Disclosure: Tim Procter worked in Arup’s Melbourne office from 2008 until 2016.
Shortly after Christmas a number of media outlets reported that tier one engineering consulting firm Arup had settled a major court case related to traffic forecasting services they provided for planning Brisbane’s Airport Link tunnel tollway. The Airport Link consortium sued Arup in 2014, when traffic volumes seven months after opening were less than 30% of that predicted. Over $2.2b in damages were sought; the settlement is reportedly more than $100m. Numerous other traffic forecasters on major Australian toll road projects have also faced litigation over traffic volumes drastically lower than those predicted prior to road openings.
Studies and reviews have proposed various reasons for the large gaps between these predicted and actual traffic volumes on these projects. Suggested factors have included optimism bias by traffic forecasters, pressure by construction consortia for their traffic consultants to present best case scenarios in the consortia’s bids, and perverse incentives for traffic forecasters to increase the likelihood of projects proceeding past the feasibility stage with the goal of further engagements on the project.
Of course, some modelling assumptions considered sound might simply turn out to be wrong – however, Arup’s lead traffic forecaster agreeing with the plaintiff’s lead counsel that the Airport Link traffic model was “totally and utterly absurd”, and that “no reasonable traffic forecaster would ever prepare” such a model indicates that something more significant than incorrect assumptions were to blame.
Regardless, the presence of any one of these reasons would betray a fundamental misunderstanding of context by traffic forecasters. This misunderstanding involves the difference between risk and criticality, and how these two concepts must be addressed in projects and business.
In Australia risk is most often thought of as the simultaneous appreciation of likelihood and consequence for a particular potential event. In business contexts the ‘consequence’ of an event may be positive or negative; that is, a potential event may lead to better or worse outcomes for the venture (for example, a gain or loss on an investment).
In project contexts these potential consequences are mostly negative, as the majority of the positive events associated with the project are assumed to occur. From a client’s point of view these are the deliverables (infrastructure, content, services etc.) For a consultant such as a traffic forecaster the key positive event assumed is their fee (although they may consider the potential to make a smaller profit than expected).
Likelihoods are then attached to these potential consequences to give a consistent prioritisation framework for resource allocation, normally known as a risk matrix. However, this approach does contain a blind spot. High consequence events (e.g. client litigation for negligence) are by their nature rare. If they were common it is unlikely many consultants would be in business at all. In general, the higher the potential consequence, the lower the likelihood.
This means that potentially catastrophic events may be pushed down the priority list, as their risk (i.e. likelihood and consequence) level is low. And, although it may be very unlikely, small projects undertaken by small teams in large consulting firms may have the potential to severely impact the entire company. Traffic forecasting for proposed toll roads appears to be a case in point. As a proportion of income for a multinational engineering firm it may be minor, but from a liability perspective it is demonstrably critical, regardless of likelihood.
There are a range of options available to organisations that wish to address these critical issues. For instance, a board may decide that if they wish to tender for a project that could credibly result in litigation for more than the organisation could afford, the project will not proceed unless the potential losses are lowered. This may be achieved by, for example, forming a joint venture with another organisation to share the risk of the tender.
Identifying these critical issues, of course, relies on pre-tender reviews. These reviews must not only be done in the context of the project, but of the organisation as a whole. From a project perspective, spending more on delivering the project than will be received in fees (i.e. making a loss) would be considered critical. For the Board of a large organisation, a small number of loss-making projects each year may be considered likely, and, to an extent, tolerable. But the Board would likely consider a project with a credible chance, no matter how unlikely, of forcing the company into administration as unacceptable.
This highlights the different perspectives at the various levels of large organisations, and the importance of clear communication of each of their requirements and responsibilities. If these paradigms are not understood and considered for each project tender, more companies may find themselves in positions they did not expect.
Also published on: