A smart sustainable system for flood damage management with the application of artificial intelligence and multi-criteria decision-making computations. (January 2023)

• Record Type: Journal Article
• Authors: Zabihi, Omid; Siamaki, Maryam; Gheibi, Mohammad; Akrami, Mehran; Hajiaghaei-Keshteli, Mostafa
• Is Part Of: International journal of disaster risk reduction. Volume 84(2023)
• Publication Date: 2023-01
• Subjects: Sustainability -- Smart cities -- Flood -- Climatology -- Decision support system; Emergency management -- Periodicals; Risk management -- Periodicals; Disaster relief -- Periodicals; Hazard mitigation -- Periodicals; 363.34
• DOI: 10.1016/j.ijdrr.2022.103470
• Languages: English
• ISSNs: 2212-4209

Abstract

Abstract: Decision Support System (DSS) is an approach for smart management of different man-made and natural phenomena such as flood disasters. In the present study, different stages of a novel DSS system are designed to achieve Sustainable Development Goals through monitoring, predicting and controlling the flood. First, the monitoring step is done based on scrutinizing hydrological gathered data through historical climatology archives. Then for the prediction stage, after clustering the precipitation and flood records in five provinces in Iran, with the application of Logistic Regression, Neural Network, and Support Vector Machine as Machine Learning (ML) computations, flood disaster is estimated based on rainfall in different climates. Consequently, according to three differen scenarios, the appropriate strategies in Pre-Flood Activities (Pre-FA), During Flood Activities (DFA), and Post-Flood Activities (Post-FA) are prioritized by three Multi-Criteria Decision-Making (MCDM) techniques. The outcomes of integrated Artificial Intelligence (AI) and Ward computations illustrate that the rainfall data of different provinces of Iran are heterogeneous because of various geographical and topographical conditions. Finally, the MCDM outputs demonstrate that through Pre-FA, DFA, and v, all strategies are ranked for implementation of early making-decision in flood disasters.