Taylor Burns: Risk management for major infrastructure projects

Taylor Burns is the Australian representative for the Austrian based company, RiskConsult. He’ll share his thoughts about risk management for major infrastructure projects at a REBOK lunchtime webinar on Tuesday 20 October 2020. Register here.

Why is a ‘big picture’ view important for infrastructure projects?

If you don’t know what the big picture is, you’re missing all the uncertainty. There’s an image by Nassim Nicholas Taleb which shows somebody walking across a river that’s just over one metre (four feet) deep on average. But averages vary, so it’s potentially really deep at one end, and really shallow at the other. If you were to walk across assuming it was the same depth the whole way, you could potentially fall in and drown. The same thinking can be applied to infrastructure projects. 

Image courtesy of Taylor Burns, RiskConsult

Why is it important for engineers and project managers to take cost and schedule uncertainties into account?

Let’s start with costs. Any quantities you might have on a project include uncertainties. For instance, with excavation, you might have less or more unsuitable materials or rock than determined by the geotechnical investigations. Before the award of a contract, costs for materials and other payment items may also vary significantly from the initial estimate at the time of tender. Risk impacts incurred throughout the project will also result in additional costs and time-related costs. These risk impacts also hold uncertainty. 

With probabilistic assessment of cost and schedules, we are able to use distributions, one of the most common being a triangular distribution. In this case you have the most likely (often the deterministic estimate) in the middle, and the best and worst scenarios either side of it. The same goes for time durations. This can be done for all items in the work breakdown structure, including the risk impacts, which bring another set of time and cost uncertainties. It’s really important to do this assessment to get an understanding of how good or bad things could be. Once the distributions are set we can do a simulation – either Monte Carlo or Latin hypercube sampling – to work out the probabilities of certain outcomes so that you can have the full picture.

When you’re doing estimates deterministically, you only get a very narrow view of what the outcomes might be, and you miss a whole lot of information about what might happen. Things don’t always work out the way you plan them – so it’s important to have the full picture.

How can a good picture of cost uncertainties help with infrastructure project management?

Having a good picture of uncertainty is especially helpful for decision makers. Knowing the likelihood of cost overruns or project delays in advance provides opportunities to take action early and mitigate risks to acceptable levels. 

This is especially helpful for large infrastructure projects that may stretch over a number of years. To further clarify the picture, it is useful to use a cost component structure. A simple example is to separate costs into base costs, risk costs (known and unknown) and escalation. Escalation is particularly important for projects spanning multiple years. These cost components are often depicted in the form of a waterfall diagram with results shown as S-Curves.

Image courtesy of Taylor Burns, RiskConsult

How would you define a complex infrastructure project?

Generally, complexity comes with the size of the project, for example, projects above $50 million. Complex infrastructure projects have many components, stakeholders and elements which need to integrate with one another, and with the potential to mutually affect each other. They are surrounded by uncertainty and ambiguity, similar to the unpredictability of many moving parts.

One example of a complex project is a bridge replacement. You may have to provide a temporary bridge, demolish the old bridge, realign the road and have multiple traffic interchanges. Another example is tunnelling. You may have a lot of unknown geotechnical constraints such as especially soft or hard material, water inflow, or heat due to geothermal activity. You could also have fault lines within the material that you're tunnelling through. You’ll need to assess the impact on the tunnel boring machine when travelling through all of these different substrates. This could include the different classes of protection needed and the time taken to bore through different substrates. In addition to the technical complexities, large infrastructure projects often require a complex or alternative contracting method.

Why should you avoid a blanket contingency in large or complex infrastructure projects?

If you were doing a cost estimate deterministically for a low value project with little risk, you might use a blanket contingency of 10 to 20 per cent based on what you perceive the risk to be for your project.

Very large or complex infrastructure projects generally have a lot of unknowns and uncertainty. I feel it’s better to assess those unknowns along with the identified risks and integrate them within your work breakdown structure and your schedule. Then you can carry out a Monte Carlo simulation to get the full picture of the potential impacts of those risks, uncertainties and escalation over time. These probabilistic results can then be used to inform what a suitable contingency might be.

Who do you think would benefit from your webinar?

Project managers and decision makers for large infrastructure projects, engineers, consultants, government, and people who take on high-value, complex projects.