The Observational Method in Tunnelling

Diane Mather, Design Manager, Inland Rail, ARTC

Alex Gomes, Chief Technical Principal – Tunnels, SMEC

Background

The design of tunnels and underground structures often involves a myriad of technical disciplines, which can range from geosciences up to architectonics and electro-mechanics, depending on the tunnel purpose. However, because tunnels are invariably built within the ground, a naturally formed material is an inherent part of the tunnel structure, geotechnical engineering and structural design stand out as disciplines ubiquitous to most types of underground works, as they are necessary for the assessment of the ground and the ground-structure interaction.

While material properties and behaviour are better understood and can be more precisely modelled in structural mechanics, the ground is a complex material and geotechnical engineering is by far not an exact science. It must deal with the intrinsic limitations associated with the engineers’ ability to explore the ground and determine its composition and properties, as well as with the constraints associated with the modelling and estimation of its behaviour during construction.

Due to these uncertainties, the tunnel designer must be cognisant that models and computation results should be understood as representing potential scenarios and outcomes, as opposed to being taken as precise predictions. In projects with complex sub-surface settings, anticipated ground conditions and response to excavation can only be confirmed once excavation is carried out. Hence, to achieve maximum economy and assurance of safety, a natural solution consists of the design of scalable engineering solutions that can be adjusted to encountered and observed conditions to meet the required performance criteria and safety levels.

The Development of the Observational Method

The suitability of the observational approach to deal with the complexity and unpredictability of ground and tunnelling engineering was identified by the early pioneers of rock and soil mechanics. Terzaghi (1943) initially suggested the “design-as-you-go” or “learn-as-you-go” method, which was further expanded and refined by Peck, who formally introduced the “observational method” as a rational strategy to deal with such uncertainties in his 1969 Rankine Lecture.

Since then, the observation method has evolved from relying purely on basic visual observations conducted on site to sophisticated procedures using modern monitoring instruments and computer-based back analysis techniques. It has become an essential element in the design and construction of underground works, especially when excavated through variable and complex geology, weak deformable rock masses and soils, and sections under very shallow or high overburden conditions. The observation method is a thorough planned methodology which is developed under a specific contractual framework to facilitate design changes during construction and provide a framework for risk management.

The CIRIA (1999) defines the observational method as follows:

“…a continuous, managed and integrated process of design, construction control, monitoring and review which enables previously-designed modifications to be incorporated during or after construction as appropriate. All these aspects have to be demonstrably robust. The objective is to achieve greater overall economy without compromising safety”

In Eurocode EC7 (EN1997-2004), the observational method is listed as one of the acceptable design methods. The use of this method can enable significant economy in particular circumstances and also be effective where estimation of geotechnical behaviour is difficult or uncertain.

The observational method requires effective and proactive response to actual encountered ground conditions and behaviour, as well as the performance of the prescribed designed ground support. This is achieved by developing (but not limited to) the following:

Employment of appropriately experienced tunnellers to undertake excavation of underground structures so  critical observation can be continually made during excavation.

• Interpretation of geological conditions including its variability, key high-risk features and potentially most adverse conditions.
• Design of ground support and estimated behaviour for representative conditions, including identification of controlling failure mechanisms.
• Objectives of the key parameters to be observed, measured and recorded during construction including high risk items (such as sensitive structures).
• Development of acceptable limits of behaviour (trigger action response plan) with pre-planned contingency response and or remedial actions, including clear responsibility of involved parties.
• Comprehensive instrumentation and monitoring plan, including a robust process of information collection, management and distribution.
• Evidence to confirm the quality of construction (support structures).

Caveats and Limitations

The observational method should not be used where there is insufficient time to fully develop and implement emergency plans or observations of actual ground support performance is difficult to obtain or is unreliable. For these cases, the CIRIA 185 recommends that the design should be based on robust assumptions with consideration of a factor of safety, in what is termed as “predefined design”. While instrumentation and monitoring are still considered for performance validation, it is not used for design adjustments, but targeted at providing confidence to stakeholders that no undesirable event or adverse impact to adjacent infrastructure occur.

Project Implementation

Before construction starts, the possible behaviours must be assessed, acceptable limits of behaviour must be established, and an instrumentation and monitoring plan must be implemented to verify that the actual behaviour of the ground, the ground support and adjacent infrastructure is within acceptable limits. During construction, monitoring should be carried out as planned and assessed at appropriate stages, and planned contingency actions put into place if the limits of behaviour are likely to be exceeded.

While the monitoring of the ground and ground support behaviour is of utmost relevance, the assessment of the impact on adjacent and overlaying structures is also an integral part of verifying compliance with the project’s contractual and stakeholder requirements. Instrumentation and monitoring allow the constructor to quantify the true impacts of construction, which in turn can assist in assessing any damage claims from third party property owners. It also provides assurance to third party infrastructure owners and other stakeholders that the impacts of construction are as estimated by the design and within acceptable limits.

During construction, ground conditions are generally assessed by a combination of face mapping and advanced exploration in the form of probe holes and use of geophysical methods. While access to the tunnel face is easy under a conventional cyclic excavation, it is not always possible when using a closed tunnel boring machine (TBM). In the latter case, data acquisition must rely more heavily on information obtained from the TBM operations and from advanced exploration ahead of the face.

Ground behaviour (or the ground response to excavation works) are typically assessed by correlating the observed ground conditions with design models, observation of natural phenomena and the interpretation of monitoring results considering potential failure mechanisms. This typically includes the measurement of displacements (and sometimes also stress and strain) of the ground, ground support and adjacent infrastructure. During construction, results of geotechnical assessment and behaviour are continuously compared with the design assumptions and established trigger levels to identify the need for adjustments and/or implementation of contingency measures.

Groundwater conditions are also a relevant aspect to be controlled and monitored, as the amount of water inflow and pressure encountered during construction may not only impact the ground and ground-structure behaviour but also lead to water-table draw-down and settlement that can impact adjacent infrastructure and the environment. During construction, ground treatment can be used at critical areas to mitigate groundwater inflow and consequential impact on the water table, especially in environmentally sensitive areas.

The proposed frequency of monitoring must consider the relative proximity and rate of construction. Monitoring frequencies should be continually reviewed so that the rate is appropriate in relation to the construction activity and the trends of the monitoring results. The requirements of Third-Party Agreements and planning approvals must also be adhered to in so far as they dictate monitoring frequency and the cessation of monitoring.

Review meetings provide an important forum where decisions on the works can be made based on the data received and the knowledge of ongoing site operations. The design of the monitoring system needs to indicate what would be an appropriate regime for routine review of data, taking account of actions which may follow from monitoring observations. This may include the specification of key outputs, such as summary graphs of key parameters which are to be produced on a regular basis.

Summary

The observation method is today an essential approach underlying the modern design, construction and risk management of tunnel and underground works, enabling cost or time savings, while maintaining an acceptable level of safety.

Whilst the application of the observational method must be scaled to the nature of the design and the perceived hazards and level of risk for each specific project, practically any tunnel project considers at least an instrumentation and monitoring plan to validate the design assumptions and observe potential impacts.

Stakeholders not familiar with the observational method may inappropriately associate it with uncomfortably low safety margins and project uncertainties, which may lead to cost and time overruns. A deeper scrutiny of the method shows however the contrary, as the observation approach includes the provision of a set of scaled solutions to the constructor, and a planned monitoring and design validation process that enable technical or contractual uncertainties to be addressed in a cost-efficient and safe basis.

As Australia currently experiences a boom in tunnel infrastructure projects and a new generation of engineers is introduced to the tunnel industry, it is vital that all stakeholders involved in tunnel engineering and tunnel projects become educated in the principles and application of the observational method, including not only technical disciplines, but also the commercial, juridical and managerial areas.

References:

1. AFTES (2005). Guidelines on Monitoring Methods for Underground Structures, WG GT19.
2. Austrian Society of Geomechanics. (2014) Handbook Geotechnical Monitoring in Conventional Tunnelling.
3. ITA Report n°009 – WG2-Research – Monitoring and control in tunnel construction (2011).
4. Monitoring Underground Construction – A best practise guide, British Tunnelling Society Institute of Civil Engineers Thomas Telford, (2011)
5. Nicholson D, Ming TC and Penny C (1999) The Observational Method in Ground Engineering: Principles and Applications. Report 185. CIRIA, London
6. Peck, R. B. (1969). Advantages and limitations of the observational method in applied soil mechanics. Geotechnique 19(2)
7. Terzaghi, K. (1943) Theoretical Soil Mechanics. Ed. John Wiley and Sons, Inc.