A hybrid model for learning from failures: the Hurricane Katrina disaster

There is a need to facilitate learning from failures in the context of natural and man-made disasters. This paper investigates the multi-faceted nature of research in disasters and the aspect of hybrid approaches in modelling within this domain. The paper applies a framework of reliability and multiple criteria decision analysis techniques to the case of the Hurricane Katrina disaster of 2005. It is shown how this hybrid model can be used through an integrative approach to perform a systematic analysis that can lead to learning from failures.

The proposed framework incorporates and integrates Fault Tree Analysis (FTA), Reliability Block Diagram (RBD) analysis and the Risk Priority Number (RPN) concept, together with the Analytic Hierarchy Process (AHP) which is used as a simulation model for decision support. It is shown how the proposed integrated framework can contribute to our understanding of failures and enhances the ability to extract lessons from failures or disasters. Such lessons are then mapped into specific decisions for prevention, and resource allocations, to help avoid a repeat disaster.