Unexpected Disasters Research Articles

A new application of deep neural network (LSTM) and RUSLE models in soil erosion prediction

Rainfall variation causes frequent unexpected disasters all over the world. Increasing rainfall intensity significantly escalates soil erosion and soil erosion related hazards. Forecasting accurate rainfall helps early detection of soil erosion vulnerability and can minimise the damages by taking appropriate measures caused by severe storms, droughts and floods. This study aims to predict soil erosion probability using the deep learning approach: long short-term memory neural network model (LSTM) and revised universal soil loss equation (RUSLE) model. Daily rainfall data were gathered from five agro-meteorological stations in the Central Highlands of Sri Lanka from 1990 to 2021 and fed into the LSTM model simulation. The LSTM model was forecasted with the time-series monthly rainfall data for a long lead time period, rainfall values for next 36 months in each station. Geo-informatics tools were used to create the rainfall erosivity map layer for the year 2024. The RUSLE model prediction indicates the average annual soil erosion over the Highlands will be 11.92 t/ha/yr. Soil erosion susceptibility map suggests around 30 % of the land area will be categorised as moderate to very-high soil erosion susceptible classes. The resulted map layer was validated using past soil erosion map layers developed for 2000, 2010 and 2019. The soil erosion susceptibility map indicates an accuracy of 0.93 with the area under the receiver operator characteristic curve (AUC-ROC), showing a satisfactory prediction performance. These findings will be helpful in policy-level decision making and researchers can further tested different deep learning models with the RUSLE model to enhance the prediction capability of soil erosion probability.

Identification of spaces with cluster infection risks in small cities in China based on spatial syntax and GIS

Among modern cities developing in a large-scale, extensive and unbalanced manner, smaller cities are relatively lagged behind due to relatively underdeveloped infrastructure, inadequate capital and technology talents, and insufficient attention from the national government, and thus they are more vulnerable when hit by unexpected disasters. The rampant pandemic of coronavirus disease 2019 (COVID-19) has made it even clearer that small cities must be equipped with stronger abilities to timely identify and prevent potential disease outbreaks. This paper takes Zhaodong City as an example to study how to better locate spaces with cluster infection risks in small cities. It combines spatial syntax, points of interest (POI), and geographical information system (GIS), and adopts hotspot analysis, average nearest neighbour analysis, kernel density estimation and other methods, to identify and locate potentially vulnerable spaces in neighbourhoods with relatively frequent people-to-people contact and thus higher disease transmission risks. Results show that there are three point-space, four line-space, and one plane-space with high risk of outbreaks in Zhaodong City, verifying the efficacy of the identification method for small cities.

Electrochemical measurements used for assessment of corrosion and protection of metallic materials in the field: A critical review

Corrosion degradation is detrimental to metal structures as it shortens their lifetime and leads to huge economic losses and unexpected disasters. Therefore, the detection and monitoring of corrosion degradation is of great importance. Herein, we briefly review the state-of-the-art electrochemical methods, instrumentation (based on virtual instrumentation), and advanced sensor/probes that are used in the field for the assessment of corrosion damage. Typical corrosion monitoring results, some of which have been obtained at Tianjin University in the past 30 years, for metallic materials and organic coating/metal systems in atmospheric, marine, and soil conditions in the field are presented. Detection methods, data analysis, and theoretical and mathematical models regarding each corrosion system are discussed, and the challenges, problems and possible solutions for each case are suggested. Lastly, future developments, such as wireless, intelligent, and automatic electrochemical measurement, that will augment the present electrochemical methods of evaluating corrosion degradation are summarized.

Examining the Performance Measurement Factors Impacting on the Effectiveness of Disaster Management System

Body The purpose of this study is to examine the direction of disaster management by administration change in Korea and identify factors that could accommodate the public's demand for an effective disaster management system. Keyword network analysis is conducted for the 3rd and 4th National Emergency Management Plan to explore the changes in the direction of disaster management policies. Through the analysis, topic words, their frequency, and matrixes for co-occurrence between keywords are extracted. Based on the extracted matrix, the degree centrality (connectivity of keywords) is identified and cluster analysis has been conducted afterward. The findings show that education training, on-site disaster response, improvement of legal system, professional training, and cooperation among related agencies are detected as key topic words. The direction of the plan is also changing to cope with the unexpected disaster situation due to the climate change.

A Comprehensive View on the effects of Hyperautomation in Business and in Employment of Humans

We, at present, are witnessing a severe and faster change in the field of intelligence than in any other period in the history of human evolution and it is most likely to evolve at the same pace or even much faster in the coming years. What was once considered ‘physical involvement of humans’ is becoming obsolete with technology advancement, global unexpected disasters like climate change, viral diseases and even political issues which in turn causing an unprecedented turmoil of larger magnitude as well as opportunity for both society and businesses alike. All these technology advancements need to be filtered down into a perfect system which leads to the evolution of new field called Hyperautomatin.This hyperautomation, still in its early stage of adaptation, impacts the lives of millions of ordinary people and businesses. This paper is about the impact of this technology and how it is impacting in varying scales.

An optimization approach for disaster relief network design under uncertainty and disruption with sustainability considerations

Human-made, natural, and unexpected disasters always cause human and financial losses to communities. Disaster management is a framework with proven performance to reduce the damage caused by disaster and supply chain disruptions. Transferring the injured people from affected areas to hospitals at the minimum possible time is a crucial goal in times of disaster. This paper develops a two-stage stochastic programming model to transport the injured people from affected areas to hospitals in the incidence of multiple disruptions at transportation links and facilities under uncertainties. Herein, economic, social, and environmental aspects of sustainability are considered, while simultaneous disruptions are managed to minimize the adverse impacts of the disasters. We aim to determine optimal locations to establish transfer points and flows between the relief network nodes with sustainability considerations. Ultimately, a case study in District 12 of Tehran, Iran is conducted to ensure the proposed model’s validity and performance. Various sensitivity analyses are also implemented to ensure the model’s effectiveness. The results indicate that disruptions in facilities and transportation links lead to increased relief time, hence has the most significant negative impact on relief operations.

Current Challenges and Prospective Benefits of Using UAVs

The modern economy completely changed by offering to industry groups the capability to reorganize their operating processes. Nowadays we become observers of similar in scale revolutionary transformations: technologies for the use of unmanned aerial vehicles substantially modified business models and new operating conditions have been developed in various sectors from industries, agriculture to the emergency, parcel delivery or movie industry. In the very near future, clients’ enterprises from multiple sectors of the economy will recognize the first effect of the use of unmanned aerial vehicles (UAVs) in diverse fields - from commercial sectors to examining and exploring the environment or to reduce the impacts of unexpected disasters. The aim of the present review paper is to analyse the opportunities and advantages of using the application of UAVs technologies. On the other hand, not only the devices (drones) themselves are of importance, but also their wider utilization to acquire extraordinary volumes of data.

Prediksi Curah hujan di Kota Pekanbaru Menggunakan lSTM (Long Short Term Memory)

Based on data obtained from BMKG Pekanbaru City in 2010-2020 there was an increase and decrease in the intensity of rainfall that occurred in Pekanbaru city. The increase in rainfall in the city of Pekanbaru will cause problems such as the occurrence of flooding of several roads and several areas in the city of Pekanbaru and the occurrence of other unexpected disasters that will cause problems and experience difficulties. In overcoming this problem, research was conducted in the form of Rainfall Prediction in Pekanbaru City Using LSTM (Long Short Term Memory) using 2 methods, namely in finding the accuracy of the error rate using RMSE (Root Mean Square Error) and MSE (Mean Square Error). The results showed that the predictions made were quite good. With the lowest error rate of 21,328 in the train and 454,901 in the test, the composition of the train data and the test data half gave the best results.

Analyzing Healthcare Facility Resilience: Scientometric Review and Knowledge Map.

In contemporary “high-risk” society, unexpected disasters (epidemics and extreme weather) and chronic pressures (aging problems) put tremendous pressure on healthcare facilities. Enhancing the healthcare facilities' resilience ability to resist, absorb, and respond to disaster disruptions is urgent. This study presents a scientometric review for healthcare facility resilience research. A total of 374 relevant articles published between 2000 and 2020, collected from Web of Science (WoS) core collection database, Scopus database and MEDLINE database were reviewed and analyzed. The results indicated that research on resilience in healthcare facilities went through three development periods, and the research involved countries or institutions that are relatively scattered. The studies have been focused on the subject categories of engineering, public, environmental, and occupational health. The keywords of “resilience,” “hospital,” “disaster,” “healthcare,” and “healthcare facility” had the most frequency. Furthermore, based on the literature co-citation networks and content analysis, the detected seven co-citation clusters were grouped into four knowledge domains: climate change impact, strengthening resilience in response to war and epidemic, resilience assessment of healthcare facility, and the applications of information system. Moreover, the timeline view of literature reflected the evolution of each domain. Finally, a knowledge map for resilience of healthcare facilities was put forward, in which critical research contents, current knowledge gaps, and future research work were discussed. This contribution will promote researchers and practitioners to detect the hot topics, fill the knowledge gaps, and extend the body of research on resilience of healthcare facilities.

KEARIFAN LOKAL MASYARAKAT DESA ADAT SADE DALAM MENGHADAPI COVID-19

Local wisdom of indigenous village communities can be interpreted as ideas, principles of life, behavior, rules or norms, taboos, and punishments derived from the experience of hereditary ancestors, one of which is to deal with life problems, namely disasters. COVID-19 is a disaster that causes difficult conditions to arise in society because it has an impact on all aspects of life. Therefore, people should fight against not only COVID-19 virus attack, but also food scarcity, and hunger. This study refers to the theory of local knowledge suggested by Geertz (1983) stating that local knowledge is born from experiences that have been passed in dealing with disasters and is knowledge owned by local communities and managed locally, in which there is a "cultural" system that aims to unite members within it. With the local wisdom that Sade Village has as a traditional village, the people have their way of dealing with COVID-19. Therefore, the purpose of this study is to identify the local wisdom of the Sade Village community in dealing with COVID-19. This research uses a descriptive qualitative method and the data is analyzed using triangulation. This study finds that local wisdom is used as a living strategy by optimally utilizing natural capital to deal with unexpected disasters. The tradition of togetherness, cooperation, and mutual assistance carried out by the people of Sade Village has made no one infected with the COVID-19 virus and can face food shortages, so this tradition needs to be followed by an example to deal with COVID-19. Another finding is the need to preserve agricultural work as a way of survival.

Management, Performance, and Measurement of Organizational Resilience

The aim of this paper is to analyze the different implications between management, performance, and measurement of organizational resilience from the different ontologies and methodologies emerging to protect organizations against disturbances, disruptions, and disasters such as risk and crisis management and continuity management. The underlying assumptions are based on the lack of clarification of the relationships among these concepts despite the strong background in various disciplines, as well as the required assessment of unknown threats, assuming measures of a general resilience to enhance abilities to face unanticipated and unexpected events and disasters. The method employed is the analytical reflective of theoretical and empirical literature review sustained in a holistic approach. It is concluded that each discipline based on its own ontology provides concepts, methodologies, and tools to implement and measure organizational resilience management performance linked to organizational levels and offering solutions.

Tunnel disturbance events monitoring and recognition with distributed acoustic sensing (DAS)

Accurately identifying disturbance events along tunnels is essential for their safe operation, which constitutes an important part of tunnel health monitoring and abnormity warning. In recent years, distributed acoustic sensing (DAS), a state-of-the-art fiber-optic sensing technology, has developed rapidly in the field of earth science and engineering. Based on the principle of phase-sensitive optical time-domain reflectometry, DAS allows detecting acoustic/vibration signals along a common fiber-optic cable up to tens of kilometers. This brings new opportunities for monitoring of long perimeters such as underground tunnels. In this paper, we propose a DAS-based method for the recognition of disturbance events along tunnels. The EMD denoising algorithm is employed to optimize vibration signals to better extract the time–frequency domain features of monitored events. Different events are then recognized via a machine learning-based multi-class classification approach. The Random Forest algorithm is applied to analyze the DAS data acquired with fiber-optic cables deployed along the tunnel lining, successfully recognizing a variety of vibration events during the construction of the tunnel including unexpected disasters such as rockfalls, with a recognition accuracy of 92.31%. This DAS-based disturbance identification method may provide a new opportunity for unmanned, real-time monitoring of tunnel abnormal events.

A Data-Driven Situational Awareness System for Enhanced Air Cargo Operations Emergency Control

Based on the constant need for safety and operational cost optimization, the air-cargo industry is continually evolving in the context of Industry 4.0. Used wisely, data can help the industry to provide critical resilience that will allow authorities to take proper measures/actions in response to unexpected disasters and secure societal protection. The “INTELLICONT” project combines state-of-the-art technologies blended with novel solutions to improve the loading/unloading time, the structural status awareness, and the safety and security of the air-cargo related operations (prior to, during, and after the flight), as well as to enhance their capabilities related to the execution of their duties. The suggested system is contextually aligned and harmonized with the existing international and EU regulations. In the present work, the remote monitoring and control system for intelligent aircraft cargo containers have been presented from the software perspective. The intelligent containers integrate three types of sensors, Structural Health Monitoring, fire suppression, and locking status indication. The focus has been given to the design and development of a Human Machine Interface (HMI) capable to visualize all related data for better and safer control of the aircraft cargo. It is shown that the system can contribute to making the air transportations safer, environmentally friendlier, faster and with the lowest possible cost.

A deep learning based hazardous materials (HAZMAT) sign detection robot with restricted computational resources

One of the most challenging and non-trivial tasks in robot-based rescue operations is the Hazardous Materials (HAZMAT) sign detection in dangerous operation fields, in order to prevent further unexpected disasters. Each HAZMAT sign has a specific meaning that the rescue robot should detect and interpret to take a safe action, accordingly. Accurate HAZMAT detection and real-time processing are the two most important factors in such robotics applications. Furthermore, the rescue robot should cope with some secondary challenges such as image distortion and restricted CPU and computational resources, embedded in the robot. In this research, we propose a CNN-Based pipeline called DeepHAZMAT for HAZMAT sign detection and segmentation in four steps: (1) Input data volume optimisation before feeding into the CNN network, (2) Application of a YOLO-based structure to collect the required visual information from the hazardous areas, (3) HAZMAT sign segmentation and separation from the background using adaptive GrabCut technique, and (4) Post-processing optimisation using morphological operators and convex hull algorithms. In spite of the utilisation of a very limited CPU and memory resources, the experimental results show the proposed method has successfully maintained a better performance in terms of detection-speed and detection-accuracy, compared to classical and modern state-of-the-art methods.

Multilatinas and COVID-19: initial actions to the pandemic

ObjetivoA América Latina não está isenta dos efeitos da pandemia COVID-19. Portanto, este estudo tem como objetivo identificar as ações das 50 principais multilatinas para o enfrentamento da pandemia.Design/metodologia/abordagemUma análise qualitativa de conteúdo foi realizada em jornais publicados na América Latina e sites corporativos de março de 2020 a abril de 2020, período em que as multilatinas realizaram suas primeiras ações em relação à pandemia.ResultadosOs resultados identificaram 106 ações realizadas pelas multilatinas nas áreas social, operacional, filantrópica e financeira. Analisamos os resultados, discutimos as implicações para as teorias de negócios e desenvolvemos três proposições teóricas.Implicações práticasprimeiro, este estudo pode servir como uma referência para os tomadores de decisão de negócios, pois oferece orientação na tomada de medidas durante uma pandemia. Em segundo lugar, destaca a importância da velocidade de resposta a desafios imprevistos. O estudo recomenda que os gestores de negócios incluam em seus planos de contingência procedimentos para enfrentar situações imprevistas, como a pandemia, com foco no aumento de sua capacidade de resposta.OriginalidadLa importancia de este estudio recae en el entendimiento de las acciones ocorres for las empresas multinacionais durante desastres inesperados y eventos raros. Este estudio cubre dos huecos importantes na literatura que não foi público no contexto da América Latina: las acciones por las multinacionales encarando um desastre de salud e como las multinacionales afronta los eventos raros y desastres.

Analysis of cognitive skill in a water discharge activity for firefighting robots

Teleoperated robots are expected to perform tasks in unexpected disaster sites, which have become increasingly common in recent years. However, only few teleoperated robots are used for such purposes owing to their poor usability compared to the handling of such tasks by humans. Further, human-machine interfaces (HMI) used in teleoperated robots are closely related to human skills. Firefighting activities require cognitive skill to make appropriate decisions. In extinguishing a fire or attenuating a leaked combustible gas, firefighting activities consist of two tasks: reconnaissance and water discharge to stop the fire or attenuate the leaked combustible gas, and these tasks possess different properties. In this study, we compared the cognitive skill of ten firefighters (five novices and five experts) in attenuating the leaked combustible gas, the latter of the said two tasks, using a teleoperated robot at a fire department. The experimental results indicated that experts were more multifaceted in their evaluation of the attenuation of combustible gas as compared to novices. In addition, as this result related to firefighter decision-making mechanisms, we were motivated to improve the HMI of firefighting teleoperated robots.

「想定外」は回避できるか 科学と社会の相克

「想定外」は回避できるか 科学と社会の相克

On the Generalization Ability of a Global Model for Rapid Building Mapping from Heterogeneous Satellite Images of Multiple Natural Disaster Scenarios

Post-classification comparison using pre- and post-event remote-sensing images is a common way to quickly assess the impacts of a natural disaster on buildings. Both the effectiveness and efficiency of post-classification comparison heavily depend on the classifier’s precision and generalization abilities. In practice, practitioners used to train a novel image classifier for an unexpected disaster from scratch in order to evaluate building damage. Recently, it has become feasible to train a deep learning model to recognize buildings from very high-resolution images from all over the world. In this paper, we first evaluate the generalization ability of a global model trained on aerial images using post-disaster satellite images. Then, we systemically analyse three kinds of method to promote its generalization ability for post-disaster satellite images, i.e., fine-tune the model using very few training samples randomly selected from each disaster, transfer the style of postdisaster satellite images using the CycleGAN, and perform feature transformation using domain adversarial training. The xBD satellite images used in our experiment consist of 14 different events from six kinds of frequently occurring disaster types around the world, i.e., hurricanes, tornadoes, earthquakes, tsunamis, floods and wildfires. The experimental results show that the three methods can significantly promote the accuracy of the global model in terms of building mapping, and it is promising to conduct post-classification comparison using an existing global model coupled with an advanced transfer-learning method to quickly extract the damage information of buildings.

Large-scale simulation of traffic flow using Markov model.

Modeling and simulating movement of vehicles in established transportation infrastructures, especially in large urban road networks is an important task. It helps in understanding and handling traffic problems, optimizing traffic regulations and adapting the traffic management in real time for unexpected disaster events. A mathematically rigorous stochastic model that can be used for traffic analysis was proposed earlier by other researchers which is based on an interplay between graph and Markov chain theories. This model provides a transition probability matrix which describes the traffic’s dynamic with its unique stationary distribution of the vehicles on the road network. In this paper, a new parametrization is presented for this model by introducing the concept of two-dimensional stationary distribution which can handle the traffic’s dynamic together with the vehicles’ distribution. In addition, the weighted least squares estimation method is applied for estimating this new parameter matrix using trajectory data. In a case study, we apply our method on the Taxi Trajectory Prediction dataset and road network data from the OpenStreetMap project, both available publicly. To test our approach, we have implemented the proposed model in software. We have run simulations in medium and large scales and both the model and estimation procedure, based on artificial and real datasets, have been proved satisfactory and superior to the frequency based maximum likelihood method. In a real application, we have unfolded a stationary distribution on the map graph of Porto, based on the dataset. The approach described here combines techniques which, when used together to analyze traffic on large road networks, has not previously been reported.

Detection of damaged buildings after an earthquake with convolutional neural networks in conjunction with image segmentation

Detecting damaged buildings after an earthquake as quickly as possible is important for emergency teams to reach these buildings and save the lives of many people. Today, damaged buildings after the earthquake are carried out by the survivors contacting the authorities or using some air vehicles such as helicopters. In this study, AI-based systems were tested to detect damaged or destroyed buildings by integrating into street camera systems after unexpected disasters. For this purpose, we have used VGG-16, VGG-19, and NASNet convolutional neural network models which are often used for image recognition problems in the literature to detect damaged buildings. In order to effectively implement these models, we have first segmented all the images with the K-means clustering algorithm. Thereafter, for the first phase of this study, segmented images labeled “damaged buildings” and “normal” were classified and the VGG-19 model was the most successful model with a 90% accuracy in the test set. Besides, as the second phase of the study, we have created a multiclass classification problem by labeling segmented images as “damaged buildings,” “less damaged buildings,” and “normal.” The same three architectures are used to achieve the most accurate classification results on the test set. VGG-19 and VGG-16, and NASNet have achieved considerable success in the test set with about 70%, 67%, and 62% accuracy, respectively.