The tsunami is the second greatest threat that can occur, having a considerable impact on coastal areas and causing flooding. Therefore artificial intelligence is a major player in disaster risk management as it enhances the ability to keep people and property safe post calamity and is said to be the future of disaster management.

Disaster management is a strategic— multilayered— approach that comprises prevention, preparation, response and recovery meant to protect the vulnerable members of the population and the infrastructure at risk from any disaster.

A tsunami ranks second among the greatest dangers as it would heavily affect coastal areas with inundation included. Thus artificial intelligence plays a major part in disaster management, for it enhances the ability to save human and material resources from disaster and is believed to be the future of managing disasters.

Figure 1. Global flood losses, 1990 to 2017 [1].

Today, artificial intelligence and geospatial technology are both highly evolved, and hence have the potential to be successful in disaster situations. Disaster mitigation planning primarily depends on the shape of an area (topology) with weather and ecology, among other components— including resources available within reach from machinery location. In enhancing emergency relief safety by operations research management science resource distribution effect on population should be taken into account.

However, different studies present varied ways artificial intelligence can be effective in disaster management; a stark difference observed is that the response of other nations to crises differs significantly from that of India's situation thus there is need for identifying information meant for devastating natural calamities.

The appropriate methods aimed at reducing impact of tragedy include vulnerability as well as readiness. In this case it’s also important to take note of resilience based on disaster management [2-4].

Artificial intelligence and geographic information systems go hand in hand when it comes to understanding floods and their dissemination pathways. A geographic information system acts as a trigger— which brings together, stores, integrates, manages and offers spatial data for the purpose of coming up with a plan that will help prevent floods soonest possible — such information is both relevant and timely; these systems are efficient and holistic in dealing with flood related issues. Having decisions that are accurate and effective means taking a systematic approach from the planning stage. This work outlines AI benefits in disaster management: helping reduce disaster impact.

A keyword list is compiled from various journals in the Scopus, Web of Science and Science Direct databases which are indexed to evaluate the significance of AI in disaster management. Moreover, a number of international probes were picked for the review.

An inventory of key terms linked with different journals in the Scopus, Web of Science, and Science Direct databases that are indexed was compiled for evaluating AI significance in disaster management. Moreover, numerous foreign inquiries were chosen for review. (Table 1 and Figure 2).

Table 1. Search string (keyword analysis in international journals, 2015–2020).

Figure 2. Keywords assessed within different international journals (2015–2020).

In the next graph, we have the key words and methodology identified from the quest carried out between 2015-2020. Among the publications, Natural Hazards was identified as having the highest number of articles followed by Sustainability, Disasters and Journal of Natural Disasters.

The study was conducted via a scrutiny activity to bring out the significance of AI and GIS in disaster control. The Scopus database was used during the evaluation that contained adequate information related to the field; as a result, scientific articles were retrieved from Scopus, Science Direct and Web of Science databases about AI to have the latest and most accurate details on AI and its uses.

The work was done in the middle part of May 2021 using the advanced search function as illustrated in Figure 3. The initial search produced 2460 publications; with 1188 being common to the three databases, they were therefore excluded after an overview of titles and descriptions which led to the removal of 1089 more studies. The remaining articles were re-evaluated— and 93 more studies removed based on their nature and scope— until we arrived at a final selection of 100 studies that met all requirements for this discussion. Additionally, critical demographic information from these articles is included with record and country data stored within journal database.

Figure 3. The intricate procedure of screening the latest articles regarding AI and disaster mitigation.

The array of scientific journals is lengthened and diversified, known to be peer-reviewed where different artificial intelligence and disaster management approaches are deliberated. This is illustrated in Figures 2, 4, and 5. Many studies were published between the years 2015-2020 from China, U.S., South Korea, Iran, Australia, and Italy.

Figure 4. Distribution of the studies published by country (2015–2020).

Figure 5. Identified studies on artificial intelligence and disaster management

The disaster management cycle (DMC) is the procedure of preparing for disasters and reducing their effects ( table 2 ). The DMC also assists government, agencies, civil society, and businesses in executing and creating community resilience.

Table 2. Disaster management phases (previously, during, and after a disaster)

The process has four distinct steps in disaster management which are later partitioned into three separate efforts: Pre-disaster activities, current, and postdisaster activities.

The DMC arranges its activities as a continuum with the least possibility of human or physical damage. Disaster prevention planning before an occurrence implies two main components: enacting laws to prohibit any disaster and rules to avoid any catastrophe, which should be strictly adhered to ensure no emergency takes place. It also states that preparation must be extensive as well as resource-appropriate — preparation is highly valued by DMC because it helps in minimizing loss of property in a disaster. In disasters, response and rescue are carried out with an objective of providing first aid plus humanitarian assistance and making an immediate assessment on the extent of damage. Post disaster recovery efforts begin including launching overall community assistance program; aimed at reinstating community life support system back to normal working condition and ensuring those affected receive compensation based on identified losses.

The application of AI is aimed at increasing the efficiency of disaster management. Sensors based on artificial intelligence can help establish a common ontology-driven communication mechanism that enables information to be shared between different components of the disaster and across a multi-agent platform in real-time as well as virtual scenarios. The success of emergency management relies on the availability of multiple information sources, their integration and analysis for decision making. The purpose of AI is to strengthen the efficacy of disaster management. Sensor-based AI components can boost information sharing by using ontologies that result in informative disaster components and a multi-agent based platform— usable in both real-time and virtual scenarios. The success of emergency management relies on the availability of multiple sources of data which are integrated into one place for decision makers to come up with intelligent decisions.

Disaster management is one of the applications of AI. Sensors are considered an AI-based tool that can help in enhancing the information communication through the development of ontologies to ensure that information reaches all components of the disaster and having a multi-agent platform that can be used in both real-time and virtual scenarios. The stake of emergency management effectiveness lies in having different information sources, combining them and making wise decisions.

Artificial intelligence is aimed at disaster management to boost and better its efficiency. Sensor technology based on AI can enhance the communication of information by establishing ontologies between them which will lead to sharing information among the components of disaster, and an multi-agent platform usable in both real time and virtual scenarios— where cognitive agents are used to facilitate the effectiveness of emergency management with intelligent decision [7] that can be made after having information from various sources.

Artificial Intelligence is commonly applied to better disaster management by ensuring high efficiency. For example, the use of AI-based equipment like sensors can help establish common ontologies among different components of a disaster system and thus enable seamless information integration between the components of the disasters as well as use in multi-agent real-time or virtual platforms. The success of emergency management depends on the availability of multiple sources of information that can be fused to generate an intelligence decision [7].

The WorldCat database publications from 1991-2018 concerning AI applications in disaster management are shown in Figure 7. Among the different phases of disaster management, disaster response is noted to have the highest amount of documented research contributions. The primary application area of AI has been noted to be in disaster response— more than any other phase. It can therefore be concluded that while the use of AI for disaster response is increasing and effective (hence enhancing the overall efficiency), the popularity of using AI for mitigation is clearly depicted by figure review which refers to written works such as books and articles that are easily accessible resources.

Social networking sites and digital technology are beneficial to the public at large as they provide an effective means of communication and information consumption. During any natural or man-made disaster, thousands of individuals resort to social media for information dissemination. Similarly, the use of social media has proven effective in various charity functions as documented by previous studies. However, comprehending different dimensions of social media during a time-critical crisis is arduous task: with vast amount of data and high velocity social media streams manual analysis is complex, hence, cascading the messages becomes impracticable due to information overload plus stream pace.

In the present climate, individuals have increased their involvement in various social media platforms and other sites to take part in information sharing about different types of disasters. Earlier instances of natural and man-made disasters like floods have shown that social media plays a very important role in the information system for both natural and societal calamities such as earthquakes, wildfires, nuclear crises or civil strife. For instance, one hundred seventy seven million tweets were written in just one day post 2011 Japan earthquake.

The graph shows World Cat publications from 1991 to 2018 on AI applications in disaster management. Among the phases of disaster management, response has the highest number of documented researches works where AI has predominantly been applied during response compared to other phases. The use of AI is steadily increasing during the response phase; it is effective and efficient in enhancing overall disaster management. The growing popularity of mitigation observed in the figure underscores AI's value through its ability to easily review various written resources like books, articles, etc. 

Figure 7. Number of publications from 1991 to 2018 [5].

Timely disaster responses are critically important— they can literally make the difference between life and death. Those at the helm of affairs need to go all out to comprehend the situation in its entirety so as to optimize the effectiveness of response actions. This calls for a high level of situational awareness that, in turn, enables the stakeholders to make well-informed decisions with priority given to setting up a people-oriented information system for managing disaster relief. The focus should be on addressing the immediate concerns and needs of the affected population. It is in this realm AI techniques find their utility: not only can they be deployed to support response efforts but also to enhance them.

Artificial intelligence, particularly machine learning (ML), greatly influences the scrutiny of big data from social media particularly during disasters. Extracting the raw social media data and turning it into usable information. The AI can mimic the working of a human brain over time with large amounts of data thus significantly improving efficiency of various tasks and processes in addition to revealing pattern recognition in the big data [37] which is sure to have a major impact. While there has been much research on calamities for utilizing multiple machine learning techniques as means of evaluating social media data, few studies have looked at recent advances systematically — none have used AI specifically for social media disaster management involving large data analysis. Deep learning [39] stands out as one of the widely adopted techniques in system forecasting. It is applied in emergency management field to reduce the occurrence of disasters which leads to losses in both life and property. Another approach that is used to solve the forecasting problem is through integration of time series hierarchy with a deep learning method, where these two are combined.

Kumar and Sud [6] came up with a mobile application, named " DHARA", to help predict floods. The application used CNN (LSTM) based long term memory. Their main objective of developing this app was to predict future flood likelihoods which in turn helps in pre planning and restoration strategies. In [8], the study proposed a rainfall prediction model that predicts data from characteristics of environmental properties associated with rainfall.

Regression is one of the artificial intelligence methods. It consists of linear, nonlinear, and logistic regression used for estimating potential damage and losses due to hazards as well as assessing risk and damage. Support vector machines are noted for their quick prediction-making ability that is combined with a risk management capacity. Neural networks come along with hierarchical clusters, k means clusters, fuzzy clusters and principal components in the development and evaluation of mitigation strategies, trainings and disaster evacuations [9–11]. Among the deep learning methods there are convolutional neural networks, deep neural networks, multilayer perception used in disaster mapping. Damage assessment uses Q-networks, genetic algorithms while information systems for disaster management use collaboration between agencies during emergencies to employ cost assessment repair method; Deep Q-networks and genetic algorithms are the most recent methods that have been employed to assess the damage and expense of repair.

The MOBILIZE platform also introduces a technology-based interface based on VR. This new virtual reality (VR) innovation enables the user to see 3D models in such a way that they seem real from data that is obtained from airborne sensors like drones. The MOBILISE web platform visualizes risks and calculates disaster effects as it provides decision support information during emergency response using the visualization of real-time data captured by Unmanned Aerial Vehicles (UAVs).

The MOBILIZE platform is user-friendly and enables agencies to effortlessly input their hazards, exposure, and vulnerability data. This local risk information is in turn overlaid with the identified risks and tracking of actions taken during risk mitigation. The dataset that keeps information about the primary risk is hosted on an Azure-based server.

In addition, MOBILIZE adopts an interface technology-based on VR. This new virtual reality innovation allows the user to see 3D models as if they were taken by air-borne sensors such as drones. The MOBILISE web platform visualizes the risks of disaster and computes consequences based on it—MOBILIZE shows actual data during disaster responses using 3D visualization which will support the documentation of rainfall plus drainage lines as layers.. (Figures 8 and 9). 

Figure 8. Visualization of risk hazard exposure mapping using MOBILISE platform in one of Malaysian state [12].

Figure 9. Risk visualization of built environment GIS data using MOBILISE platform in one of Malaysian state [60]. 

Disaster management aims at three things: sustaining life, infrastructure and a viable environment. Geographic Information System (GIS) is an effective tool to support these components in disaster management which include disaster prevention, preparedness for disaster, response towards the disaster and recovery after the disaster. Each phase is interrelated with other phases and uses GIS for integrating spatial information and establishing a system for managing disasters [13]. There are three steps in every stage of disaster management within the GIS process. The first step involves acquiring data from reliable sources while the second step entails measurement of spatial data whereby different graphical formats are used to represent geographic entities along with their attributes— numbers, graphs or tables among others. This has to be done through a Geographic Database (GDB) [70].

The development leads to efforts and sometimes even deaths. In the time of emergencies, GIS acts as a way to organize important details visually.

GIS is a framework that holds geographic information and analyzes it. It consists of multiple elements, each composed of software, hardware, data, people, and methods. While other types of GIS software are available for purchase— commercial, open-source, personal— the most widely used program is Esri ArcGIS; custom programs use Google Earth Pro, Mapinfo Pro or Google Maps API plus batchGeo.

Geospatial analysis offers a visual and graphical representation through maps, graphs, and tables designed to improve research on flood alerts. The maps present different ways of predicting and planning for disasters before they take place— which in turn increases the efficiency of disaster agencies in taking pre-impact emergency actions while also highly reducing the environmental effects (Page 64). Early warning systems — an alternative to remote sensing plus satellite imagery — have the capacity to evaluate the effectiveness of disaster control efforts. Erial maps help in identifying the disaster area by outlining the region and locating vulnerable places where minimal damage would occur. These technologies help in reducing flood damage to the area affected by floods and limiting harm that can be brought about to the community and land itself. Such entities as Malaysian remote sensing institutions have recognized the importance of utilizing artificial intelligence and geographic information system technology for flood risk mapping.

Disaster management has three main goals: preserving life, infrastructure and the environment. GIS can help in all these areas and is used in disaster management from mitigation through preparedness, response and recovery. The different phases are interrelated and use GIS as a technology for data integration and analysis during development of a geographic information system that supports disaster management [14]. There are three steps at every stage of disaster management within GIS processes. First step is collecting reliable data which involves measurement after data has been captured so that spatial information can be provided in a graphic form (maps). In addition to numbers there are also graphs plus tables while complexes mean directions on how the use this data. This is done using Geographic Database (GDB) [15].

Disaster management has three main goals: Life preservation, saving infrastructure and conserving the environment. GIS is an asset that can help in ensuring these prerequisites of disaster management which include prevention of disaster, preparedness for disaster, response to disaster and recovery from it as each phase is interlinked with another & uses GIS for information analysis thereby establishing a system resource based on disaster management [14]. There are three stages at every level of disaster management that form the three steps in the GIS process. The first step involves data acquisition from reliable sources; spatial data is then measured and presented in a understandable form— such as numbers or graphs — along with tables and charts developed using complex spatial relationships stored inside Geographic Database (GDB) [15].

Finally, the information is completed and delivered to the receiver. In the end, it is disseminated using statistical methods. The skills of geospatial analysis are important plus very useful for spatial analysis. They play a major role in enhancing disaster management. The last and final step is to compile the information and deliver it to the address provided; ultimately, the distribution of this geospatial analysis data will be through statistical means. These capabilities on geospatial analysis are critical and important for spatial analysis plus that they contribute significantly to enhancing disaster management.

GIS has proven successful in identifying the areas favorable to develop flood mitigation systems and evaluating the effectiveness of the currently applied flood mitigation systems. The study by Puttinaovarat and Horkaew [82] introduced a system through a computer network aided by remote sensing, GIS and DL. This system was to be used with a flood prevention strategy ( see Figure 10 ) where these methods help in detecting floods at their initiation time on real-time basis which then helps in reducing false alarm numbers thereby investigation times. Darabi et al [83] employed machine learning techniques to determine the flooded area in Amol, Iran— based on geographic variables that predicted where the area would be flooded from — as an application of ML in flood management.

The use of Geographic Information Systems allows one to carry out physical and numerical analysis at every stage of disaster management cycle. These studies of past floods serve as a preview for future trends.

Many techniques have been used to assess flood risk and damage. These approaches include the AHP method, logistics regression, the analytic network method, statistical measurement, random forest and flood zoneing; a technique based on hydraulic science that consists of two maps: damage zones map and flood areas map. The first one locates the place where a flood is going to take place while the second describes the damage that will be caused by it[16]. Flood zoneing has become popular nowadays with high importance for human safety as it helps in reducing floods which are among the worst disasters killing thousands of people every year [17]. In addition, GIS coupled with HEC-RAS — hydrologic models that simulate rainfall — have successfully created river maps for Warsaw (Poland), Columbia University (USA) and Dhaka (Bangladesh) along with many other flood-prone states.

In another context, GIS has proven to be successful. In identifying areas where it is desirable to have flood mitigation systems developed and also in evaluating the effectiveness of already established flood mitigation systems. The study by Puttinaovarat and Horkaew [18] discussed a system which was meant for fostering floods mitigation through a computer network that combined remote sensing, GIS and DL. This system was to be used alongside a flood prevention strategy ( Figure 10 ). These methods help in detecting floods at the time they occur which reduces the time taken for investigations by eliminating false positives. Darabi et al.'s [19] team used ML techniques to find out the area of Amol, Iran which would be flooded based on geographic variables that predicted where flooding would occur.

Successful applications of GIS include the identification of areas suitable for developing flood mitigation systems and assessment of the efficiency of existing flood mitigation systems. Puttinaovarat and Horkaew [20] elaborated on a study that aimed at enhancing flood mitigation through a computer network system integrating remote sensing, GIS and DL. This system was to work hand in hand with a flood prevention strategy ( Figure 10 ) by ensuring real-time detection of floods, which minimizes time for investigations due to reduction in false positives. Darabi et al.'s team [21] used ML techniques to find the flooded area in Amol, Iran— where geographic variables predicated an area would be flooded based on those predictions. The recognition of flood danger was due to the distance to the channel, land use and runoff generation as major causes [22]. An obtained vulnerability map highlights the high degree of importance for flood prevention planning in vulnerable areas..[23].

Figure 10. Design of the online flood information management system [24].

The study that was undertaken aimed at evaluating the use of artificial intelligence in disaster management and its applications in enhancing disaster management. The research looked into different approaches and techniques used to identify phases of disaster management as well as an analysis that supports AI being useful in improving disaster management which indicated current interest focusing on how to respond and mitigate a disaster. In a finding, geospatial technology is still developmental but it has proven to be appropriate; not only did it provide a relevant solution for the topic but also introduced GI Science as another significant subject with which to approach disasters— therefore advocating for the need to take a geographic perspective since today GIS and RS have both matured into powerful tools offering a new approach toward understanding disaster situations.

The study sought to explore artificial intelligence in the realm of disaster management and how it can be used as a tool to improve disaster management. During the investigation, different approaches and techniques that could be used to identify phases of disaster management were discussed, and from the analysis, it can be argued that AI is very beneficial for improving disaster management; we also found that current interests lie more on response and mitigation but geospatial technology is still at infant stages which could have been useful because it falls under GIScience. Hence importance must be given to the development of a geographic perspective since today GIS and RS have both come into play as new tools offering new approaches in understanding disasters.

Disasters unveil a host of other concerns for the affected community post disaster, especially floods. Floods can lead to different types of diseases which are contagious and the most common is a virus that causes skin infection. The use of GISs is pivotal in any calamity and can readily deliver critical data that would aid in surveillance and minimizing impact of the crisis. In December 2019 a new virus — COVID-19 — was discovered. This virus leads to a respiratory disease and has high transmissibility globally. When the COVID-19 crisis was declared as a pandemic it led countries to adopt strategies focused on containment so as to slow down its spread. GIS is what provided factual information that can be used in the battles against previous virus such as Ebola epidemic, bird flu; it might also be possible to make use of mapping and plotting to help identify and track clusters of COVID-19 through integration with population. Although we are fighting COVID-19, these tools play a major role in disaster management and planning aimed at overcoming this deadly disease.

The study investigated the involvement of AI in disaster management and its use as a tool to enhance disaster management. In particular, it addressed different approaches and techniques used to identify phases of disaster management from which the analysis argued for the suitability of AI in improving disaster management. It was noted that current interest is more on response and mitigation of a disaster while geospatial technology is still in developmental stages but has provided a relevant solution to the topic— hence GI Science deserves significant attention; thus, there's need for a geographical perspective which has evolved today with GIS and RS now being powerful and offering new approach to understanding situations of disasters.

In the future, we anticipate that AI technology will be able to obtain a dataset with more details which will without doubt assist in disaster mitigation. But remote sensing data and GIS have paved way for a new scientific research field— information technology— and thus artificial intelligence can greatly contribute in addressing both natural and man-made disasters. The performance of any AI system is obviously contingent on two basic components: the level at which private data is managed (with security) and the level of computational abilities initiated into the system algorithms. In addition, effectiveness also depends on other major technological applications that support optimization strategies of successful management teams.

The authors declare that they have no conflict of interest

No funding sources

The study was approved by the University of, Mosul, Iraq

1. Pielke, R. Tracking progress on the economic costs of disasters under the indicators of the sustainable development goals. Environ. Hazards 2019, 18, 1–6. doi:10.1080/17477891.2019.1573863

2. Abid, M.; Isleem, H.F.; Shahzada, K.; Khan, A.U.; Kamal Shah, M.; Saeed, S.; Aslam, F. Seismic Hazard Assessment of Shigo Kas Hydro-Power Project (Khyber Pakhtunkhwa, Pakistan). Buildings 2021, 11, 349. doi:10.3390/buildings11080349   

3. Abid, S.K.; Sulaiman, N.; Wei, C.S. Building Resilient Future: Information Technology and Disaster Management—A Malaysian Perspective. In IOP Conference Series: Earth and Environmental Science; IOP Publishing: Jaipur, India, 2021.   doi:10.1088/1755-1315/775/1/012017

4. Costache, R.; Bui, D.T. Spatial prediction of flood potential using new ensembles of bivariate statistics and artificial intelligence: A case study at the Putna river catchment of Romania. Sci. Total Environ. 2019, 691, 1098–1118. doi:10.1016/j.scitotenv.2019.07.056

5. Chen, J.; Li, Q.; Wang, H.; Deng, M. A Machine Learning Ensemble Approach Based on Random Forest and Radial Basis Function Neural Network for Risk Evaluation of Regional Flood Disaster: A Case Study of the Yangtze River Delta, China. Int. J. Environ. Res. Public Health 2020, 17, 49. doi:10.3390/ijerph17010049

6. Sun, W.; Bocchini, P.; Davison, B.D. Applications of artificial intelligence for disaster management. Nat. Hazards 2020, 103, 2631–2689. doi:10.1007/s11069-020-04183-1

7. N. binti Sulaiman, N. P. N. binti Mahmud, U. Nazir, and S. K. Abid, "The role of autonomous robots in fourth industrial revolution (4IR) as an approach of sustainable development goals (SDG9): industry, innovation and infrastructure in handling the effect of COVID-19 outbreak," in IOP Conference Series: Earth and Environmental Science, vol. 775, no. 1, p. 012017, June 2021 doi:10.1088/1755-1315/775/1/012017.     

8. Khouj, J.M.M.; López, C.; Sarkaria, S. Disaster management in real time simulation using machine learning. In Proceedings of the 2011 24th Canadian Conference on Electrical and Computer Engineering (CCECE), Niagara Falls, ON, Canada, 8–11 May 2011; pp. 1507–1510.   doi:10.1109/CCECE.2011.6030702     

9. Munawar, H.S.; Ullah, F.; Qayyum, S.; Khan, S.I.; Mojtahedi, M. UAVs in Disaster Management: Application of Integrated Aerial Imagery and Convolutional Neural Network for Flood Detection. Sustainability 2021, 13, 7547. doi:10.3390/su13147547

10. Oakley, M.; Himmelweit, S.M.; Leinster, P.; Casado, M.R. Protection Motivation Theory: A Proposed Theoretical Extension and Moving beyond Rationality—The Case of Flooding. Water 2020, 12, 1848.  doi:10.3390/w12071848

11. Kabir, M.Y.; Gruzdev, S.; Madria, S. STIMULATE: A System for Real-time Information Acquisition and Learning for Disaster Management. In Proceedings of the 2020 21st IEEE International Conference on Mobile Data Management (MDM), Versailles, France, 30 June–3 July 2020; pp. 186–193. doi:10.1109/MDM50115.2020.00037 

12. Doshi, J.; Basu, S.; Pang, G. From Satellite Imagery to Disaster Insights. no. Nips. 2018. Available online: http://arxiv.org/abs/1812.07033 (accessed on 17 June 2021).     doi:10.48550/arXiv.1812.07033

13. Vetrivel, A.; Gerke, M.; Kerle, N.; Nex, F.; Vosselman, G. Disaster damage detection through synergistic use of deep learning and 3D point cloud features derived from very high resolution oblique aerial images, and multiple-kernel-learning. ISPRS J. Photogramm. Remote Sens. 2018, 140, 45–59.    doi:10.1016/j.isprsjprs.2018.03.009

14. A Collaborative Multi-Agency Platform for Building Resilient Communities. 2019. Available online: https://mobilisesarawak. uthm.edu.my/ (accessed on 20 May 2021). 

15. Termeh, S.V.R.; Kornejady, A.; Pourghasemi, H.R.; Keesstra, S. Flood susceptibility mapping using novel ensembles of adaptive neuro fuzzy inference system and metaheuristic algorithms. Sci. Total Environ. 2018, 615, 438–451.  doi:10.1016/j.scitotenv.2017.09.277

16. Sabri, S.; Yeganeh, N. Flood Vulnerability Assessment in Iskandar Malaysia Using Multi-criteria Evaluation and Fuzzy Logic. Res. J. Appl. Sci. Eng. Technol. 2014, 8, 1794–1806. doi:10.19026/rjaset.8.1045 

17. Paul, P.K.; Aithal, P.S.; Bhuimali, A.; Tiwary, K.S.; Saavedra, R.; Aremu, B. Geo Information Systems & Remote Sensing: Applications in Environmental Systems & Management Geo Information Systems & Remote Sensing: Applications in Environmental Systems & Management. Int. J. Manag. Technol. Soc. Sci. 2020, 5, 11–18. doi:10.46281/ijmtss.v5i1.211 

18. Fariza, A.; Rusydi, I.; Hasim, J.A.N.; Basofi, A. Spatial flood risk mapping in east Java, Indonesia, using analytic hierarchy process—Natural breaks classification. In Proceedings of the 2017 2nd International conferences on Information Technology, Information Systems and Electrical Engineering (ICITISEE), Yogyakarta, Indonesia, 1–2 November 2017; pp. 406–411. doi:10.1109/ICITISEE.2017.8250477

19. Thomas, C.F. An Introduction to Geographic Information Systems. Libraries 2020, 183–204. doi:10.5860/libraries.114.3.183

20. der Sarkissian, R.; Zaninetti, J.M.; Abdallah, C. The use of geospatial information as support for Disaster Risk Reduction; contextualization to Baalbek-Hermel Governorate/Lebanon. Appl. Geogr. 2019, 111, 102075. doi:10.1016/j.apgeog.2019.102075

21. Hoque, M.A.A.; Tasfia, S.; Ahmed, N.; Pradhan, B. Assessing spatial flood vulnerability at kalapara upazila in Bangladesh using an analytic hierarchy process. Sensors 2019, 19, 1302. doi:10.3390/s19061302

22. Vojtek, M.; Vojteková, J. Flood hazard and flood risk assessment at the local spatial scale: A case study. Geomat. Nat. Hazards Risk 2016, 7, 1973–1992.   doi:10.1080/19475705.2016.1140335

23. Curebal, I.; Efe, R.; Ozdemir, H.; Soykan, A.; Sönmez, S. GIS-based approach for flood analysis: Case study of Keçidere flash flood event (Turkey). Geocarto Int. 2016, 31, 355–366. doi:10.1080/10106049.2015.1032085

Demir, V.; Kisi, O. Flood Hazard Mapping by Using Geographic Information System and Hydraulic Model: Mert River, Samsun, Turkey. Adv. Meteorol. 2016, 2016, 4891015. doi:10.1155/2016/4891015