THE MINISTER FOR DEFENCE
THE COLLINS CLASS SUBMARINE
AND RELATED MATTERS
Printed by CanPrint Communications Pty Ltd
20 June 1999
Minister for Defence
REPORT TO THE MINISTER FOR DEFENCE ON THE
Following your request to us of 12 March 1999, we have pleasure in submitting to you our report on the Collins class submarines and related matters.
In submitting this report, we should record our thanks to all those who participated so willingly, frankly and competently in our inquiries, without which this report would not have been possible.
We trust you will find it of use.
|Malcolm K. McIntosh, AC, Kt||John B. Prescott, AC|
TABLE OF CONTENTS
The Collins class submarines constitute, on the one hand, probably Australia’s most important strategic asset for the decades starting 2000, and on the other, Australia’s most ambitious and technically advanced defence industrial project ever.
Because of well publicised deficiencies in their performance, there is a widespread view that the submarines themselves are "duds" and the project is well over both time and budget.
Defence (the Navy, the Defence Acquisition Organisation (DAO), the Defence Science and Technology Organisation (DSTO), and the other elements involved), with the Australian Submarine Corporation (ASC) as the principal contractor, is now at the stage of having 5 of the 6 boats on order in the water and 3 delivered to Navy, but with substantial reservations as to their performance, fitness for purpose and conformance with the contract specifications. The latest boat to be delivered (the third) was some 2 years and 4 months behind the original schedule, but there have been no significant increases in the real contract or project prices.
Our aim was to examine the problems with the submarines and make recommendations on:
We were not asked to chart in detail how we got to where we are or to assess the historical roles of particular individuals or groups. Rather, we were asked to advise the Minister on what he might best do with the situation he inherited, so we have focused our efforts on the future more than the past.
To gather data and test hypotheses we have sought to talk to or visit all the major players involved directly in the project, as well as some of its critics and other interested parties. In all we have had some 42 visits and discussions and received 15 submissions. We believe this has been adequate to provide a balanced picture of the project and its prospects.
There have been many other inquiries into the project (some highly classified and some, such as the Auditor General’s report, public) and we see little value in repeating much of the detail and history that others have documented. In considering those reports, however, we became aware of failures to act on some prior reports, even where they were generated out of concern that remedial action was required. Where action was taken, more might reasonably have been expected. In other cases the reports were not referred to sufficiently senior levels to put remedial action beyond doubt.
Notwithstanding the well-publicised technical problems, much good work has been done. In fact almost everyone we met was totally committed to the project and doing what they believed would bring maximum benefit - at least within their terms of reference and responsibilities. Rather the difficulties we will describe stem from a lack of overarching capacity to deal with the scale and complexity involved, given the changes in mission and technology that should have been recognised as inevitable in a project of this ambition and duration.
Australia sought, for good reason, to undertake a project well beyond its previously established capacity in Defence, but despite a great deal of success in some aspects, has so far not established the total organisational and systems capabilities nor some of the individual competencies required. We will make recommendations for further actions that we believe will result in the Collins class becoming fully effective. They will also be relevant to achieving success in future major Defence projects.
2. WHAT’S WRONG
The essential and the visible problem with the Collins Class submarines is that they cannot perform at the levels required for military operations. The underlying cause is a myriad of design deficiencies and consequential operational limitations relating to the platform and combat system.
As a result of the operational limitations, it is becoming increasingly difficult to sustain the very high levels of crew training and morale needed for submarine operations. This situation is exacerbated by the timetable for taking the Oberons out of service, such that there is a serious gap in Australia’s continuing submarine capability.
Behind the technical problems and delays are a series of commercial and contractual constraints and managerial shortcomings.
The Technical Problems
We are not especially critical that there are some technical deficiencies. To some extent it is inevitable in a new class of equipment as complex as a submarine that there will be design deficiencies - things that get overlooked or really emerge only when the entire boat is brought together. That said, we have been astonished at how many there still are some 6 years after the first boat was launched, the range and extent of them, the seriousness of some of them, the areas in which they have occurred, and how slowly they are being remedied.
As each of the boats completes its trials, Form TI338A is issued, which lists all the defects. Many are routinely remedied, but some have proved to be more systemic.
Not all is negative. Things that were originally regarded as high risk (eg, the all-Australian build, making the high tensile steel, the welding carried out in Australia, but not Sweden, and the ship’s control and management system) have exceeded expectations. There has been much high quality work carried out and the internal layout and the house keeping of the boats are of a high order.
Further, some things that were seriously defective have been fixed. For example a significant and difficult issue with the propeller shaft seal has now been resolved.
The Propeller Shaft Seal
Excessive leakage (exceeding 300 litres an hour) in the primary mechanical seal was experienced, especially at depth, during contractor’s sea trials on HMAS COLLINS. The design was assessed as not robust enough for the power, speed, depth and pressure ranges required.
The Submarine Project also found considerable dissatisfaction amongst other Navies using the same seal (poor leakage rate and excessive corrosion). Many Navies were replacing the seal with one similar to that used in the Oberon class submarines. ASC continued to argue that the specification was excessively low and could not be met in practice, but the Project Office continued to press the issue.
The shaft seal sub-contractor ceased to manufacture and was taken over by the company that manufactured the Oberon seals. A new seal, similar in design to the Oberon seal, was installed and tested on HMAS FARNCOMB in 1998. It achieved a leakage rate 5 times better than the specification required and more than 100 times better than the seal originally fitted.
All other boats are now fitted with the new design seal.
The most serious of the issues that have yet to be remedied are: the diesel engines, noise, propellers, periscopes and masts, and the combat system. Some of these are interrelated. There are differences in opinion as to the detail and significance of some measurements of these deficiencies and the importance of having to depart from some contractual specifications, where these are no longer relevant or are unobtainable; meanwhile, other, difficult to achieve, criteria have been enforced.
To our surprise, deficiencies have occurred in items, that should have been relatively straightforward, had testing, even along the lines of that routinely applied to merchant ships, been undertaken.
The Fuel System
The fuel system has proved to have unwieldy arrangements to separate the compensating seawater from the fuel oil in the tanks and is not easily managed by the submarine’s engineering crew.
As a result, seawater has entered the diesel engines causing inefficient combustion (and possibly vibration), corrosion of fuel injectors and pumps, and eventual seizure of pistons in cylinders. It has also limited the endurance of the submarines below that specified in the contract as the crews have had to leave about 30% of the fuel in each tank to prevent water contamination of the diesel engines.
If the design of the fuel oil system is appropriate, the operational constraints involved in operating it to the design criteria have not been sufficiently impressed on the ships’ companies. On balance, it seems inappropriate for such constraints to have been required. The system should have been more flexible.
The Diesel Engines
The Collins class submarines are each fitted with three diesel engines. They are a specially manufactured 18-cylinder version of the Hedemora Turbo Charged B210 series engines. Hedemora engines are used in the off-shore oil and gas industry, for emergency generators with high responsiveness requirements and locomotive applications.
However, in the Collins class, the diesel engines have proven to be unreliable and far from robust. From delivery of HMAS COLLINS and HMAS FARNCOMB until October 1998, more than 750 defects were recorded against the diesel engine systems.
They suffer continuing ancillary equipment failures, thought to be due in part to excessive vibration partly due to water contaminated fuel. Failures have included broken pistons, broken gear trains, seized fuel injectors and fuel pumps, broken rocker arms and generator couplings and, most recently, a damaged crankshaft that must be replaced. All the submarines have been affected. One has suffered extensively; others to varying, sometimes serious degrees. Repairs have taken an unacceptable time.
Others of the problems have been caused by poor design modifications to the 18-cylinder layout, poor quality control in manufacture of components and assembly, and poor subcontractor back up.
There has also been a high consumption of lubrication oil, which prevented the submarine achieving its endurance requirements. This now appears to have been rectified by using stronger scraper rings.
In overall terms a recent report by Lloyds Register observed, "The number of diesel engine related failures sustained onboard HMAS FARNCOMB and indeed the Collins class submarines as a group is totally unacceptable to the point of being quite unique in modern medium speed diesel marine applications."
Clearly not all the implications of transferring a diesel designed for other purposes to a military marine environment were fully evaluated. Nor was it understood that entrusting the supply of the main diesels to a relatively small company, Hedemora, would so constrain efforts to fix any problems that did arise. Further, this situation has been compounded by ASC restricting the range of subcontract expertise it would utilise.
In our view, the problems with the diesel engines will be resolved but they are not just ones of engineering detail. They have prevented full trials being progressed as scheduled in respect of other issues and have prevented a full run up of crew training. They have diverted technical expertise and management time from more challenging areas and thus are more serious than at first appears.
The effectiveness of a submarine is highly dependent upon its noise signature. It affects operations in two fundamental ways. First, a submarine’s radiated noise determines how easily an enemy may detect it and, second, it reduces the efficiency of the submarine’s own sonars, making it more difficult for the submarine to detect the enemy.
Noise is thus a fundamental and extremely sensitive characteristic of a submarine. We are astonished and appalled by the extent of information leaked into the public domain on the Collins class noise characteristics.
It is also a complex issue as it must be assessed at different modes of operation of the submarines (diesel or battery), different speeds and depths, across the whole frequency spectrum, against different sea states and against the characteristics of the detectors that are likely to be used against the submarine.
Acoustic deficiencies in the Collins class have been suspected for some years but have only recently been confirmed due to the maturing of the Navy’s acoustic test range. Even then, while some have been resolved, others are outstanding.
Those unresolved include the operation of some onboard equipment, the propeller and the flow characteristics of the hull. The propeller does not meet the specified cavitation requirements of the contract. Furthermore the propeller is excited by disturbed flow caused by inappropriate fin and casing shapes which also separately constitute a noise problem.
There is also an issue as to whether the noise deficiencies should be compared to contracted requirements as distinct from the Navy’s current operational requirements. Certainly ASC is interpreting its contractual obligations as narrowly as possible in this area and this is symptomatic of issues between ASC and Navy in some other areas. The fact remains that the submarines have not complied with the contract and in some areas the Navy requires an even higher standard.
In our view, the fact that these noise issues have arisen reflects some serious shortcomings.
The cavitation and flow characteristics problems reflect inadequate tank testing on the hull. In fact, it seems this was less than that carried out for many merchant vessels.
Also the origin lies at least partly in specifying not merely performance criteria for the propellers, but also in requiring them to be cast in a metal that to date has proved unfit for the purpose.
However, it is currently expected that modifications in fibreglass to the fin and the hull together with new propellers and some other tracking down of mechanical noise sources will resolve these issues - possibly by year end.
The Collins class propellers suffer from fatigue failure as well as from the cavitation problems just described. In part this results from the propellers being specified to be cast in a special material called Sonoston for its unique properties of sound dampening, which reduces resonance (but not cavitation). However, this material is also brittle.
This situation has in turn been aggravated by a substantial increase in the time the submarines are spending on the surface, which puts more strain on the propellers. First, the change in homeport from Sydney to Fremantle requires more surface transiting before the submarines can dive and, second, the vessels are doing more transiting to and from Adelaide because of the need for additional trials and repairs, along with an even longer time to dive off Adelaide. We believe the propellers should be more than robust enough to cope with this situation.
The use of Sonoston may also be a factor in the propellers not meeting the specification for cavitation, as they are extremely difficult to manufacture. Due to the properties of the material they are currently hand finished and are not of consistently high precision. In turn there is potential for reduced power performance.
In any event, aside from noise issues, the propellers have failed from cracking.
HMAS COLLINS’ propeller requires to be replaced before further service, but it is a class issue, which requires resolution on all follow-on submarines.
Again this is a serious issue as a catastrophic failure to a propeller will put a boat at risk even in non-hostile circumstances. Its origin lies in the factors we have described. It will be fixed, but on all boats this may be an extended process and it is of sufficient concern to have caused the Navy to seek US assistance outside the contract with ASC.
While recognised subcontractors (Pilkington - Barr and Stroud) were chosen for the attack and search periscopes, the design requirements imposed have impaired their operational performance in a manner not contemplated, nor fully evaluated.
Severe vibration is experienced on the attack periscope at moderate speeds when operating the submarine below periscope depth with the periscope fully extended. The vibration is due to eddy shedding (turbulence) on the top of the periscope. It is thought to be a design problem associated with the unsupported length of the periscope tube.
Also, focusing problems are experienced in both periscopes when changing magnification from one power to another. This affects the periscope operator’s vision, which is a ship safety issue when estimating and calculating the closeness of targets.
A grey band across the middle of the field of vision of the search periscope (called the "double dove" effect) affects the periscope operator’s ability to detect targets, both air and sea, at a distance. There is also an internal reflection of the sun’s image at certain elevations and azimuth angles, which can temporarily blind the operator. This significantly reduces the ability to keep a safe periscope watch.
There is not yet a fix at hand for these items. While in the short term the Navy might accept periscopes which are modified to reduce the vibration, focus, double-dove and sun-spot effects (by modifying operating procedures), a complete redesign may be necessary. Some compromise with the specifications might be necessary to allow a proven, off-the-shelf design to be used.
When the prime contract was signed, the Navy thought that it did not have a need to equip all boats with satellite communications. In retrospect, this was an amazingly short-sighted decision and now must be rectified.
Additionally, the communications mast has proven to be unreliable.
Communications with headquarters is therefore slow and cumbersome, placing the submarine in a very vulnerable position while it is transmitting.
It is a problem relatively easily addressed.
The Combat System
The combat system is at the heart of the issues associated with getting an overall favourable outcome for the Submarine Project. The level of performance achieved with the combat system to date is unsatisfactory and this situation will not be resolved unless new commitments are made to change the present direction. Whilst some interim arrangements organised by the retiring Chief of Navy outside the contract with ASC will help, there is in our view a need to go much further than that.
Combat systems (along with most other large, software-intensive systems) have a notorious history of disasters. The main problem is the extremely rapid rate of technological change, which can give rise to new technologies which could do the job far better emerging during the course of the contract. As a result, most serious procurement agencies are experimenting with very different procurement strategies for software-intensive projects.
By including the combat system with the platform in the single prime contract, with a unique military specification, Defence left itself wide open to such technological problems. Others (eg, the USA and the UK) have done much the same with their naval combat systems so it is hardly surprising Australia did likewise and we are not being critical. The difference is that they have recognised the problem and have taken the major step of cancelling that route and moving to a much more reliable route based on commercial, off-the-shelf technology (COTS). Instead we persisted as follows:
Although known about earlier, the problem was first seriously highlighted in 1993, but the opportunity was not then taken to make fundamental changes in the configuration of the system or in the choice of subcontractor. This has had several regrettable consequences:
The descoped combat system as now contracted from ASC and its principal subcontractor Boeing is severely restricted in classification, "track" management and weapon control, and is considerably below satisfactory capability for an operational submarine. Further, the Navy believes that it is behind the revised schedule for March 2000 delivery and that it will still not be operationally acceptable. The US Navy backed augmentation program will provide short term relief for one or two boats (subject to additional funding being approved).
Basically the system does not work, the quality of information from individual sensors has been compromised and their display on screen is inferior to that of the signals actually processed. Relatively routine interrogation of targets causes failures in the displays and inordinate delays occur in bringing multiple sets of information together in the manner planned. The number of targets that can be dealt with at one time is far less than specified or required. In fact the tracking, classification and display of sonar targets is less effective than on the Oberon class. No overall satisfactory solution is yet committed.
Some details of the shortcomings follow:
Due to the "integrated" architecture of the design, the processing of sonar information is split between the sonar equipment, which processes the data and the tactical data handling system (TDHS), which controls, displays and allows the data to be manipulated. Because the full capability of the TDHS is not yet available, only about 25% of sonar functionality is currently available to the operators even though all of the data is being processed. This includes sonar functions, which are essential for use in operational patrols.
The combat system has been designed so that each multi-purpose console should display information from the sonar sets, radar, electronic support measures, periscope video etc. The design also intended that many functions would be performed automatically, thereby reducing the load on the operators.
For reliability, the intent was that if one console failed, the functions would continue to operate on another. To achieve this, a distributed operating system was developed, which required the replication of data to all consoles. This introduced another level of complexity - concurrent access to data by all console computers and prevention of consequential computer deadlock.
Information transfer has slowed to such a degree that the multipurpose consoles are inadequate and too cumbersome for the operators. Similarly, the automatic processors are proving to be inefficient and actually increase, rather than decrease the operator load.
The slowness and cumbersome nature of the combat system and lack of effective automatic processors do not give the submarine commander the "edge" he needs on an operational patrol, to remain undetected or, if fired upon, to react swiftly enough.
The Seriousness of the Shortcomings
Despite their substance and duration, we have found few design deficiencies for which a remedy is not fairly readily available, although some will be substantial and costly. A major issue, which has probably held up progress more than any other, is whether these costs will fall to the contractor or the Government.
Noting the sensitivity of the performance of the submarine, we do not propose to quantify or discuss further the operational consequences of the deficiencies. What is however, the most obvious and debilitating consequence is a very low level of reliability. Not only has no submarine yet gone to sea with anything like its full complement of operational capabilities, but each invariably returns with even less. The rates of failures are far too high and are extending the trials and limiting the extent to which the crews can practice tactics rather than merely work around the failures.
We have considered carefully whether the submarines could be "sent into danger". Obviously this is a relative question, depending on the one hand, on how complex and technically advanced the threat to the submarine would be, and on the other, how important a military contribution the submarine could make. In our view, the circumstances would have to be extremely serious indeed to risk the submarines in their present state.
The situation is actually worse than this because, in anticipation of the Collins class being fully operational, the Oberon class has been run down to the point where there is only one operational boat, HMAS OTAMA, which will soon need a long and costly major refit if it is to remain in service. Not only is there a risk that there will be a significant period when there will be no fully operational submarines available, but there will be a period when Australian submariners will get no operational experience. Such experience is slow and difficult to restore. It is extremely important that the Collins class be made fit for purpose quickly to avoid such a debilitating hiatus.
That said, it is also our view that, once the deficiencies are remedied, the Collins class has the potential to be an extremely potent strategic and tactical defence asset for Australia, which should serve the country well and retain a technical edge, with periodic upgrades, for decades.
3. DOES IT MATTER?
At the most obvious level, Defence has spent at 31 March 1999, $4742 million or some 93% of the expected total project cost (including $4310 million, or some 98% of the expected total contract price), for which there are five boats in the water, but none performing anywhere near adequately. As well as funds set aside for the project, some $2.6 million of an anticipated $90 million has also been spent, (and some $23 million of the $90 million has been committed) mostly with the US Navy as a hedge and in anticipation of upgrading the boats well before it might otherwise have been necessary. Clearly, there is an obligation to Australian taxpayers to get value for money in the form of operationally effective submarines, and through them, to get a commensurate contribution to Australia’s defence.
Australia’s defence policies have rested, in recent times, not only on the obvious notion of the defence of Australia and our direct interests against a potential aggressor, but also on the benefits through trade and investment of a stable and increasingly prosperous region. Australia, being a substantial developed democracy, anchors the south of the region and, with the right structure of its forces, can help to discourage instability and its escalation, without being seen as having either expansionist capabilities or intentions of its own. The key in all instances is to maintain a technological edge in certain military capabilities. From past history and common sense, such a position by Australia will not be seen as a threat or as encouraging a regional arms race.
Within this context, the F-111 aircraft and the submarines can be seen as the two substantial strategic assets in the Australian inventory. Neither are capable of threatening invasion, but both, in their different ways have enormous capability to limit expansionist capabilities of others and therefore help to deter any such aspirations. Considerations of this kind justify seeking to make the Collins class submarines extremely capable platforms with a significant edge over anything else in this region.
This means that not only do we need to fix the current problems as soon as possible, we need to do it in a manner that maximises the longer term potential of the Collins class.
4. WHY ARE THINGS SO WRONG?
The analogy most often used to describe to us the Submarine Project is World War I -- each of the parties is said to be firmly entrenched in a bunker and repelling all approaches. While, like most analogies, this contains an element of exaggeration, the positions of the parties (the operational RAN, the procurement project office, the in-service support project team, the prime contractor and the principal sub-contractors) are certainly far more antagonistic, defensive, uncooperative and at cross-purposes than should be the case in a project like this.
In any large, complex, multi-party project, there are, of course, differences of opinion, which are often quite hard fought. Usually they are resolved by the common view of the aim of the project and a liberal dose of common sense and good will. Even when the issue of who bears the costs is ultimately resolved in court, there is often agreement to get on with the work in the meantime. For various reasons, little of this urgency and pragmatism seems present in the Submarine Project.
Successfully completing a project of the scale and complexity of delivering the Collins class submarines into operational service requires:
We see weaknesses of varying degrees in all these aspects.
The aims of the project are far from commonly understood and applied. At the highest level, the original aim of the project was to produce 6 submarines with well trained crews and supporting infrastructure capable of performing their strategic role (as it was then understood) with at least two boats on station well distant from Australia, more or less indefinitely. Given the growing sophistication of the region in which the submarines were to operate, this required a significant technological edge over existing capabilities. All of this was reflected in the specifications drawn up at the time (around 1984).
Since then, technologies have changed, the region has changed and Defence’s ambitions for the boats have changed accordingly, but there has been no sensible mechanism for incorporating such changes into the contract.
Thus there are now considerable differences between the parties over what is really required. At one end are adherents to the original specification, which is not being reached anyway. At the other, are those who seek some variant that drops things that turn out to be more difficult or less important than originally thought, and includes new things that are considered more important.
Simplifying somewhat, Navy has an urgent need for adequate boats, a requirement for some upgraded capabilities now, along with a need to retain crews and a leading technological edge in the longer term. These requirements can be met only by a degree of flexibility in how the submarines are procured and supported, which is certainly not available in the present arrangements.
ASC on the other hand has no motivation to provide more than what it interprets as its contractual obligations, especially when the Commonwealth has established it will not pay more than the original contracted price. It is also peeved that the Commonwealth chose the combat system contractor but then did not act as firmly as it wanted when the original non-performance in that area became apparent.
Moreover, this is aggravated by some other perception issues. In complex situations different individuals view issues not just on the basis of the apparent facts before them but on how they perceive those facts according to their own qualifications, vocation and experience. Such factors are the filters which give people quite different perspectives of the same situation. People behave according to their capabilities and according to what their experience tells them, along with their understanding of their obligations.
In the present case the project is dominated by Kockums (as principal shareholder of ASC and the submarine designer). Kockums are experienced in building boats for Baltic operations under instructions from Swedish authorities, and have a different view of what is expected than they might have with experience in building vessels for long transits and other operational profiles for different authorities. It is hardly surprising that there have been conflicting interpretations placed on the same events, and this has been a factor in getting a poor common understanding of the project aims and in not getting issues resolved quickly between ASC and Navy.
Caught somewhere in the middle is the Project Office with the role of acting on behalf of the Navy and administering the project. While well cognisant of the operational Navy’s wishes, it is also concerned that anything more than very minor amendments to the contract could let the prime contractor "off the hook" and lead to substantial blow-outs in time and cost. It is under pressure from the Navy to modify the submarines and from the prime contractor to make concessions and alterations. While working hard on the immediate issues, it seems to be poorly placed to resolve such issues at the macro level.
Though some have tried, no one has stepped in to effectively fill the void in setting continuing overarching mission objectives. This is a senior management deficiency with important consequences.
Finally, the sub-contractors feel themselves excessively constrained by the now-aged specifications and by the prime contractor. They are also in a vacuum as to what some of the real priorities are.
Competencies, Organisation and Management
More cohesion about mission objectives would have focussed more attention on the need to assemble all of the individual skills and competencies required and on arrangements for coordinating these.
The technical competencies required for the project are generally available, but in certain areas, have been made unavailable by the structure of the contract or the interests of the parties; or they have not been applied because of an overly rigid adherence to an input specification rather than a performance requirement. The former has been a factor in delaying attention to propeller shaft seals and diesel engine issues, while the latter has affected resolution of combat system problems.
These are worrying management issues but they are also organisational ones. Closer coordination of the various Defence interests to produce the final result is an imperative. For example, the Defence Science and Technology Organisation’s role has not been sufficiently defined, despite their great deal of expertise in relevant areas; an earlier and more formal role might have been expected. However while fragmentation and distribution of responsibility has contributed to the lack of a full response in some instances, even without this complication it remains true that insufficient action has been taken to overcome the issues we have described.
This is especially the case given the number and clarity of previous reports highlighting problems with the Collins class. These reports alone should have brought home to Defence management at senior levels a full understanding of the shortcomings to be addressed, the need for better coordination to achieve this, and for more intensive follow up.
The point we are making here is not that management difficulties or lack of skills in Defence or Navy are responsible for the fundamental failure of the contractor to deliver, but rather weaknesses in management have resulted in issues not being addressed as they should have been.
It is possible only to speculate on what other factors may have contributed to more not being done, but these may include:
This is a damaging situation for Defence that must be addressed by internal changes and by recognising that something must be done to improve relationships with ASC. The latter, however, requires, as much as anything, that it be brought home to ASC that its future depends on a change in its approach - a factor which the chairman of Celsius (the parent of ASC’s principal shareholder) effectively acknowledged in his speech after the launching of SHEEAN.
This brings us to the point that the contract is, in our view, no longer fully effective as a mechanism for bringing the parties together.
This is a complex area as the reasons for the Commonwealth's approach to setting up a fixed price contract are readily understood. At the time the contracting strategy was devised, Defence had experienced a long and generally increasingly costly period of cost-plus contracts and there was a trend elsewhere in the world for open competition for fixed price contracts.
Aside from the protection a fixed price contract gave the Commonwealth, it was also seen as giving the contractor an incentive to complete the contract more efficiently and earlier, to keep both the direct savings and the implicit interest.
For a relatively routine product or one where the specifications are clear and unambiguous, and where payment is made mostly on delivery, this can work well.
However, for a large, complex and new project, for which a design does not exist in detail, and for which generous up-front payments are made, its effect can be deleterious. Particularly in the later stages, it can encourage the supplier to contest the specifications and their interpretation and to avoid responsibility wherever possible to protect profit. Conversely, it can encourage the buyer to incorporate everything possible into the interpretation of the contract.
Fixed price contracts also tend to put substantial financial, and therefore technical, barriers between customers and sub-contractors. These deter the prime contractor and can prevent the customer from seeking an alternative sub-contractor with greater or different expertise part-way through the contract. Further, many sub-contractors are specialists in important aspects of the operation of the submarines (eg, sonars, communications), where the fine-tuning of equipment and the tactics for its optimal operation can be closely interrelated. The contract should provide for such relationships between the users and the sub-contractors.
Difficulties in these areas and adversarial relationships epitomise the Submarine Project and are clearly limiting and delaying the remediation of the design deficiencies.
In other respects the policy, commercial and contractual arrangements in place for this project are somewhat unusual and they have adversely affected outcomes in various ways.
The concept that risk should be mitigated for the Commonwealth by the contractor having to take on performance guarantees is frustrated by the ineffectiveness of the Commonwealth enforcing such guarantees against itself as 48.5% owner of the contractor.
The fixed price aspects previously discussed are aggravated by the unsatisfactory provisional cost (PC) terms in the contract. The amount for PC items was limited because of the fixed price philosophy, but it has also proved to be inadequate to cover items identified but not costed at the date of the contract and also items that were not able to be fully specified at that time. This has eroded the limited contingency of the original contract.
In any event, that contingency was quite inadequate at 2.5%. Even though the Commonwealth funds the impact of variations in exchange rates outside the contingency, a contingency at this level is unrealistically low for a project of this magnitude, complexity and duration. A figure in the range of 10-15% would be normal in the private sector but it would have to cover exchange rate variations as well.
By imposing its selection of the combat system subcontractor on the principal contractor the Commonwealth started out on the wrong foot with ASC and this situation only became aggravated as the Commonwealth agreed to successive modifications to the contract to reduce the performance requirements of the combat system as we have described. Being a close witness to the original contract not being enforced against Rockwell/Boeing has not helped ASC’s attitude in negotiations over other obligations.
More seriously, the structure of the contract is one which imposes both performance criteria and some detailed specifications as to how this performance should be achieved. The latter is not the appropriate way to enforce the former and it has been a factor in the difficulties with, amongst others, the combat system, propeller, and periscopes. It was aggravated by the lack of insistence on all the testing required to reduce risk, notably full tank testing of hull models and early checking of more routine equipment to avoid unnecessary problems.
In summary, the primary prerequisites for a successful project -- clear and shared aims, competent, committed and well managed parties, properly communicating, and brought together by an appropriate contract along with related mechanisms to bring it all together -- are all missing.
All these things are capable of being addressed but this is not yet happening as fully it should.
5. WHAT SHOULD BE DONE?
We will deal with each of the matters we have raised in turn. Basically we expect the technical issues in respect of the platform will be resolved by a combination of efforts by ASC, the Defence Acquisition Organisation, Navy, DSTO and the US Navy -- but there will need to be more effort to act in concert.
Those in relation to the combat system will require action beyond that currently in hand and we will propose this.
We will also propose actions within Defence to deal with the other matters which have affected the way the project has developed and in a later part of this report we will address the question of the implications for further major Defence procurements.
The Diesel Engines
The Navy, DSTO and ASC are working to determine the causes of the vibration and reduce it to acceptable levels. This involves a range of measures including finding the optimum engine running speed.
ASC is reviewing its diesel engine sub-contractors’ quality systems. Replacement gear trains are being manufactured and installed by the contractor, and are being supplied with proper quality evidence.
In the short term, the Navy will continue developing and testing fuel-water coalescers. In the longer term, it is expected a more manageable fuel supply system will be designed, but ASC is not yet committed to this.
The diesel engine contractor, Hedemora, is providing training in Sweden for its Australian support contractor. This together with efforts by Navy’s improved Quality Assurance "watchdog" is expected to improve reliability.
ASC is testing an oil scraper ring modification on the diesel engines at sea and, if successful, will install the modification to all other engines.
The success of these initiatives depends on a continuing commitment by ASC to resolve the issues. It will also be desirable for the Navy to utilise its own resources and those of DSTO to ensure a fuller understanding of the rectification measures which should be adopted.
Noise and Propellers
ASC is redesigning and manufacturing some mechanical parts, which emit noise from a number of systems.
The composite fibreglass casing structure at the fin and aft near the towed array winch, are being redesigned to improve flow around the hull utilising DSTO and USN facilities. Modifications to HMAS COLLINS are being fabricated and installed. These will then be tested at sea on the Navy’s acoustic test range and once proven the new design will be fabricated and fitted to the other submarines.
ASC is currently producing a second-generation propeller, which may assist with the cavitation requirement possibly at the expense of some radiated noise.
Also as a hedging strategy, the Submarine Project is funding the modification of one of the current propellers and an ex-USN submarine propeller in the US, to improve cavitation and radiated noise performance.
The Project Office is using project contingency funds to progress this work pending approval of additional funds under Project Sea 1446, which is the augmentation program to also achieve some higher interim capability in the combat system.
This work will alleviate some of the deficiencies in performance and approach, to which we referred, but success again depends on ASC's continuing commitment to the task along with the assessments made at DSTO and by the USN. The estimated date for resolution of noise issues following testing on HMAS COLLINS is 31 January 2000. The rest of the class will be upgraded in turn.
We think these efforts will be effective in meeting immediate requirements but the nature of the work on the hull form and propellers is such that there is likely to be requirements for future funding to allow further iterations of design. This should not be a cause of criticism and in fact it should be expected there will usually be continuing studies to improve the acoustic signature of these assets.
Modifications are being made to the periscopes to reduce some of the problems mentioned. These will improve performance but more substantial replacement work may be required later at additional cost.
A minor project is in progress to fit the submarines with satellite communications. Design changes to the communications mast are being progressed by the contractors.
While there are apparent remedies for each of the platform deficiencies, they will require vigorous management if they are to be remedied in a more timely fashion than has occurred to date.
The Combat System
This is the principal technical challenge still to be resolved. In dealing with the issues the Chief of Navy and other senior staff have sought to do two things. Firstly they have put in place an interim fix under Project Sea 1446 and secondly they have examined what more substantial arrangements should be made for the future.
It is clear that the interim fix is just a short term band-aid to get some necessary capability quickly and that the modifications being made to the contracted "integrated system" will not lead to a satisfactory outcome. In fact the Chief of Navy’s assessment recently was, "The history of the development of the CS software since 1993 further indicates that it is unlikely that the CS contractor Boeing Australia Ltd (BAL) will ever be able to deliver useful CS software using the present architecture, particularly the architecture within the TDHS (Tactical Data Handling System)." 4
There is widespread support for this view. Put simply the Boeing system being installed cannot be transformed to an effective performance level without much of the software and some of the hardware being replaced or superseded. Moreover other planned submarine enhancement projects will be too late to provide a remedy.
The challenge now is to keep the submarines in service whilst the software package and some hardware is progressively changed from the original installation to that which would be installed today if starting out afresh.
In our view, Defence should move quickly to do this, reducing as far as possible expenditure on the Boeing system from now on. A preferred new system would be configured with less integrated architecture and would utilize more robust commercial off the shelf (COTS) equipment that can be replaced independently as technology advances. It would be simpler in that only the display output rather than the raw data itself would be distributed to the different consoles.
This could be achieved by:
Quite modest cost and time estimates have been given to us by potential contractors for this work. There would nevertheless be some risks in proceeding in this way, given the negotiations that would be required with Boeing, the problems in deferring further the dates scheduled for having Collins class boats in service (but possibly not realistic dates) and the lack of Australian experience with the equipment that would be ordered. However several contractors are prepared to put their reputations on the line and the "virtual submarine" would provide a valuable test bed for any proposal.
There are several possible contenders. Boeing might well bid and might have an advantage with some aspects of what it has done so far. CSC, who have written most of the key applications software and who have a version running on commercial technology, might feel able to prime, perhaps with some help from the USA. Thomson and Sonartech (with Krupp Atlas) have both also indicated a willingness to consider standing as primes. In writing a new contract for the combat system alone, it should be possible to incorporate the flexibilities common in modern software-intensive projects and any such system seems certain to be a substantial improvement over the monotonically downgraded variants of the original specification now on offer.
It is, of course, difficult without firm quotes to be precise about how much it would cost and how long it would take to install and set-to-work a new system. It should be possible to recover some of the cost from the present contractors, both as a share of what is left to pay and as compensation for failure to deliver to time and cost and the damage that failure has done to the Navy. It would certainly be affordable if the contingency provision were increased to more normal levels.
Operational Requirements and Contracts
On balance we believe it is appropriate to proceed in three parts for the present project and to adopt some different arrangements in future:
In the longer term, there will be opportunities to improve the platform beyond that in the current specification in respect of its acoustic profile and other operational characteristics. In many of these areas, the US and UK are undoubted leaders. Defence has wisely opened links to both, with most effort so far being undertaken by the USN. Some of this work will require the exclusion of other nationals, which will constrain the future role of Kockums. While no doubt attractive, care will be needed to ensure that such work does not undermine the contractual obligations of ASC to produce contractually satisfactory platforms and related services.
The main argument put to us against the approach we are recommending is that any major change to the contract would allow the prime contractor to negotiate substantial delays, price increases and weakening of the specifications and warranties. We are not convinced. In implementing these arrangements we would expect to release ASC from its obligations under the contract (on price, delivery, warranties etc) in respect of the combat system, but not otherwise. We would seek to maximise the relief from Boeing given there is little point in them incurring any but the least possible expenditure on the current arrangements. In fact some allowances have already been settled with Boeing and there may be scope for more.
It is our view that for coherence and an orderly introduction into service, the Submarine Project requires a much stronger commitment and involvement by the Navy from now on, which would bring together procurement, in-service support, the personnel authority and the operators. We recognise the different competencies required by each of these functional areas and consider a full merger of the organisational structures for these areas inappropriate.
Rather, as the officer with the responsibility for the ultimate outcome - warfighting with the submarines - we think the Chief of Navy should chair a committee, comprising the most senior officers from each of these functional areas. The Submarine Committee would meet monthly to identify and solve problems and to plan for all aspects of the orderly introduction of boats into service.
In practical terms it is difficult to see how the non-project management elements of the total task can be coordinated by anyone but the CN. And while much of this work will be delegated to others, it is the CN who is the procurement team’s customer and he needs to be very active at the interface. He cannot assume he will get what he needs without being involved.
We would charge the CN to work closely with those with overriding authority to plan defence capability and with the head of procurement. These responsibilities should be actively in place from the earliest stages of any project and would intensify well before planned delivery.
The interface must be a continual and constructive one to ensure proper regard is paid to alternative means of achieving desired outcomes and to ensure progress is effectively monitored so issues are dealt with in a timely fashion.
It must be ensured that there is "buy in" in operational areas, the project office and amongst the key contractors to the overall mission.
Further, given the importance of the submarines to Defence, the size and complexity of the project and with its current problems, we think the project would benefit from a more senior project director, who would be better placed to resolve issues on his or her own authority and to attract direct attention from the most senior officers of the Department. This may seem excessive, but, provided it is not allowed to bypass the rest of the Defence system, seems warranted by the importance of the submarines to Defence. We would take other steps as well, outlined later, to strengthen the Defence Acquisition Organisation.
The Role of ASC
In the short term there is a clear need for the Commonwealth to reassert its expectations to ASC, both as a customer and as a shareholder. However we have little confidence that this will of itself deliver the changes needed. Better definition by the Chief of Navy and the Head of Procurement of what they expect by way of resolution of technical issues, additional performance capabilities and in-service support will certainly help. In the end, however, ASC’s future will depend on what it does for itself.
In relation to in-service support three points emerge:
While the 1990’s have seen two large indigenous shipbuilding projects occurring simultaneously (the ANZAC ships and the Collins submarines), future decades will see much lower levels of shipbuilding as other defence assets are needed (eg, replacement combat aircraft are already being bid for by the Air Force). It seems inevitable that there will be a shakeout in shipbuilding capacity in Australia to two, or more likely one, major yard specialising in defence work. Even then, it is likely to depend to a large extent on upgrades and repairs and maintenance for its viability.
It is not for us to speculate on which yards, much less which owners, might go and which might survive. We merely note the various views put to us that there are long-term difficulties in sustaining such heavy manufacturing activities (as opposed to the rather less intensive and even picturesque aspects of the fleet base) in central Sydney, and that Williamstown is "gentrifying" such that there are likely to be other, more profitable uses for the site. Neither of these features may be decisive, and we strongly recommend that the Government leave it to the market to sort out.
Particularly, however, if only one yard is ultimately sustainable, it is extremely important that it remain in majority Australian ownership and untied to overseas companies, except for particular projects. While it may be tempting to create the widest possible field of buyers and to encourage overseas investors, we believe that, if a monopoly or near monopoly does develop, and it ties Defence to a single overseas source, Defence will pay many times over in subsequent contracts for any increase in purchase price for the Government’s assets. The sale of ADI and the AIDC shares in ASC should be managed with this in mind.
It will also be a factor in determining whether ASC should have some substantial Australian engineering company injected into it or some other structure formed by the market to take over the real and intellectual assets, to provide on-going support of the submarines, and for any further build program.
In the short-term, no changes should be made that would allow ASC to escape its contractual obligations. In the longer term, which alternative would be best for the Commonwealth and its future requirements will depend to significant degree on the outcome of the sale of Australian Defence Industries (ADI).
The distinctive feature of the Collins Class Submarine Project is the overall magnitude and complexity of what is involved, not any insurmountable difficulties with individual aspects of it.
The lessons advanced in this part of the report are based on the perspective that scale and complexity are factors which require special attention, and an important point from the Submarine Project is how hard the lessons are to learn in a small country like Australia with, relatively, so few large indigenous defence procurements.
In the 1980s, each head of the procurement organisation in Defence got, typically, one major indigenous project each, together with 3 or 4 off-the-shelf procurements, most often from the USA. Taken together with the high proportion and higher turnover of military staff in the procurement organisation, expertise and corporate memory, not to mention opportunities to experiment with alternative procurement strategies, are understandably low.
By way of contrast, during his 7 year tenure as Chief Executive of BHP, Mr Prescott was accountable for the procurement of around 30 major projects across several industry sectors, and during his 5 years as Chief of Defence Procurement in the UK, Dr McIntosh was responsible for three submarine projects and several surface ships, as well as major items for the other two Services, including tanks, self-propelled artillery, fighter aircraft and a new class of nuclear weapon, all indigenous or joint developments.
This contrast suggests three things. First, that the procurement organisation must be structured on the most professional possible lines and provide attractive careers for all its staff, including a somewhat smaller leavening of military staff. One of us has reported on measures to this end elsewhere. Second, that it create opportunities for members of its staff to be seconded to large commercial organisations in Australia to get a different and very businesslike perspective on large complex procurements and to large friendly overseas defence procurement organisations for a direct contrast and learning experience. And third, that the head of the organisation be brought in from outside the organisation from time to time. This happens in many of the countries with which we like to compare ourselves -- the UK started the practice by bringing in Mr (now Lord) Levine, the Americans do it regularly as part of their political system, the French do it as a natural part of the way they run their industries, and the Germans do it as a point of principle.
Without reflecting in any way on the diligence and competence of the present incumbent, we believe Australia would benefit greatly from doing so also. Further, we believe the position should be at junior Secretary level to more adequately reflect the skills and experience the job needs, on the one hand, and the authority and responsibility it should exercise on the other.
At the more particular or project level, there are many lessons from the Submarine Project, which might apply in whole or in part to future major indigenous defence projects, especially if both normal commercial pressures for performance and government policies are to be enforced.
Whilst Defence is in the special position of having its funds provided as a single line appropriation, rather than separately identified and itemised, it cannot afford to regard cost over-runs as entirely its own business. Certainly, when cost over-runs occur in Defence, there are no calls on funds outside the Defence Appropriations, but a great deal of damage can be done within the Defence portfolio by displacing or deferring other high priority activities. The community then does not get as much defence for its money as it should. One alternative to damaging deferrals is to set aside funds in the expectation of over-runs. This is an unsatisfactory approach. It implies that over-runs are predictable and runs the risk of holding back money that should have been spent on high priority tasks. The proper method is to cost properly in the first place.
It was put to us that Defence has become increasingly casual about seeking and expecting project cost increases. Whilst requests for additional funds have been made to the Defence Minister on short notice for other projects, the Submarine Project has been more austere. Rather there has been a very strict adherence to achieving the best result within the originally authorised amount and, despite all the issues raised, the project is not yet overspent.
It has been suggested to us that the Appropriations be broken down and far more detailed scrutiny be made by the Department of Finance, and perhaps Parliament, on a routine basis. Apart from requiring more bureaucrats, who would then be inclined to make further work for themselves, this is quite contrary to the Government’s own policies on accrual accounting and management by outputs and outcomes. More importantly, given the difficulties Defence has in sustaining the high level of technical and other procurement expertise itself, we think it extremely unlikely that either group could acquire sufficient expertise to add much value, although each would certainly add dramatically to the paperwork and overheads, including costs. We believe there are better alternatives and that some of our recommendations will also help.
In this case, having the AIDC act for the Commonwealth on the seller’s side has not been very effective because it knows little or nothing about submarines and could exert no effective influence on Kockums or Celsius. Indeed, we were advised that none of the Directors or the Chief Executive of the AIDC has the security clearances to be told the nature of the technical problems, although it is far from clear what they could have done about them if they had known. Had the shares been held by a Commonwealth agency that knew about shipbuilding, such as ADI, they could have influenced the technical issues, but, as a potential competitor with ASC, this would have led inevitably to conflicts of interest. It is far better not to create such circumstances in the first place.
In the case of the submarines, having selected an unfamiliar contractor for a unique project, Defence would have benefited enormously by hiring for the duration of the project some experts from the US or UK, who had recently built submarines and could have advised on practical aspects as "buyers friends".
In the final analysis the outcome will depend on the clarity with which the total mission is understood, the collective competencies of those involved, the effectiveness of organisational systems for bringing these together, and on the Commonwealth having sensible contractual freedom to make appropriate changes, while still having regard to its contractors' rights.
These lessons are not new and many have been applied effectively by Defence in other projects. They seem, however, to have been disregarded to a greater or lesser extent in the Submarine Project under its particular procurement strategy. We have tried to remedy this where we can in our recommendations.
TO THE COMMUNITY
All major projects, both civil and military, of the scale and complexity of the submarines have design defects and other problems, sometimes on a comparable scale. We note, however, that the Submarine Project has been subject to leaks and public vilification on an unprecedented scale.
We are concerned that, amongst the leaks, information of serious importance to the effective operation of the submarines may have been released. Clearly the community has a right to know how its money is being spent and about the nature of the problems, but this must be subject to security concerns. It is almost unheard of in other reputable navies for details of noise profiles to be the subject of widespread public speculation, for example, and details of operationally sensitive matters should be rigorously protected here also.
The saving grace is that, at this stage of their development, the submarines are changing monthly and whatever was leaked has probably now changed and is already obsolete. To take an obvious and important example, the noise profile has changed (for the better) and further changes are proposed.
In any event, and given the interest already generated, it would undoubtedly help to have more rather than less information readily available to the media and the public. We propose monthly media updates outlining progress in terms of milestones achieved and problems solved and a status report of problems remaining outstanding. At least this would put problems in context and limit their selective treatment by those with such interests.
While useful in the interim, we are well aware that only by taking strong action to solve problems, and by actually solving them, will public concerns be properly allayed. We hope that the measures we have proposed elsewhere in this report will help in both these respects.
8. SUMMARY AND RECOMMENDATIONS
The Collins class submarines are well designed for Australia’s special requirements and have generally been soundly built. They are, however, bedevilled by a myriad of design deficiencies, many of which should not have occurred, and most of which are taking far too long to remedy. Together, they are seriously restricting the operational usefulness of the boats.
Apparently sensible remedies for nearly all defects have been presented to us, which gives us confidence in the ultimate performance of the boats. The main issue is to improve the managerial and contractual structures so that the deficiencies are recognised and addressed much more quickly and robustly.
The key deficiencies are:
On the contractual side, we recommend:
On the management side, we recommend:
In our view this is the minimum change scenario. The augmentation program Project Sea 1446 can be completed at less cost than further extending the life of HMAS OTAMA. The other recommendations we have made in respect of the submarines can be expected to be completed within what would have been a reasonable contingency for such a project in the first place.
In the longer term, there are related issues, in respect of which we recommend:
These are our principal recommendations. As well, throughout the report, we have provided advice and guidance, which we believe would be useful to those charged with implementing such aspects as the Minister and the Government may accept.
THE COLLINS CLASS SUBMARINE PROJECT
AS AT APRIL 1999
Of the 6 Collins class submarines ordered from the Australian Submarine Corporation under a contract signed on 3 June 1987:
Each submarine is approximately 3300 tonnes dived weight, 78 meters long and 8 meters in diameter at the fin.
They constitute an entirely new class of submarine, being substantially larger and longer than the rest of the world’s diesel electric submarines, and quite different in their operational profiles to the nuclear powered ‘attack’ submarines.
The Collins class is required to carry out the following missions:
VISITS AND MEETINGS