With the impacts of climate change becoming more devastating each year, there’s never been a more important time for government agencies, NGOs, and businesses in the UK to make the right decisions when it comes to optimising infrastructure to balance societal and environmental needs. That’s where the expertise of Interim Dean of Kent Business School, Professor Jesse O’Hanley, comes in.
As the Head of the Department of Analytics, Operations & Systems, can you tell us more your research and its applications?
My research is primarily within the field of operational research, which is applied mathematics for decision-making. My particular focus is on environmental conservation and sustainability, but I also have a broader interest in facility location and logistics.
One of the areas I’ve done a lot of work in is improving river connectivity. Rivers worldwide have many barriers ranging from big hydropower dams to culverts and weirs. These block fish from accessing different parts of the river, which is a severe problem in the UK, Europe, and North America. Using operational research approaches, I can devise algorithms which take into account multiple factors, such as how passable they are for fish, their relative position within the river network, and the cost of removal, and use these to figure out which barriers to remove to restore habitat connectivity.
How might this approach help us respond to climate change?
Flip the river connectivity problem around and you can see how operational research techniques can be used to determine where to locate infrastructure for clean energy, such as hydropower, in an eco-friendly manner. I’ve done research on this in the UK and in Brazil. There’s strong interest in the Amazon, Congo, and Mekong basins due to large amounts of hydropower being installed with little planning. Better planning could significantly reduce environmental impacts and improve capacity to generate renewable energy.
How is the emergence of AI technology affecting your area of work?
There’s a big overlap between AI and operational research. Much of what is called AI is actually machine learning, and many machine learning problems are optimization problems, which fall under operational research. Computer science may use different terminology but we’re often using the same approaches, which are fundamentally about making things better. The development of AI will see major advances in data acquisition and data assimilation. Reinfocement learning, a type of AI that can learn from data to plot optimal decision-making trajectories, could also prove transformative for optimizing systems and decision-making.
What have you learned from working with industry partners?
I have worked with a range of organisations and businesses, including with Eurostar and the Port of Dover, to help them apply operational research techniques to solve their problems. Rather than forcing a technique onto a business problem, I find the best approach is to really get to know a business and their problems first. I ask questions to encourage businesses to articulate their objectives, how they measure success, and what control points they have. It’s about having a conversation to define the problem, and then finding the right operational research technique for it. It’s also key to speak their language and make the results practical and usable. Often, it’s the modelling process that is more beneficial to the business than the output itself, as they learn so much about the system along the way. I’m also aware that when working with companies, going for super advanced, high-tech solutions isn’t always the best idea; you need to balance technicality with usability. If a solution isn’t usable, it’s a waste of time for everyone.
Having been part of Kent Business School for almost 20 years, what would you say makes it stand out?
I love the School and particularly love my department. Our community of researchers and academics has really built up a rapport over the years and we’re one of the best operational research, management science and business analytics groups in the country. I also like Kent and Canterbury, and I like our students. I think it’s a nice place to work in that sense.
What advice would you give to those aspiring to work across disciplines as you have?
Before discovering operational research, I intended to become a medical doctor. It was only when I took a class on decision making whilst studying Biological Sciences that I realised what you could do with mathematics and completely changed course. What’s great about expertise in operational research is that it’s not a one trick pony; it can be applied to better understand problems and improve decision making in any organisation. So despite my change of career path, I was still able to find a way to use it to pursue my interests in the environment.
For those aspiring to pursue a career in operational research, my advice is to get good at statistics and be aware of all the different techniques in operational research, but also focus on the areas you’re good at. It’s also important to be flexible and adaptable and stay on top of emerging trends, especially when it comes to AI.