology transfer and license construction programs in Argentina, Turkey and Portugal and understood the importance of a close relationship and unimpeded communication with the customer’s yard. Navantia hasn’t had the same experience of technology transfer and its contract was with the AWD Alliance, rather than directly with two of the three yards building the modules.

Third, the differences between Osborne and Ferrol need to be allowed for down at dockyard level: in Ferrol the F-100s are assembled on an inclined slipway and then launched before internal fit out is complete. At Osborne they will be assembled on a flat hardstand. This is the easiest and cheapest way to fabricate the blocks and install piping, wiring and machinery, and make any necessary repairs. Doing it after the modules are assembled increases the cost by a factor of five; doing it once the ships are in the water raises the cost by a factor of ten. (And that is why the AWD Alliance hasn’t stated a launch date in advance: it wants to keep the ships on the hardstand for as long as it can).

However, this difference in approach means dockyard expertise and experience are vital, which is one reason why drawing and manufacturing data is so important. This should have been recognised at the outset and schedule allowances made; if they weren’t, this speaks to the inexperience of both the AWD Alliance and Navantia in handling technology transfers.

ASC has managed so far to side-step some of Williamstown’s problems. First, the South Australian government helped establish a Maritime Skills Centre at Os-borne several years ago to prepare a new construction workforce for the challenge of the AWDs and subsequent projects. Second, as a member of the AWD Alliance it has been closer to Navantia in a contractual and physical sense.

Interestingly, Forgacs has largely managed to avoid the skills erosion problem. The industrial demographics around NSW’s Hunter River area have helped the company maintain a stable workforce and retain and grow key skills. The BAE Systems yard at Williamstown has enjoyed none of these benefits.

When projects come under technical or schedule stress any latent weaknesses of inception or implementation start to emerge. However, it’s important that major difficulties don’t blind stakeholders and observers to the solutions that are available and the corrective action that is being taken. 

Complex projects demand a constructive relationship between the major players with a focus on the project’s needs and goals. Difficulties encountered in previous Australian projects (the Collins-class submarine for example) have sometimes resulted in a breakdown in these vital relationships. That has had the fatal side effect of driving the players into confrontation, defending entrenched commercial positions which prevent any constructive resolution of the project’s difficulties.

If there’s any comfort to be taken from the Ministers’ announcement on 26 May it is that last year’s difficulties at Williamstown generated a relatively constructive response. They focussed the attention of the key players on the problems in the project, drew forth an honest assessment of the difficulties, their causes and solutions, and have resulted in what looks like a workable plan to deal with them.

In this one respect, at least, the AWD project is performing much better than others which have gone before. The challenge now will be to ensure the stakeholders continue to work for the good of the project; this will require some honest self-examination as well as some straight talking between the players.