Situated in one of the most seismic regions of the world is Santiago, Chile. With its large collection of tall buildings set against the backdrop of the Andes Mountains, it was a superb location for the 16th World Conference on Earthquake Engineering.
I attended the conference along with four colleagues, who presented papers on a range of topics including viscous damping modelling, buckling restrained braces, and lessons learnt on restoration projects.
In a change from the usual programme, this year’s conference featured a number of debate-style sessions, including one on the promises and pitfalls of performance-based earthquake engineering (PBEE) – a topic which propagated throughout the rest of the conference, and one I found to be particularly thought provoking.
What is performance-based earthquake engineering?
PBEE is an approach to the seismic design or assessment of structures or systems in which a pre-determined set of performance metrics are specifically targeted. For example, one might aim to design a structure so it’s resilient enough to be occupied immediately after a certain specified intensity of earthquake ground shaking. Or perhaps a time-based objective such as limiting annualised economic losses to below a particular level. This approach contrasts the more conventional code-based methods, which tend to be highly prescriptive and have only loosely-defined performance objectives.
The driver behind this change in earthquake engineering philosophy is largely due to the advent of several large earthquakes in California in the late 80s and early 90s – in particular the 1994 Northridge Earthquake. Whilst casualties were relatively low, the economic impact was massive (not dissimilar to the 2011 Canterbury Earthquake in New Zealand). This triggered the earthquake engineering community to look at how buildings could be designed to not only consider life-safety, but also a broader set of performance objectives now known as the three Ds: Deaths, Dollars and Downtime.
There are a range of advantages to PBEE put forward by its proponents. One of the most important is that it allows for the evaluation of performance metrics that are more meaningful to clients and other decision-makers. For example, evaluating the expected probability of collapse of a building during its design life gives a much more informative measure of ‘performance’ when compared to simply evaluating its %NBS (percentage of new building standard) score.
It also allows for performance objectives to be tailored to suit the needs of a particular project. An example of this could be the design of storage racks in a warehouse. They are unlikely to pose a significant risk to human life, but the economic impact of failure due to loss of stock could be extremely costly. So in this case, much more emphasis would be placed on designing the structure to minimise direct economic losses.
The key argument against PBEE, is that it simply cannot deliver what it promises. This is primarily because earthquake engineering is a discipline fraught with uncertainties.
There is uncertainty in the characteristics of future earthquakes expected at a site, uncertainty in the response of a building subjected to ground shaking, and uncertainty in the damageability of structural and non-structural components; amongst many other aspects that may need to be considered in evaluating seismic performance.
As a result, any evaluated performance metric is likely to be predicted with only a low level of statistical confidence.
Those criticising PBEE suggest that earthquake engineering should be more empirically-based and rely on our understanding of what has and has not worked in the past.
It should also be acknowledged that the cost of undertaking a performance-based design is likely to be significantly higher than a code-based design, and one must therefore also consider whether or not this increased initial cost is likely to be recouped.
The way forward
Supporters of PBEE acknowledge there are limitations to the approach; however, they also argue that it allows a better understanding of seismic performance and is the best way forward to meet modern societal demands. This requires the approach to continue to improve and evolve as new data and techniques become available.
In the slightly paraphrased words of one of the World Conference debaters – California-based Farzad Naeim: “What can we do? We can either improve or do nothing”.
Given those two options, I know what I would choose!
Editor's note: Find out more about our latest thinking on seismic resilience here.