A Conversation with Paolo Bocchini

Bocchini discusses computational research, probabilistic analysis, and how his models are being used to improve infrastructure systems, predict regional Ebola outbreaks, detect blockages in oil pipelines, improve the reliability of weapons systems, and more

Paolo Bocchini Research

You've got a few major research projects going on, but let's start with the latest funding from NSF. Can you tell me a little bit about that?

This is a big project. The total funding is $2.2 million, which is a very large sum for works in the NSF engineering division that are not experimental. It officially started on September 1 and, eventually, there will be at least 16 people working on it, including faculty, staff, post-docs, and students. We are spread over three universities, seven departments, and multiple research centers/networks, but Lehigh's CEE and ATLSS are leading the entire effort.

We are studying the interdependencies of infrastructure systems after extreme events. Engineers civil engineers in particular have been working for a very long time on making our structures and infrastructures able to withstand extreme events. Indeed, we need to observe that we have become very good at this, but there are always unexpected events that cause failures. Most importantly, we need to focus on ensuring that the functionality of the infrastructure is restored quickly after the occurrence of the extreme event, to support the community recovery.

So what's one real world example?

There are many examples, but what happened at Fukushima in Japan after the earthquake and tsunami epitomizes the importance of the post-event infrastructure functionality. The power plant was able to withstand a huge earthquake and a tsunami. Neither of those extreme events caused the failure. Instead, the root of the issue is that electricity could not be restored for the nuclear reactor's cooling systems. This caused the explosion and the nuclear contamination that followed. It was not the extreme events, but the inability to respond and restore infrastructure functionality promptly. So observing several cases like this, many researchers in the civil engineering community started focusing on what happens after the event. In its essence, the idea of resilience brings into the equation also the emergency response and recovery phases. For instance, it is good to have structures that can be repaired very quickly and economically in case of failures. Similarly, it is important to retrofit bridges or powerlines that are most crucial after the event, even if they are not the largest or most in need of maintenance.

When exactly did resilience become a major area of study within civil and structural engineering?

There was a seminal paper published in 2002 that was a milestone for civil engineers. It was written by a large group of professors at different universities and simply suggested that we should start looking into this topic with a quantitative approach. However, not many people followed that advice initially: when we started working on this at Lehigh, there were only a few other groups. Today, everyone is working on this topic. So, just within the last five years it really took off from a niche issue to a mainstream one.

A few of your current projects have an interdisciplinary component to them, correct?

Yes, I am very interested in hazard and multi-hazard modeling on a regional scale. The models that the scientific community had developed for a single site are not appropriate when looking at an entire region and trying to determine, for instance, if an earthquake or the combined impacts of a hurricane and a storm surge could knock out multiple infrastructure components in a region simultaneously. With a colleague in bioengineering we are investigating how to apply similar regional hazard models to the spread of diseases, specifically Ebola.

There is also a collaborative project with colleagues in mechanical engineering, which may officially start in 2016. We have an international collaboration with the Petroleum Institute in Abu Dhabi, UAE, and our goal is to enhance techniques to detect and remove blockages in various types of pipeline networks. Some of them are very large, from the Middle East to Northern Europe, so their analysis presents computational issues, compounded by poor and diverse data. We work on efficient ways to interpret the data, to enable effective localized interventions.

And you're working with Dan Frangopol on a project for by the U.S. Army?

Yes, we're working with weapons systems, trying to apply some of our expertise in bridge life-cycle analysis to this other field. It was a fascinating challenge, because initially we were not sure if this extension was feasible. On paper, it sounded like it should, but phase one of the project was, both for the Army and for us, only a feasibility study, a way to see if we were really interested in and able to explore this topic.

Thankfully, the results have been very encouraging and we have just entered the third phase of the project. We were able to completely support a couple students on the funding from the project. The goal was to use our models to better invest some portions the available defense budget. Large scale tests were done on various weapons systems, and we input that data into our models to help predict the weapons systems' life-cycles, set optimal maintenance schedules, and analyze their reliability over time.

How have you seen computational research change over the years, and what's Lehigh's role in that?

As you can see, I am working on very different applications, but the common theme is that they involve computational modeling and simulation. This is the lowest common denominator of what we do. These are the two core pieces of expertise that my research group has and that we try to leverage for different types of applications.

I really think that we are at a turning point for civil engineering and structural engineering. We are proud of the fact that we have the Fritz Lab, which is an ASCE historical landmark and still a fully operational lab, but on the other hand, which other discipline in engineering can say that major lab instrumentation that was built 50 years ago is still fully useful?

There is something beautiful in this, but also something concerning. We need to evolve to remain relevant, and Lehigh has embraced this. We will always need to break beams and columns (and we have constantly upgraded our experimental facilities over the years, especially at ATLSS), but if we do not want to be marginalized, we need to redefine what research in civil engineering means, and I am very glad that the department has decided to do this. The fact that I and several other colleagues are here means that the entire department sees the role computational modelling and simulation increasing in the future. The research on disaster resilience that I am doing now would not have been considered research in structural engineering 15 years ago, but I firmly believe that it is an important part of the future. It will help us remain a relevant national player for the years to come.

By John Gilpatrick



You Tube