Randall Essex commenced the meeting by introducing the background to the need for Geotechnical Baseline Reports (GBR) in the US. He explained it was primarily derived from the US litigation culture and inherent uncertainties in ground conditions. In the 1970s, the US construction industry came to appreciate contractually that sub-surface conditions could vary remarkably at a project site.
In a competitive bid environment when contractors are required to make their own interpretation of the geotechnical data, they are compelled to make a potentially over optimistic interpretation of the geotechnical conditions and use that interpretation as the basis for the price of their bid. The consequence was that the parties would be in dispute when the contractor encountered differing conditions from those used as the basis of their tender. Consequently, costs spiralled and no tunnel project was being completed without a significant ground conditions claim.
It was evident from litigation hearings on ground condition claims that were presented before the Army Engineers Board of the contract appeals process, that these hearings were run by lawyers for lawyers and defence testimony was presented before lawyers.
Dissatisfied with the existing process for resolving ground condition claims, the industry started to consider and develop better approaches and to improve the contracting tools so that the parties to the contract could decide the outcome of the claims and not the lawyers. The concept of improved contracting practices began in the US back in 1974 with a cornerstone document by the US National Committee on Tunnelling and Technology. It was based on practitioners coming together to see what they could learn from case histories. This identified four principle contract tools that the industry should be applying.
The problem is that Geotechnical Interpretative Reports were written by geologists. So when the geologists reported that the ground conditions could be this, or could be that, or they could be something else, then geologically that is correct as a factual statement. The difficulty was that that such statements had no basis in a contract document.
The GBRs were being written for inclusion in the contract, but they weren’t been written effectively. Thus, the necessary focus in the US became “how do we write better Geotechnical Baseline Reports”. This issue was addressed by the American Society of Engineers’ Underground Technology Research Council (UTRC).
Randall Essex explained that in 1991 he was asked to start up the committee on geotechnical reports, with a remit of providing guidelines on how to write geotechnically binding interpretations to be included in a contract. The consultation documents (The Yellow Book) were finally published in 1997 and the recommendations have now been implemented on a number of projects across North America, Canada and elsewhere.
The committee identified that the concept of a GBR was dependent upon geological processes, constructability and behaviour as a consequence of the contractor’s selected means and methods; and that it was only one component in a very important network of contract documents. If the GBR was not intimately connected with the plans, the specifications and most importantly the payment provisions then it had no benefit. The GBR is therefore the most important contract specification written. It’s the first document that a bidder reads and it is the first document that a dispute review board (DRB) reads when a dispute is brought before them.
Chapter 7 (Owners Perspective’s) may be the most important chapter in the document. Owners need to appreciate that this is not a guarantee or warranty on ground conditions. What it attempts to do is establish a framework for resolving disputes and allowing the contractor to move ahead. Chapter 8 (Roles and Responsibilities) defines the roles and responsibilities of everyone who is involved in the process.
Recommendation of the Yellow Book
The Yellow Book starts by defining geotechnical baselines as descriptions of anticipated relevant physical conditions and relevant behaviours that we expect will be encountered using one or a number of different tools for construction. Because they allocate behavioural conditions they also allocate financial responsibility to deal with those behavioural conditions. The balance of the risk sharing is determined by where the baseline is set. So the GBR is the document that says to the contractor these are the financial tools that we are going to use to pay you for the work that you will ultimately have to perform.
The benefit of having risk sharing is that the owners really only pay fairly for the encountered conditions. The amount of risk we allocate to the parties is ultimately determined by what is said in these baseline reports. So the GBR contains contractual assumptions that are not geotechnical fact. The GBR is therefore an administrative tool to administer the site conditions clause. When a contractor says he encountered a changed condition, the base line report answers the question “different from what?”.
Randall Essex explained that to add clarity and certainty to GBR’s there are certain words and phrases that have to be eliminated from baseline reports, all be it that they may be truthful in a geological context based on what we know. Phrases like: “these things could occur” or “this might happen”.
Also the author must be cautious of qualitative works and should avoid phrases like; “high groundwater”, “frequent boulders” or “occasional joints”. The GBRs must therefore contain qualitative terms that are measurable in the field and those chosen terms must use wording that is integral with the wording in the specification.
A further recommendation is that the geotechnical engineer who commences the GBR should also be the engineer who writes the specification and who ultimately completes the GBR, to ensure compatibility and to avoid inconsistencies or ambiguities. The owners also need to recognise that where these baselines are set will directly impact the price for the work so they also need to be very involved in the interpretation, understanding and in the setting of those baselines at the beginning of the job.
The guideline on GBR’s is a relatively short pamphlet, it was intended to be a short pamphlet to set a guideline for a relatively short GBR. In general, a GBR should not be any longer than about 45 pages.
Lessons learned
Randall Essex then went on to explain the lessons learned in the US, where the industry has been following the guidelines of the Yellow Book for the past six or seven years. The first positive aspect of GBRs is that they can be used to identify and fill in gaps.
Gaps in the data could result for example from the difficulty and cost effectiveness in carrying out investigations in difficult mountainous terrain. Filling in the gaps in the data with baseline conditions specified in the GBRs is a far better process than telling the contractor that the risk is yours, as these days contractors will not price that risk.
Baselines can however also be double edge swords. For example on a hard rock TBM job if you tell a contractor that he is responsible for temporary support and you talk about the fracture zones that he is going to encounter, the expectation that the ground may be well fractured may be conservative in terms of temporary support, but it may lead to over optimistic predictions of TBM progress rates. So the lesson learned is that you simply can’t be conservative and you have to be realistic or you may end up misleading the contractor on geotechnical conditions that he reasonably relied upon.
Another lesson learned is that property ranges do not work. If properties are described as varying considerably and that this variation within the data is presented as the baseline, then that is not a baseline. A contractor cannot price large variations in ground conditions and design parameters as the baseline. So in the US it has been found that if you show graphical representation of the data as a histogram, this shows the statistical characteristics of the data and it provides far more meaningful data from which to interpret behaviour than just the range.
The theory of reasonable reliance
Before there were GBRs, there was a statute in US law that stated that the contractor had to reasonably rely on a contract indication for him to be impacted by that indication. So if there is an indication in the contract, the contractor has to rely on it when pricing the data in order for him to claim for the adverse consequences of a change to that indication. If a contractor cannot demonstrate that he relied on a contract indication when he priced for the work, all be it that he can now demonstrate that a change to this condition has subsequently had an effect on him, then a claim based on reasonable reliance will not be successful.
Some recent developments
Substantially the environment in which we can see GBRs being used has been within the design, bid, build arena. So we have recognised that design and build is in fact an improvement on design, bid and build contracts because it engages proper collaboration with the design team and the contractor. We therefore feel that this collaboration has to occur in the development of the GBR as well. So there is another book that was published in 2001 by Bryer and O’Neil. This drew on lessons learned from GBRs that were implemented in a design/build environment and its principal recommendation was for a two-stage GBR preparation.
The first stage is the production of a design team GBR where the owner does not know the details of the contractor’s chosen means and methods. This document has gaps in it and the second stage requires each contractor to fill in the gaps based on their interpretation of the effects of their chosen means and methods.
This requires the contractor to identify what behaviours they deem are relevant to the success of the work and what parameters they relied upon, based on the defined physical conditions. The bidder’s response is called the Geotechnical Report for Construction (GRC). The contractor’s response to the GBR may indicate that he has judged ground conditions over optimistically.
In such instances the opportunity was made available for a back and forth negotiation where the owner was able to state that he did not agree with anticipated behaviours proposed by the contractor and it allowed him to request a quote for the effect on price of the bid if the baseline was amended to better represent the owner’s anticipated behaviours.
It is recognised that a lot more experience is required in implementing GBR’s in a design build context and it is similarly recognised that the “Yellow Book” has raised the profile of risk sharing in the post 9/11 era.
For example insurers in the US will no longer insure underground construction contracts that are worth more than US$250M. The GBR’s are therefore an absolutely essential tool to level the risk sharing playing field and if owners don’t implement a fair contracting process then insurers may not underwrite the construction.
