In recent years, confidence in tunnelling has been eroded by stories in the press of the problems encountered on a few high profile projects. These have included concerns about safety, cost and programme overruns, leaving few potential clients willing to carry a share of the risk. With the high demand for improved infrastructure into the new millennium and the constraints posed by the environment and public alike, the requirements for new tunnelling projects should be set to increase – but this does not appear to be the case.
A recent US conference identified areas which need to be considered in an attempt to address the issue of risk and deal with its management to increase customer and contractor confidence alike. These are not new ideas, and include: allocation of risk between parties; adoption of ground reference (baseline) conditions; and development and implementation of new technology
Unfortunately, adoption of the philosophies behind risk allocation and management have been diluted, with clients preferring to place all the risk on the contractor, while UK practice on ground reference conditions, developed some years ago in CIRIA 79, has not been universally adopted in the UK.
The need to develop and implement new technology, whilst widely acknowledged, requires drivers like new work, encouragement, financial input and the willingness of clients to understand and accept risk. At the John Pier tunnelling project, the interpretation, allocation and management of the risks associated with construction were the main driving force for the adoption of a risk sharing arrangement. The tunnelling part of the project has been completed to the satisfaction of client, designer and contractor, being under budget and ahead of the scheduled completion date.
John Pier tunnelling project
The project was undertaken at Workington on the Cumbria coast and forms part of North West Water’s (NWW) programme to improve coastal waters to comply with the EU Urban Wastewater Treatment and Disposal Directive that prohibits discharges of untreated wastewater to sea after December 2000. Six unsatisfactory sewage outfalls serve a population of 29 000 and discharge untreated wastewater into the sea at or near to low water. The aim of the scheme was to remove the remaining four outfalls and deliver flows from Workington to a new treatment works facility.
The proposed solution incorporated a 2km long, 2.9m diameter tunnel, founded at between 11.4m and 19.1m below ground level (BGL) connecting to a 25m diameter on-line pumping station known as John Pier pumping station.
Management of risk between parties
Historical borehole information made it clear that the John Pier Tunnel would prove to be a difficult undertaking given the variable nature of the ground, combined with experience of tunnelling in similar conditions. The allocation and management of risk was a major consideration during contract development between client and contractor. Recent projects for this client had been undertaken using a lump sum priced contract. However, the view within the design team was that an appropriate allocation of risk and, in certain circumstances, a sharing of risks with the contractor on critical parts of the contract, would be crucial for the successful completion of this scheme.
Six tunnelling contractors pre-qualified. All stated that they would elect to undertake the project under a partnering arrangement with risk sharing. It was therefore recognised by all parties that, from a technical and risk management view point, the most appropriate form of contract would be a sole source negotiated form of contract. A review of the budget estimate and an analysis of the anticipated programme indicated that this option was also financially viable.
The New Engineering Contract (NEC) target contract with activity schedule was used as it allowed for flexibility in terms of the type of engineering, the level of design responsibility and pricing methods. The advantage of this type of contract strategy meant that risks could be identified, quantified and managed between client/designer and contractor. It would also assist in the management of the project in terms of developing lines of communication to facilitate problem solving and prevent overspend and overruns.
A strategy for a negotiated contract was agreed with contractor AMEC, which had extensive experience of tunnelling in the ground anticipated and also had the most appropriate equipment for the job, with one of its TBMs having been specifically modified for a similar previous project. This meant that personnel with the appropriate experience from both the contractor and design team would be selected, a crucial factor as the most onerous ground conditions were anticipated within 300m of the start of tunnelling.
A Heads of Agreement document was agreed and signed by the contractor and client in discussion with the designer. This defined the main principles of the contract and laid down the activities to be scheduled as part of the target price and conformed within the contract documentation as follows:
|
“The requirements for new tunnelling projects should be set to increase – but this does not appear to be the case” |
In order to manage costs and maintain focus on the execution of the contract, incentives were incorporated to align the contractor’s objectives with those of the client’s A payment mechanism was devised to allow for risk sharing between parties, with any cost savings resulting from the mitigation of risk to be split on an agreed basis between client and contractor. Fixed priced items, considered to be of low risk to the project, were also incorporated in the target price.
In essence, the contract was formulated in such a way as to provide a mechanism whereby the contractor’s profit was optimised within agreed target levels by delivering the project within programme and price and to a specified quality.
Route selection
The general route of the tunnel had been determined by the requirement to intercept the outfalls, and followed a north/south line roughly parallel to the coast. The route was also determined with the aim of complying with the following key criteria:
1. Availability of land for construction and maintenance
2. Avoidance of railway crossings
3. Avoidance of bad ground/minimising settlement impact
Once the land for the construction of the four shafts along the tunnel had been identified, this determined that the main tunnel would run parallel with the local railway but was not required to cross beneath it.
In terms of the potential for bad ground, the results of the initial investigations revealed a possible solution to a potentially major problem. The Workington area has a long and well documented history of coal mining activity. Evidence from previous investigations indicated that thicker drift deposits had accumulated in a buried channel associated with a previous course of the River Derwent. The trend of this channel was in a north/south direction in the same area as the proposed tunnel, confirmed during investigation.
However, even the selected route was not without problems, as mine shafts were noted to lie close to proposed shaft locations along the tunnel line. In order to mitigate the risk of encountering these structures, a geophysical survey was commissioned to define these locations and the route of the tunnel was modified. The southern section of the route was designed to avoid having to move the main trunk road into Workington to the west below the carriageway of an industrial area.
An advantage of choosing this modified route through the industrial area was that the impact of potential settlements could be more easily managed , incurring minimal disturbance. As part of the ground information included in the contract, a prediction of settlement above the tunnel was calculated using a simplified model6. It was agreed that surface settlements of 30mm was an inevitable consequence of undertaking the works using best practice (assuming a 2.5% face loss based on previous performance of the TBM proposed for the scheme) and therefore damage resulting from settlements less than this figure were the responsibility of the client.
Site investigation
An extensive site investigation was undertaken in several stages to define the general stratigraphy, resolve any apparent anomalies and explore areas of interest. It comprised more than 200 cable percussive and rotary boreholes carried out for several contracts in Workington, including the John Pier Tunnel. In the areas where glacial gravels and clay were anticipated, boreholes started at 300mm diameter with the intention that this be continued to at least tunnel horizon to quantify the presence, distribution, composition and strength of cobbles and boulders which could be encountered. The boulders recovered had rock strengths up to 350MPa.
In addition, boreholes were undertaken in the River Derwent to establish ground conditions at a potentially critical area for the tunnel drive. Cone penetration tests were targeted at an area of alluvial deposits at the location of the final tunnel reception shaft because these materials, comprising interbedded clays and silts, may have been susceptible to liquefaction during tunnelling works. The additional testing enabled the risk to be quantified and monies allowed in the contract. Pump tests established the presence of a highly permeable gravel to the north of the site adjacent to the Derwent, tending to a less permeable silty sandy to the south.
|
“The advantages of using a conditioning medium to aid better handling and removal of spoil through the machine had already been identified” |
Implementation of new technology
In the last five years, several major tunnelling projects have been undertaken in the north west of England as part of NWW’s sea change programme using the Lovat 131 EPB machine. These projects highlighted the complexity of tunnel construction in variable ground and raised a number of areas where improvements in technology would be beneficial.
Potentially excessive wear on the face, tools and screw conveyer was identified as a major problem resulting from the highly abrasive nature of the glacial deposits. In an effort to reduce the wear on these components a material called ‘Trimay’, used in the coal mining industry as a lining to the chain conveyors, was adapted to provide a coating for the tools and head of the machine, subsequently demonstrated to reduce significantly the abrasion on the machine.
Conditioning the ground
The advantages of using a conditioning medium to aid better handling and removal of spoil through the machine had been identified during previous tunnelling contracts. Foam injection was used to modify ground behaviour by decreasing permeability. It reduced cutting head torque and enhanced muck handling properties. Polymer TK50 (developed by Morrison Mud) injection was used as a muck stiffener to enable rates of progress to be maintained in open, waterlogged ground like sand, gravel and mixed face conditions.
The use of these materials in the tunnel industry are in their infancy. Key issues continue to be resolved, such as the optimum place for injection and determination of quantities. This was an area where further research and trials would prove to be advantageous for the John Pier Tunnel. Additionally, the need to develop a conditioning material capable of reducing water ingress into the tunnel face from highly permeable gravels was identified, with high flows anticipated close to the River Derwent. Considerable efforts were made to develop these materials in collaboration with the contractor. A study group was set up between client/designer/contractor to develop the new technology. The cost of the development was incorporated into the original contract.
Large boulders entering the plenum chamber in front of the bulkhead also presented a problem for the machine and the existing cutters. This problem was solved by adding grizzly bars in front of the flood gates to prevent large particles from entering and blocking the screw. In an effort to handle the boulders once in the plenum chamber, the position of the screw was modified from a central orientation to bottom right of the plenum chamber, allowing the heavier particles to be taken up the screw. This modification also had the advantage of removing clay sticking to the sides of the chamber more efficiently, which reduced the quantities of foam required to condition the clay.
Construction
Because of their previous experience, geotechnical engineers from the contractor and designer staff were assigned to the scheme during construction. They recorded the nature of the ground encountered and reviewed it against anticipated conditions, summarised in the Ground Reference Conditions. From this information it was possible, in conjunction with the contractor, to improve the anticipated rates of progress and gain a better appreciation of the quantity of conditioning material required to control the face without producing excessive spoil. Any potential ground risks could be identified before problems arose.
The 2km long tunnel was completed in 25 weeks, six weeks ahead of programme. Rates of tunnelling increased from an average of 9m/day in the first drive to 34m/day on the final drive, largely as a consequence of improved ground conditions along the drive. This is also reflected by the decrease in downtime from an average of 24% to 6.7% as the tunnel progressed. Early completion of the tunnel allowed NWW to achieve savings of 10% on the original target price.
The project was successful for a number of reasons:
During construction, the partnering form of project management stimulated exchange of information and discussion, generating continual improvement of the tunnelling process.
Related Files
General Site Location
Longitudinal Profile
Schematic of the Lovat
