Towards Next-Generation Catastrophe Modelling

Researcher: Nikola Blagojević, PhD
Faculty Advisor: Professor Jack Baker, PhD

Existing catastrophe modeling software combines hazard, exposure, and vulnerability models to assess future disaster losses. While this approach is effective for evaluating direct losses, such as repair costs and casualties, it faces challenges when assessing indirect losses, such as those caused by building and infrastructure downtime or supply chain disruptions. The primary limitation lies in treating assets as isolated elements of the built environment. However, research developments over the past decade have led to the creation of regional recovery models that account for functional interdependencies among assets. These models enable more accurate estimates of functional downtime by incorporating factors such as regional resource constraints, access to repairs, and infrastructure interdependencies.

The iRe-CoDeS framework and the accompanying pyrecodes software model the built environment as a network of interacting components across various infrastructure systems and building stocks. Interactions are captured by simulating the flow of resources during post-disaster recovery. This framework advances existing catastrophe modeling software, paving the way for next-generation tools that enhance the assessment of indirect losses. More specifically, the proposed framework and software can be used to assess time to achieve functional recovery of buildings considering the regional context, assess business interruption losses, support post-disaster regional recovery management and pre-disaster recovery planning.