Major underground construction projects by nature have the potential for significant cost and schedule overruns as the conditions in which the work is undertaken are likely to be less than fully understood. Over the years, there have been projects whose performance when measured against the original budget and schedule have been less than optimal for a variety of reasons, including such projects as the Heathrow Express, Channel Tunnel, East Side Access, Hallandsas, and more. Research undertaken by Bent Flyvbjerg and others has generally concluded that a major infrastructure project that includes underground work will have an average cost overrun of approximately 40%.
Over the years there have been numerous efforts to improve the contractual models used for underground construction, including the development and implementation of geotechnical baselines, which developed out of work that started in the United States in the 1970s, the International Tunneling Association’s long-standing efforts to develop a framework for the contractual sharing of risk, the Code of Practice for Risk Management of Tunnel Works published by the British Tunnelling Society and the UCA of SME’s recently published Recommended Contract Practice’s for Underground Construction. In addition, more local efforts such as the development of underground specific contracts in some countries and the development of the NEC (“New Engineering Contract”) in the United Kingdom have also attempted to address the equitable sharing of contract risk.
Given the proliferation of articles and sessions at conferences concerning this topic, there is still a healthy debate concerning this subject. While there may be no “one size fits all” solution, there are several themes that need to be addressed to arrive at a contract suitable for a particular project. Almost every project is unique and the development of the appropriate contract environment should be addressed for a specific project as there are many factors at play, not least of which are the owner’s attitude to risk sharing as well as their experience, the specific type of underground work to be performed and the legal procurement framework that may exist in the location of the project or be attached to the funding of the project.
Risk Management Toolkit
There are several themes that need to be addressed when developing the contractual environment for a specific project. This toolkit includes the following items:
Risk assessment, allocation, and risk register
Baseline and Differing Site Condition (DSC) clause
A payment mechanism for changed ground conditions to minimize claims
DRB or other alternative resolution process
Procurement strategy
Collaborative project environment
To demonstrate how the decisions during contract development can impact project outcomes, some recent project examples will be examined.
Water Supply Project
This 160-mile-long water supply project included approximately 28 miles of tunnel that was to be constructed predominantly by drill-and-blast or mechanical means with one TBM drive tunnel included for the first bored tunnel crossing beneath the Bosphorus Channel separating Europe and Asia. The underground work was spread across three separate contracts, but all of them were handled in a similar manner as indicated below:
Risk Register – No risk register was required for the project although during the design period the designer maintained its own internal risk register. No formal risk analysis was required to be performed for the project.
Baseline – A geotechnical baseline report was not prepared or utilized. A geotechnical data report (GDR) was prepared and a re-measurable Bill of Quantities (BoQ) was used to establish a baseline.
DSC Clause – Included.
Payment Mechanism – Re-measured BoQ covered all potential ground conditions.
Procurement Strategy – Design-bid-build, lowest responsive and responsible bid.
Collaboration – No formal partnering or other strategy included.
Dispute Review Board (DRB) – Not included.
Management of Ground Conditions – Non-TBM Tunnels
On this project, the ground conditions were baselined using a predetermined bill of quantities that included payment items both for excavation and support of various classes of ground. These ground classes and the support to be installed as they were encountered were based on the use of the Q system. In addition to the six support and excavation classes identified on the contract drawings (Figure 1) and for which quantities were included in the BoQ (Figure 2), a suite of additional support elements such as rock bolts, steel arch ribs, mesh and shotcrete were included to be used with the agreement of the owner’s engineer who was on site supervising the work.
The ground class distribution was based on the Q values estimated from the cores received from the ground investigation together with surface outcrop mapping and historical data related to tunnels in the similar ground conditions. An assessment was made of the percentage distribution of the Q values along a tunnel’s alignment. No specific locations for where the different classes of ground were to be encountered were shown on the contract drawings and these percentages were converted into linear measurements for the tunnel to enable easy reconciliation of payment for the encountered ground conditions. The contract did include a clause for recalculation of rates inserted at time of bid should quantities be exceeded by 25%.
The contract specifications included specific requirements and responsibilities related to the mapping of the excavated ground and how this was linked to the payment items. The contractor was required to map the ground and provide the mapping to the engineer, who reviewed and ultimately approved the Q value derived from the mapping. The mapping also included recommendations for additional support for the engineer to review and approve or otherwise to trigger payment of the additional support items.
Project Performance
Despite a GBR not being used, the baseline established through the BoQ and the various support and excavation classes, the overage clause, and the fact that the contractor got paid based on the ground encountered against the rates established at time of bid, changes in the ground conditions could be dealt with relatively simply from a cost perspective. As such the risk was clearly allocated.
On this project, the schedule was dealt with in a unique way related to financing made available and as such is not addressed in this discussion. However, the contract contained provisions to enable the schedule to be adjusted should quantity overages exceed 25%.
For one of the tunnels the encountered conditions were significantly different than those predicted from the ground investigation. This was partly caused by a disconnected investigation process with the client undertaking the drilling, a local consultant undertaking the mapping and a separate consultant developing the Q value. Upon re-evaluation of the cores it was discovered that errors were made by various parties leading to unrepresentative estimates being developed. However, the contract form used enabled the changes to be processed relatively easily with the contractor. The discussion between the Engineering JV and owner was somewhat different.
On the negative side of the equation, the lack of formal risk assessment led to unrealistic expectations concerning cost certainty and meant that there was very limited contingency allocated to the tunnel projects. The absence of shared commitments and unrealistic expectations from the owner led to strained relationships diverting significant man-hours from the main task of constructing the project. Cultural factors and owner experience were not sufficiently appreciated when developing the procurement strategy.
The tunnel excavation projects all suffered cost and schedule overruns for various reasons including artificial restraints imposed due to cash flow constraints imposed by the Owner. However, the management of the construction contracts related to ground conditions was straightforward.
Wastewater Project
This project included the construction of 26 miles of tunnels beneath a major port city’s harbor connecting five primary sewage treatment plants and diverting the sewage to a new treatment plant and ultimately an ocean outfall.
Risk Register – No risk register was required for the project although during the design period the designer maintained its own internal risk register. No formal risk analysis was required to be performed for the project.
Baseline – A geotechnical baseline report was not prepared or utilized. A geotechnical data report (GDR) was prepared and a re-measurable bill of quantities was used to establish a baseline.
DSC Clause – Specifically excluded by contract language, putting risk outside of the BoQ on the contractor.
Payment Mechanism – Re-measured BoQ but only for the rock support indicated on the contract drawings.
Procurement Strategy – Design-bid-build, lowest responsive and responsible bid.
Collaboration – No formal partnering or other strategy included.
Dispute Review Board (DRB) – Not included.
It is worth repeating here the contract language related to changed ground conditions:
“The Contractor shall be deemed to have examined and inspected the Site and its surroundings and to have satisfied himself, before submitting his Tender, as regards existing roads or other means of communication with and access to the Site, the nature of the ground and sub-soil, the form and nature of the Site, the risk of injury or damage to property, the nature of materials (whether natural or otherwise) to be excavated, the nature of the work and materials necessary for the execution of the Works, the accommodation he may require and generally to have obtained his own information on all matters affecting his Tender and the execution of the Works.
“No claim by the Contractor for additional payment shall be allowed on the ground of any misunderstanding in respect of the matters referred to in sub-clause (1) of this Clause or otherwise or on the ground of any allegation or fact that incorrect or insufficient information was given to him by any person whether in the employ of the Employer or not or of the failure of the Contractor to obtain correct and sufficient information, nor shall the Contractor be relieved from any risk or obligation imposed on or undertaken by him under the Contract on any such ground or on the ground that he did not or could not foresee any matter which may in fact affect or have affected the execution of the Works.”
Management of Ground Conditions
Six separate tunnels were constructed, five using TBMs and one by drill-and-blast. Two of the five TBM drives were lined with pre-cast segments and the remaining three TBM drives as well as the drill-and-blast tunnel utilized predefined rock support systems based on the mapped Q value. A re-measured BoQ was utilized that included separate pay items for the different support classes and contained additional rock support elements to be paid upon approval from the engineer.
As for the project in Istanbul, the contractor was responsible for mapping the excavated rock and then the engineer endorsed the contractor’s mapping, Q Value assessment and need for any additional rock support elements over and above what was called for on the contract drawings for the rock class. As the tunnels were generally some 400 ft deep and beneath the sea, probe drilling was mandatory ahead of the face, as was grouting when inflow exceeded 20 L/min. Probing and grouting were also included as measurable bill items.
During mining operations, significant groundwater inflow was encountered during the advance probing operations. The ground being mined through was massive granite with some faults and joints. The groundwater encountered was generally associated either with an open joint being intersected in which flows could exceed 200 L/min and be subject to full hydrostatic pressure, or with fractured rock bordering fault gouge material in the fault zones. Although the TBMs could potentially operate through the open fissured rock, the presence of reclaimed ground above the tunnels meant that grouting had to be undertaken to traverse these zones to minimize the potential for excessive settlement in the overlying reclamations. Where water inflow was associated with faults, it was rapidly apparent that the water would need to be cut off to avoid the fault gouge material sloughing and raveling as the TBM mined through it. Considering some faults zones were over 300 ft in width, this was extremely challenging to manage. As can be seen in Table 1 significant drilling and grouting operations were necessary to complete the TBM mining. operations.
Project Performance
Despite the inclusion of measurable items, the quantities were rapidly exceeded and significant cost and schedule overruns were experienced. On one of the tunnel drives this resulted in a 483-day delay on a 774-day contract duration together with a $37 million cost overrun on a $97 million awarded contract sum. With the aforementioned ground conditions clause forming part of the contract, the management of these conditions was far from straightforward.
CSO Project
This project included the construction of 42,000 lf, 27-ft diameter hard rock TBM mined tunnels for a CSO diversion facility.
Risk Register – A risk register was developed during the design phase and utilized to manage risk during that phase but thereafter no formal risk management process was maintained and the risk register was not carried through to construction or operations.
Baseline – A GBR was used and the ground risk was allocated to the client.
DSC Clause – Included.
Payment Mechanism – Not included for DSCs.
Procurement Strategy – Design-bid-build, lowest responsive and responsible bid.
Collaboration – Informal partnering
Dispute Review Board (DRB) – Not used.
Management of Ground Conditions
Both a GDR and a GBR were provided for the project. In addition, expected ground conditions based on defined rock classes were shown on the contract drawings. The contract drawings also included three distinct support classes based on Q values. Separate payment items were included for each support type. Unfortunately, there was no specific mechanism identified in the contract documents that defined how the support class was to be determined to enable payment to be made for the different support types. Unlike the two previously discussed projects there was no requirement for a specified person to map the ground, nor was there any mechanism identified for concurrence and determining whether the support installed was appropriate or not.
Project Performance
The lack of clear link between support classes shown on the drawings, support installed and payment resulted in an ongoing dispute throughout the contract period. The contractor’s interpretation was that he should be paid for the support installed whereas the owner considered payment should be made against the ground conditions encountered and hence the Q value derived from face mapping. A resolution was reached but this led to ill feeling between the parties involved during the excavation period.
Ground conditions encountered also varied from those described in the GBR, sub-horizontal jointing led to rock falls behind the TBM cutterhead and on the trailing gear, which resulted in the contractor installing pattern rock bolting throughout the entire drive as a minimum, for which he believed he was entitled to payment. Again, the lack of any contingency items or clear payment mechanism led to disputes concerning cost and schedule. The lack of a DRB or some form of partnering meant there was no alternate form to discuss and resolve these matters.
Transit Tunnel Project
The scope of work for this project included 10,500 lf of 23-ft diameter tunnel in soft ground excavated using a slurry TBM with pre-cast segmental linings.
Risk Register – A risk register was developed during the design phase and then carried through to construction. The contractor was also required to prepare and submit a risk register during the proposal phase, which was then merged into the owner’s risk register to assist in managing project risks. Formal risk analysis was undertaken by the owner at various stages throughout the project’s life cycle to assist in contingency management for both cost and schedule.
Baseline – A GBR was not used, but a geotechnical interpretive report was developed and the ground risk was allocated to the client. A baseline was established based on the contractors submitted schedule and operational cycle analysis.
DSC Clause – Included.
Payment Mechanism – Mixture of unit rates and lump sums with an overall contract value established as a lump sum.
Procurement Strategy – Design-bid-build, request for proposal, negotiated contract value.
Collaboration – Informal partnering.
Dispute Review Board (DRB) – Included.
Management of Ground Conditions
It was originally intended to use a GBR to establish a baseline for this project. However, the selection of a slurry TBM and the lack of ability to map ground conditions or even observe the conditions except under compressed air conditions during stoppages for cutterhead maintenance led the owner to a different solution. It must be stressed that this owner has significant experience in underground construction work and had a robust and well established risk management process and so felt comfortable departing form accepted industry practice. The basis for the baseline was developed during the negotiation phase and essentially utilized the contractor’s submitted and accepted contract schedule supported with detailed TBM operation cycle times and a detailed assessment of anticipated cutterhead tool consumption. As ground conditions varied from a full face of rock, through mixed face to glacial till with boulders average advance rates, instantaneous penetration rates, availability and utilization factors etc., were developed for these different ground conditions. The schedule had to be broken down to reflect these rates for the various tunnel reaches and separate unit rates were included for payment of tunnel excavation.
A separate payment item was developed that established the basis for payment for cutterhead interventions. The risk associated with this activity was shared with 800 hours of intervention time included in the contractor’s bid price and schedule and predetermined rates negotiated for intervention hours in excess of these for both critical and non-critical path interventions. Very detailed payment items were developed that established the basis on which such additional payment would be authorized.
Finally, a predetermined liquidated delay cost for critical path delays and any awarded time extensions was established during the negotiations.
Project Performance
The chosen process meant that there was limited argument on whether differing ground conditions were encountered. The requirement to share all TBM operational data with the owner, the daily tunnel meetings to discuss upcoming risks and previous day’s performance also contributed to a collaborative working environment. The cutterhead intervention quantity was exceeded and the allowance item provisions were triggered, leading to payment for those hours deemed compliant. There were no DSC claims related to tunneling and that part of the project was completed on schedule and within the budget established. The TBM performance parameters were generally realized with the average advance rates achieved being slightly faster than predicted, although more discs and cutters were changed than forecast.
The contract provisions used on this project were tailored to fit the project needs, the client’s experience and the very experienced tunneling contractor selected through a prequalification and negotiation process. A high degree of trust was required as the assessment of whether the TBM performance was affected by the ground conditions was dependent upon the parameters being recorded. Although the contract provisions included clear risk allocation, a DSC, a DRB, a payment mechanism for DSC and a well-defined baseline, it can be considered non-representative of current industry practice, despite its success. One point to note is that the GIR had originally been developed to be adopted as the GBR and it was modified slightly to describe the assumptions used by the owner during the design phase and describe the expected conditions to be encountered without baselining any specific quantities.
Conclusions
Four specific case studies have been briefly highlighted above and there are many, many more contractual variants that could be examined for their success or otherwise. The case studies presented were picked to highlight how the inclusion or otherwise of six specific elements identified at the outset can affect project outcomes. It is not enough to consider each element in splendid isolation; they must be considered as a whole so that all parties to a construction contract – the contractor, the owner (which includes the designer and the owner’s representative during construction) etc. – clearly understand how the risk is allocated, how excavation and support types and payments will be assessed and paid for, what constitutes a differing site condition, how it will be assessed and paid for, and the mechanism for resolution of any dispute that may arise despite the best intentions of the contract drafters.
Unless an owner simply states that all risk is allocated to the contractor for everything and is prepared to accept the inevitable premium bid prices associated with such an approach, clear, unambiguous, and equitable allocation of risk needs to be established so that each party can price it. Ground risk should in most cases remain with the owner, to the extent that the conditions baselined differ during the actual construction. A GBR should always be considered as the preeminent and the preferred method of both describing the basis on which the design was prepared and establishing that risk allocation. However, it must be stressed that the GBR should not be considered as a geotechnical document; it is primarily a risk allocation document. As described above, there are other ways of establishing a baseline provided it is established and it is clear. Depending upon the choice of baseline, a DSC clause, and a payment mechanism to be used to establish the value of any DSC need to be included.
Regarding procurement strategy, while a negotiated procurement may assist in developing a relationship between the owner and the contractor, the success of the project is not necessarily linked to this, nor the procurement type. A collaborative working environment, which can be fostered using formal or informal partnering and the use of a DRB, enables issues to be raised early and resolved in a potentially less adversarial environment.
To close, a well-thought-out contract strategy that address each of these elements, while not guaranteed to avoid encountering changed ground conditions, constitute the building blocks to its successful implementation.
Comments are closed here.