Ian heads up the risk modelling team at Arthur J. Gallagher International. Having completed a computer science degree at Sussex University, he worked in the IT industry as a software engineer before deciding on a career change to catastrophe modelling.
Ian Scott, Senior Risk Modeller, Arthur J. Gallagher
After completing his master’s degree in GIS with an internship with the catastrophe modelling company AIR Worldwide, he started work at Willis in their proprietary catastrophe modelling team. Three years later, the Analytics team at AJG approached him to set up a risk modelling team. They’ve been running vendor and their own proprietary models plus producing risk and exposure maps.
How long have you been using Ordnance Survey data?
We used OS data a lot during my master’s degree and then at Willis we had various OS datasets although as the majority of my cat model development work was focused outside the UK, the OS data was primarily used in exposure mapping. It’s been at AJG that I’ve used OS data the most. We’ve integrated OS data into our risk reporting procedures and analysis tools.
What do you find most useful?
I use AddressBase Plus the most. It has given us the location and building type for every structure in Great Britain. Being able to classify risk for more than 30 million buildings at an address level has greatly increased the accuracy of our risk reports.
Why did you become interested in geographic information?
Near the end of my career as a software engineer, while I was researching different opportunities, my older sister gave me a book: Divine Wind by Kerry Emanuel. It details the history and science around hurricanes and the computer modelling of them. It also highlighted the importance of geographic information in determining cat risk. I found it fascinating and it prompted me to learn more about catastrophe modelling which relies heavily on accurate and up-to-date geographic information.
What, in your opinion, are the key issues facing the financial services industry over the next three years?
The rapid increase in the quantity of data in all areas has led to massive data management challenges. Big data covers a range of ideas and techniques to manage and make use of this information. Using this data and keeping it up-to-date and relevant is a substantial task that has the potential to greatly increase the quality of financial services on offer. We’ve been investigating a number of options for big data analysis and the Analytics team is working closely with our data management team to implement a big data solution in order to offer the services throughout our company.
How can geographic information help?
Geolocated information increases the usefulness of reported data. A status update about a flooded house combined with latitude/longitude can help real-time reporting of flood extents quickly. Geolocations help to validate the accuracy and completeness of information.
Big data gathers information from many disparate sources and attempts to bring them together with common themes. Geolocated data greatly simplifies this process and can link sources of seemingly unrelated information around a common location.
Tell us something not many people know about you?
Back during my undergraduate degree, I had a six-week summer job at a large insurance broker digitising soil maps of the UK. All it involved was using a mouse to draw polygons around soil type boundaries. It was perhaps the dullest work I’ve ever done (including packing shelves). It was so dull; it put me off geography and the insurance industry for more than eight years! Learning that the process can now be done automatically, I knew it was safe to try the industry again.