Towards a decision support system for disaster management

Abstract

IT-based Decision Support Systems (DSS) can provide appropriate countermeasures to respond to disasters. In order to handle different and exceptional situations, there is need for a generic DSS, which avoids fixed rule-based recommendations. Moreover, recommendations for countermeasures need to be coherent and account for dependencies and factors influencing the choice of measures. This paper proposes Case-based reasoning to compose the entire DSS and presents an approach to modeling sequences of countermeasures using High-level Petri Nets. The modeling includes (sub-)events, endangered objects, and countermeasures implemented providing reasonable information content and making the choice of measures comprehensible. The nets represent solutions of stored cases. The idea is to merge the nets of the retrieved cases in order to gain new sequences of countermeasures for recommendation without losing the original firing sequences of each involved Petri Net.