More than 21km of sewer interceptor tunnelling and associated works are planned for completion before the Olympic Games, to be held in Sydney in September 2000. With civil construction programmed at less than 25 months, this will be a serious test for the 'alliance' concept. Shani Wallis reports from Sydney.
Four hard rock TBMs, five roadheaders, several raiseborers, and associated equipment will be used under the 'alliance' method of contracting in Sydney to meet a strict completion and cost schedule for the Northside Storage Tunnel project. The target date is August 2000, and the estimated cost at May 1998 prices is US$250 million. The project comprises 21.3km of 3.8m- to 6.5m-diameter TBM-driven tunnel through Hawkesbury Sandstone, another 2.2km of access declines, nine vortex and ventilation shafts 1.8m to 4.5m in diameter and between 55m and 160m deep, and all the M&E equipment for both construction and operation of the facility (Figs 1 and 2).
The Northside Storage Tunnel is a major component of Sydney Water's 20-year WaterPlan 21 strategy to clean up the harbours, bays, rivers, and beaches within its jurisdiction and represents 12% of a US$633 million Waterways Package by the New South Wales Government to establish targets and initiatives for sustainable wastewater management in the state. As an early action measure, the tunnel was given priority as part of the state government's preparations towards a successful Olympic Games in 2000.
There are up to 20 sewer overflows into Sydney Harbour/year at the four major overflow points intercepted by the Northside Storage Tunnel. With a storage capacity of 500,000m3 and a pumping capacity of 350 million litres/day, the tunnel is designed to intercept these overflow points to reduce the number of overflow events by 80% to 90% - about two a year. The surcharge will be retained in the storage tunnel until it can be processed at the underground treatment plant in North Head and the treated effluent is discharged via the plant's long sea-tunnelled ocean outfall.
An innovative, fast-track approach to contracting, far removed from traditional contracting, was required. "We needed a concept that would get the principal players in the contract engaged all together and at the very beginning of the project," explained Allan Henderson, Deputy Project Director for client Sydney Water and member of the Northside Storage Tunnel project alliance leadership team (management committee).
"We wanted to establish a single entity through which the planning, design, accounting, administration, construction, commissioning, and closing out of the contact would be managed as one, thus ruling out double and triple administration and accounting costs. The result would be a united effort for project realisation. We have found this in the alliance concept and, while many of us involved are having to abandon management practices we grew up with, early experience shows very positive advantages."
Alliance agreements are not new. The concept was developed in the oil industry and has been adopted for large civil engineering projects. It seems, however, to be a first for a major tunnelling project. Perhaps alliancing will improve the reputation tunnelling has for delays, cost over-runs, and disputes. Although much of it is exaggerated by the popular media, the industry needs all the help it can get to raise its public profile. Compared with other forms of collaborative contracting, such as design+construct, BOOT schemes, or partnering, the alliance concept is fundamentally different in that it really is a single administrative unit. As Henderson explained, the concept is based on:
All direct costs are paid for as they are incurred by Sydney Water from its existing US$258.1 million/annum capital works budget. These costs include purchase of all necessary equipment, including the TBMs. Any payments made on behalf of the alliance by the participants are reimbursed and the equipment becomes the property of Sydney Water.
Meanwhile, margins and corporate overheads are established up-front by the alliance participants. These are based on percentages of the agreed 'business as usual' direct cost estimate of the project. Payments of margins and overheads to the alliance partners (the consulting engineers and contractors) are then made in pre-programmed milestone payments and are adjusted according to reward or penalty criteria applied to the performance according to the following five project objectives and targets:
While the formulae for evaluating reward and penalty in association with cost and time were readily developed, arriving at criteria for evaluating the additional non-cost based objectives, particularly concerning environmental and community performance, has been challenging (Fig 3). "We believe it is the first time that these less tangible aspects of contracting have been incorporated into the risk reward and penalty programmes," said Henderson.
At the end of the project, any financial profit or loss will be shared by all alliance participants. This - as managed through a series of alliance workshops and support teams - is the principal incentive to keep the alliance alive through the course of the project.
Alliancing looks fine in principle and, initially, all appears to be progressing well in Sydney. However, like other methods of risk sharing and partnering, it can be belittled by comments like "all forms of contract are workable if things are going well." In discussions with TunnelTalk in Sydney, Henderson voiced his support: "Alliancing does seem to us to be the best possible concept for getting all the parties to pull in the same direction and to share equally the benefits or penalties of overall project management. Putting the principles into action has required all of us to re-examine our attitudes and preconceptions."
One of the first appointments once the alliancing concept was adopted was an expert alliance facilitator to support the development of the approach among the partners. JMJ Associates got the job. Legal firm Mallesons Stephen Jaques was engaged to assist in the establishment of the alliance and, in particular, development of the alliance agreement.
As a concept based primarily on people and their commitment to assigned objectives, the alliance is built on a management frame headed by the Integrated Project Team (IPT), with the Alliance Implementation Team (AIT), comprising department managers and other members of the IPT, managing and maintaining the alliance process. A Project Alliance Leadership Team of senior executives of the alliance participants monitors the performance of the IPT and provides guidance and support to the AIT.
The aim is to create a highly motivated and integrated workforce through team member induction, breakthrough workshops, follow-up workshops, breakdown workshops (where teams explore problems and are empowered to deal with them), high performance fundamentals workshops, and transition workshops (to establish priorities for managing transitions between various phases of the project).
Recruiting organisations to enter such an alliance is a rigorous process, but it is more straight forward than might be envisaged. In this instance, the call for proposals from the industry, as well as the evaluation and recommendation of alliance partners, was handled by a committee established within Sydney Water.
Mandatory criteria for prospective alliance partners included: experience in high-speed, large-scale tunnelling, preferably in urban environments; technical experience and capability in tunnel and cavern design and construction; and an understanding and affinity for operating as an alliance partner. Shortlist criteria included: robustness of the proponent group, experience and ability to integrate environmental management into design and construction, experience with and commitment to quality assurance, and commitment to complete the project by or before the target date.
The invitation to consultants and contractors to join the alliance was international and the response, according to Henderson, was positive, with several international engineering and contracting groups submitting proposals. The successful candidate was introduced to the risk/reward model and the direct cost framework by the evaluation committee of Sydney Water. Once that was successfully concluded, approval of the recommended alliance partners was sought from the Sydney Water Corporation Board of Directors and the alliance agreement was signed.
In October 1997, eight consortia submitted proposals to form the alliance with Sydney Water. A shortlist of two was prepared, and in January 1998 an alliance agreement was signed. The successful consortium comprised Transfield, one of Australia's largest construction companies with considerable tunnelling experience; Montgomery Watson Australia, part of the international Montgomery Watson Group, with specialist experience in water and wastewater treatment; and Connell Wagner, a consultant with specialist underground excavation and tunnel design experience.
The eight member Project Alliance Leadership Team, with two members from each of the four participants, was set up. The central project office was located close to the Tunks Park construction site.
With 'notice to proceed' finally granted by the government in July 1998, work began at the main access/working sites. The plan is to advance the main tunnel using four refurbished TBMs, three open-type Wirth TBMs of 6.6m, 6.3m, and 3.8m, respectively, and a 6m-diameter double-shielded Robbins TBM. Three of these machines will work from the Tunks Park access point, two in either direction in the main tunnel and the other in the Scott Creek branch tunnel, and the fourth will work westward from North Head.
An early design decision, together with TBM excavation, was to maintain the tunnel alignment at a deep level in order to stay in good quality sandstone and avoid tunnelling through depression zones of soft, water-charged sediments at the shallower levels. Some 34 boreholes were sunk to assess ground conditions and rock mass classifications, and in-situ stress measurements and cuttability tests were performed further to clarify geological predictions. A pattern of 40m- to 80m-long probe holes will also be drilled ahead of the TBMs as an extra precaution where tunnelling passes under the harbour waterways.
With continuous conveyor muck hauling behind each TBM, the machines are expected to advance at a reliable programme average of 190m/week, working 24hr/day (two 10hr production shifts and a 4hr maintenance shift), six days/week through the familiar Hawkesbury Sandstone. Over 22km of high capacity conveyors in horizontal and vertical configurations are needed by the Northside Storage Tunnel Project.
The tunnels in the Hawkesbury Sandstone will generally be unlined. Galvanised steel and fibreglass rockbolts, wire mesh, and wet mix shotcrete (possibly steel-fibre reinforced) will provide immediate support if necessary. Where required, an in-situ concrete lining will be cast into the invert of the storage tunnel.
The first TBM operation on the project, however, was excavation of the 1,477m-long spoil conveyor tunnel at the North Head working site. Tunnel muck from the North Head TBM will be lifted by a vertical conveyor in a 160m-deep shaft and conveyed to the Little Manly Point barge loading jetty via the conveyor tunnel. The 3.8m-diameter Wirth TBM started a 4.5% downhill gradient drive from the North Head site in early October 1998, finishing in early December and achieving a best 10hr shift advance of 46.3m.
Other advance works have mobilised ahead of the main tunnel TBM assembly, including excavation of the 400m-long access decline at Tunks Park and 1,600m of spiral access declines at North Head. They are being excavated by two Voest Alpine AM105 and one Mitsui S300 roadheaders. These, plus two more Mitsui machines (an S300 and an S200), will excavate various additional drives and underground chambers.
At North Head, the upper 400m spiral decline gains access to the existing underground sewage treatment plant, which is to be modified to incorporate additional screening and pumping capacity to handle the captured overflow. A lower 700m spiral gives permanent vehicular access to the storage tunnel level, where the TBM assembly chamber and new tunnel pumping station inlet works will be built. A second 500m spiral decline gives access to the new tunnel pumping station level, where a 14m-wide x 25m-high x 20m-long pumping chamber will be built.
At Tunks Park, the roadheader will also excavate two underground bunkers about half way down the decline, with a maximum holding capacity of 9,000m3. These will hold tunnel muck generated by the TBMs during the night until it can be loaded on to mucking barges in the daytime. All three TBMs at Tunks Park will work simultaneously for about three months.
Currently, the upper 400m spiral access at North Head is complete and excavation of the lower spirals is under way; the Little Manly Point spoil conveyor tunnel is finished, as is the 400m access adit at Tunks Park; and the roadheaders are excavating the TBM assembly chambers and working areas.
Work is continuing to make sure that all four of the main tunnel TBMs are working by mid-1999. The first away, in late May 1999, will be the 3.8m-diameter Wirth machine on its second drive from Tunks Park to Lane Cove River. The last to start, in late July, will be the 6.3m-diameter Wirth machine in the Tunks Park to Scotts Creek branch tunnel, while the two on the North Head to Tunks Park drive are expected to start in mid-June. Work is planned to be finished by early 2000.