A Social-Based Decision Support System for Flood Damage Risk Reduction in European Smart Cities

Abstract

Today, Decision Support Systems (DSS), which include monitoring, prediction, and control sections, are assumed to be tools for smart, sustainable management of disasters such as floods. On the platforms, first, a process is designed for receiving valid data before, during, and after a flood. Then, with the application of artificial intelligence (AI) models, the essential features can be predicted. Meanwhile, the main predicted factors should be determined according to the goals of each research project. In the present study, two-stage machine learning models will be used, including damage values in cities and rural regions and social impacts. In the first step, damages will be estimated by machine learning computations based on rainfall (mm), hourly flow of the river (m3/s), type of vegetation, density, etc. In a parallel way, after the determination of structural equation modeling (SEM) of social parameters in flood and their weights, in the second step of AI modeling, the social feedback factor will be forecasted based on effective achieved features. Finally, with the application of controlling systems such as the Decision Tree (DT) model, a fast reaction system is designed.