Dissertations / Theses on the topic 'Early Warnig System'

Failure to recognize and respond to early signs of deterioration in hospitalized patients can have significant implications associated with delays in treatment. This lack of recognition was the impetus for rapid response teams in the United States and the recommendation by the Institute of Healthcare Improvement for use of early warning scores. This project was designed to evaluate the pilot implementation of an early warning score on 2 units in a community hospital in the Northeast. The practice-focused question was used to explore how patient outcomes changed following implementation of an early warning score (EWS) compared to patient outcomes associated with a rapid response team alone. The translating evidence into practice model informed this project. Supporting evidence from existing hospital data was collected for rapid response, code blue, and mortality. Analysis using the chi-square test of homogeneity compared post-implementation with baseline data. The findings indicated the differences between the proportions were not statistically significant, indicating the metrics did not change appreciably following the implementation of the early warning score. While the evaluation analytics of this pilot did not demonstrate significant change in the outcome measures post-implementation, the results may be useful for the facility when performing a future evaluation of the EWS. It is possible that the results of the 2 units were not representative of the facility, and it is therefore recommended to repeat the evaluation using data from the entire facility for a longer period. Increasing the capacity for early recognition in decline has implications for social change through improvement in safety and quality of health care for all hospitalized patients.

Risk management in healthcare has solved a wide range of healthcare-related issues in Saudi Arabia. However, the limitation of risk management teams working under special conditions (needing to solve critical health-related issues) has highlighted the urgent need for an early risk warning system (ERWS) in healthcare. The influences of changing weather conditions demand that diabetic patients and doctors in Saudi Arabia have a continuous check on health conditions. The number of diabetic patients is increasing rapidly in Saudi Arabia. Hence, risk management teams in healthcare must be supported with a system that alerts to changes before the changes become a significant risk/problem. Our proposed approach does the following: 1) predicts changes in BP and blood sugar level within hospital environment at runtime. 2) Continually checks patient health status with respect to health condition at runtime. 3) Alerts to the changes as detected (e.g. risk or unknown parameter), and also provides feedback for patient and doctor. We present a computational model that defines the interaction and communication of the system components and describes the prediction and checking process in our proposed approach. We designed the architecture for our proposed approach with respect to the computational model. The thesis proposes an early risk warning system approach, which predicts and checks patient health conditions with respect to the ideal conditions according to medical standards. The health status of a patient will be communicated to doctors and patients on an emergency note if the predicted values are outside normal conditions. In this way, the risk can be mitigated before the occurrence of damage to patient health at runtime. To implement the proposed approach, neural networks is used for developing the prediction component using Java programming. The results of this research successfully predicted the health condition of a patient by checking outputs against medical standards. The risks defined in this research include hyperglycaemia, hypoglycaemia, hypertension and hypotension. Appropriate results were obtained for almost every patient when checked with four input parameters for 200 patients. Consistent results were produced by the risk prediction component and the alerts were generated after every five (5) seconds to communicate to the patients and doctors at runtime. Health status of all 200 patients can also be seen to check the changes in health conditions in the hospital environment. Finally, a case study with different scenarios based on changes in patient health status with respect to ideal conditions revealed evaluated the approach.

Tese (doutorado)—Universidade de Brasília, Faculdade de Economia, Administração e Contabilidade e Gestão Pública, Programa de Pós-Graduação em Administração, 2017.
Submitted by Raquel Almeida (raquel.df13@gmail.com) on 2017-10-24T17:50:03Z No. of bitstreams: 1 2017_JorgeHenriquedeFriasBarbosa.pdf: 3875803 bytes, checksum: 2fd1608eb0ac0d76f29b924898b06b59 (MD5)
Approved for entry into archive by Raquel Viana (raquelviana@bce.unb.br) on 2017-10-31T12:15:52Z (GMT) No. of bitstreams: 1 2017_JorgeHenriquedeFriasBarbosa.pdf: 3875803 bytes, checksum: 2fd1608eb0ac0d76f29b924898b06b59 (MD5)
Made available in DSpace on 2017-10-31T12:15:52Z (GMT). No. of bitstreams: 1 2017_JorgeHenriquedeFriasBarbosa.pdf: 3875803 bytes, checksum: 2fd1608eb0ac0d76f29b924898b06b59 (MD5) Previous issue date: 2017-10-31
Esta tese é composta por três artigos que cobrem tópicos sobre o tema de early warning system para crises bancárias e distress bancário: uma pesquisa bibliométrica sobre early warning system (EWS) para crises bancárias e distress, um estudo empírico que estima um early warning system para distress de bancos brasileiros com regressão logística e um estudo empírico que constrói um early warning system com técnicas de aprendizagem de máquina supervisionada. O primeiro artigo apresenta um panorama do estado da literatura sobre EWS para crises bancárias e distress bancário por meio de uma revisão bibliométrica da literatura apresentando as principais ideias, principais conceitos, principais relacionamentos com outros tipos de crises, principais métodos utilizados, principais indicadores de crises e de distress. Foi realizada uma pesquisa em nas bases da Scopus e da Web of Science, onde, a partir de critérios de seleção, foram encontrados 124 artigos que foram devidamente classificados e codificados mediante importantes critérios para a área de estudo. Foi apresentado a evolução dos estudos na área, as gerações e tipos de EWS e os principais indicadores micro e macroprudencias apresentados pelos estudos da amostra. Como um resultado das lacunas da literatura na área é proposta uma agenda estruturada, visando guiar novos estudos por meio da apresentação de lacunas com grande potencial para ser explorada e reforçar o estado da arte em EWS. Adicionalmente, os resultados demonstram que mais estudos são necessários em EWS com relação à determinação dos horizontes de tempo para as previsões do modelo, com relação a estudos que tratam da América do Sul, América Central e África. Futuros estudos também devem considerar a possibilidade de utilização de modelos de aprendizagem de máquina, inteligência artificial e métodos computacionais, pois ainda existem poucos estudos e os resultados são promissores. O segundo artigo contribuiu com algumas inovações, como a construção e utilização de uma nova base dados de eventos de distress de bancos brasileiros, incluindo 179 eventos considerados como distress bancário de acordo com a definição de ?, incluindo 8 casos de RAET, 9 casos de intervenção, um caso de PROER, 11 casos de privatizações, 32 casos de incorporação e fusão, 13 casos de transformação em outros tipos de instituições financeiras, 32 caso de transformação de bancos em outros tipos de instituições, 21 casos de cancelamento e 52 casos de liquidação extrajudicial. Foi construído um painel de dados a partir de 54.087 balancetes de 359 bancos, englobando o período de julho de 1994 a novembro de 2016, juntamente com dados do setor bancário brasileiro e dados macroeconômicos. Para tratar do problema de eventos raros. O presente estudo utilizou a abordagem SMOTE (Synthetic Minority Over-sampling Technique) que pode aumentar a performance do modelo em termos da área sob a curva ROC (Area under the Receiver Operating Characteristic curve - AUC), uma técnica que que maximiza a área sob a curva ROC (AUC - area under the curve). Outra contribuição do segundo estudo foia comparação de modelos de acordo com o horizonte de tempo das previsões, característica importante para um EWS. Verificou-se que o modelo com o horizonte de tempo de 6 meses foi o modelo com maior área sob a curva ROC, para os dados da amostra utilizada, considerando-se o período de julho de 1994 até novembro de 2016. No terceiro artigo, foram utilizadas duas técnicas de aprendizagem de máquina supervisionada para construir EWSs: random forest e SVM (support vector machines) que obtiveram resultados superiores ao modelo de regressão logística apresentado no segundo estudo. Ambos os modelos de aprendizagem de máquina superam a regressão logística, em termos de acurácia, área sob a curva AUC (Area Under the Curve –AUC), sensibilidade (valor preditivo positivo) e especificidade (valor preditivo negativo). E o modelo random forest também superou o SVM em termos de acurácia, área sob a curva (AUC), sensibilidade e especificidade. Verificou-se também que os modelos random forest apresentaram melhor qualidade de previsão com as janelas de tempo de 32 e 34 meses, mostrando-se adequados às necessidades das autoridades.
This thesis consistis of three articles covering topics in early warning system (EWS) for bank crises and distress: an empirical study that estimates an early warning system for distress of Brazilian banks with logistic regression and an empirical study that builds an early warning system with techniques Of supervised machine learning. The first article presents an overview of the literature on EWS for bank crises and bank distress through a bibliometric review of the literature presenting the main ideas, main concepts, main relationships with other types of crises, main methods used, main crisis indicators And distress. A survey was carried out in the databases of Scopus and the Web of Science, where, based on selection criteria, 124 articles were found that were duly classified and codified by important criteria for the study area. The evolution of the studies in the area, the generations and types of EWS and the main micro and macroprudential indicators presented by the sample studies were presented. As a result of the literature gaps in the area, a structured agenda is proposed, aimed at guiding new studies through the presentation of gaps with great potential to be explored and to reinforce the state of the art in EWS. In addition, the results demonstrate that more studies are needed in EWS regarding the determination of time horizons for model predictions, in relation to studies dealing with South America, Central America and Africa. Future studies should also consider the possibility of using machine learning models, artificial intelligence and computational methods, as there are still few studies and the results are promising. The article contributed some innovations such as the construction and use of a new database of distress events of Brazilian banks, including 179 events considered as bank distress according to the definition of ?, including 8 cases of RAET (Temporary Special Administration Scheme), 9 cases of intervention, one PROER (The Program of Incentives for the Restructuring and Strengthening of the National Financial System) case, 11 cases of privatization, 32 cases of incorporation and merger, 13 cases of transformation in other types of financial institutions, 32 cases of transformation of banks into other types of institutions, 21 cases of cancellation and 52 cases of extrajudicial liquidation. A data panel was constructed from 54,087 balance sheets of 359 banks, covering the period from July 1994 to November 2016, together with data from the Brazilian banking sector and macroeconomic data. In order to address the problem of rare events, the present study used the Synthetic Minority Over-sampling Technique (SMOTE) approach that can increase the model’s performance in terms of the Area under the Receiver Operating Characteristic curve (AUC), a technique that maximizes the area under the ROC curve (AUC). Another contribution of the second study was the comparison of models according to the time horizon of the forecasts, an important feature for an EWS. It was verified that the model with the time horizon of 6 months was the model with the largest area under the ROC curve, for the data of the sample used, considering the period from July 1994 to November 2016. In the third article, two supervised machine learning techniques were used to construct EWSs: random forest and SVM (support vector machines) that obtained results superior to the logistic regression model presented in the second study. Both models of machine learning outperform logistic regression in terms of accuracy, area under the AUC curve, sensitivity (positive predictive value) and specificity (negative predictive value). And the random forest model also surpassed the SVM in terms of accuracy, area under the curve (AUC), sensitivity and specificity. It was also verified that the random forest models presented better quality of prediction with the forecast time horizons of 32 and 34 months, being adapted to the needs of the authorities.

The nation has placed a spotlight on improving graduation rates for all students. The current study analyzed retrospective, longitudinal student data from the fifth largest school district in Oregon to create an Early Warning Indicator System (EWS) to identify students on track to graduate and those who are not. This study creates an EWS system using the student demographics and the ABC’s of (a) attendance, (b) behavior, and (c) coursework to identify students who are on track and those who are not. I employed logistic regression model to build a prediction model using middle school data (N = 2,041) that examined predictors established in sixth through eighth grade with high school graduation. The dependent variable, four-year graduation was coded as graduate or non-graduate. The independent variables were (a) gender, (b) race, (c) ELL status, (d) SPED Status (e) attendance rate, (f) ODR’s, and (g) number of F’s in English Language Arts and Mathematics. Attendance rate was the strongest predictor of high school graduation. Overall the model predicted graduates with 89.7% accuracy and non-graduates with 33.6% accuracy with the total model predicting 69.5% of graduates and non-graduates.

This thesis examines how early warning signals perform when tested on climate systems thought to exhibit future tipping point behaviour. A tipping point in a dynamical system is a large and sudden change to the state of the system, usually caused by changes in external forcing. This is due to the state the system occupies becoming unstable, causing the system to settle to a new stable state. In many cases, there is a degree of irreversibility once the tipping point has been passed, preventing the system from reverting back to its original state without a large reversal in forcing. Passing tipping points in climate systems, such as the Amazon rainforest or the Atlantic Meridional Overturning Circulation, is particularly dangerous as the effects of this will be globally felt. Fortunately there is potential for early warning signals, designed to warn that the system is approaching a tipping point. Generally, these early warning signals are based on analysis of the time series of the system, such as searching for ‘critical slowing down’, usually estimated by an increasing lag-1 autocorrelation (AR(1)). The idea here is that as a system’s state becomes less stable, it will start to react more sluggishly to short term perturbations. While early warning signals have been tested extensively in simple models and on palaeoclimate data, there has been very little research into how these behave in complex models and observed data. Here, early warning signals are tested on climate systems that show tipping point behaviour in general circulation models. Furthermore, it examines why early warning signals might fail in certain cases and provides prospect for more ‘system specific indicators’ based on properties of individual tipping elements. The thesis also examines how slowing down in a system might affect ecosystems that are being driven by it.

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior
Intrusion Detection Systems (IDS) are designed to monitor the computer network infrastructure against possible attacks by generating security alerts. With the increase of components connected to computer networks, traditional IDS are not capable of effectively detecting malicious attacks. This occurs either by the distributed amount of data that traverses the network or the complexity of the attacks launched against the network. Therefore, the design of Internet Early Warning Systems (IEWS) enables the early detection of threats in the network, possibly avoiding eventual damages to the network resources. The IEWS works as a sink that collects alerts from different sources (for example, from different IDS), centralizing and correlating information in order to provide a holistic view of the network. This way, the current dissertation describes an IEWS architecture for correlating alerts from (geographically) spread out IDS using the Case-Based Reasoning (CBR) technique together with IP Georeferencing. The results obtained during experiments, which were executed over the implementation of the developed technique, showed the viability of the technique in reducing false-positives. This demonstrates the applicability of the proposal as the basis for developing advanced techniques inside the extended IEWS architecture.
Sistemas de Detecção de Instrução (Intrusion Detection Systems IDS) são projetados para monitorar possíveis ataques à infraestruturas da rede através da geração de alertas. Com a crescente quantidade de componentes conectados na rede, os IDS tradicionais não estão sendo suficientes para a efetiva detecção de ataques maliciosos, tanto pelo volume de dados como pela crescente complexidade de novos ataques. Nesse sentido, a construção de uma arquitetura Internet Early Warning Systems (IEWS) possibilita detectar precocemente as ameaças, antes de causar algum perigo para os recursos da rede. O IEWS funciona como um coletor de diferentes geradores de alertas, possivelmente IDS, centralizando e correlacionado informações afim de gerar uma visão holística da rede. Sendo assim, o trabalho tem como objetivo descrever uma arquitetura IEWS para a correlação de alertas gerados por IDS dispersos geograficamente utilizando a técnica Case-Based Reasoning (CBR) em conjunto com Georreferenciamento de endereços IP. Os resultados obtidos nos experimentos, realizados sobre a implementação da técnica desenvolvida, mostraram a viabilidade da técnica na redução de alertas classificados como falsos-positivos. Isso demonstra a aplicabilidade da proposta como base para o desenvolvimento de técnicas mais apuradas de detecção dentro da arquitetura de IEWS estendida.

Excessive droughts on the African continent have caused the Swedish Meteorological and Hydrological Institute to launch a program of gathering data in hopes of producing models for rainfalls and droughts. A sensor capable of gathering such data has already been chosen, however there remains the problem of conveniently retrieving data from each of the sensors spread over a large area of land. To accomplish this goal, a small, cheap and efficient wireless capable module would need to be used. A possible candidate is the new WiFi-module from Espress if designated ESP8266. It is an extremely cheap and versatile wireless SoC that is able to perform the task of a wireless communications adapter for the sensor unit. The point of this thesis is to investigate the suitability of IEEE 802.11 for the task, and produce a piece of firmware for the ESP8266. The firmware shall enable it to be attached to a sensor and operate as a wireless mesh node in a self-organizing WLAN sensor network, enabling data retrieval via WiFi multi-hop deliveries.

Purpose: The aim of this project was to evaluate data from a survey sent to nurses in regards to the current practices and workflow of the Early Warning System (EWS) currently in use at Banner Health. Methods: A descriptive cross-sectional design was used to collect feedback from Registered Nurses. The survey was open between February and March of 2016. Likert style and open-ended questions demonstrate evidence supporting the following study questions: 1) What are the barriers to documentation that triggers the EWS? 2) To what extent is the EWS useful and usable? 3) What were the features of the EWS implementation? Results: Findings demonstrated three barriers associated with EWS protocol compliance to include increased workload (78%), previous negative responses from providers (62%), and alert fatigue (48%). Provider responsiveness to notification of the Early Warning Score was shown to be effective "most of the time" and "about half of the time" at 71%, with 12% indicating "sometimes" and "never". Deployment of the Rapid Response Team (RRT) when indicated by EWS algorithm showed only 9 (14%) of nurses always call the RRT, 7 (11%) call about half the time, and 16% indicated they never use the RRT. "Real time" charting occurred 50-75% or less than 50% of the time for 96% of respondents. Open ended questions support recommendations for future practice to include: implementation of a pop up alert for easy recognition of changes in EWS, tailoring parameters based on individual patient characteristics, automatic direct paging to medical providers, and elimination of the level of consciousness parameter. A validated usability survey provided data with a mean response rate (n=58). Nurses (84%) agree the EWS is useful and usable. Ease of use, efficiency, and comfort with EWS software showed 90% agree. System interface responses demonstrate 23% dislike using the interface, and 21% felt the system interface was unpleasant.Conclusions: Findings demonstrate EWS system usability and usefulness. Recommendations for improvement include implementation of a "pop up" alert for easy recognition of changes in the Early Warning Score and/or automatic direct paging to medical providers and nursing will increase effective use. Barriers to EWS protocol documentation include increased workload, previous negative response from providers, and alert fatigue. "Real time" documentation of physiological parameters is essential to successful triggering of an Early Warning Score.

This thesis adopts a combined sociological and health services research approach to examining the implementation of standardised risk assessment tools, ‘early warning systems’, in medical wards. The data collection involved, over a three-year period from 2006 to 2008, ethnographic observations and 37 semi-structured interviews with staff in four UK hospitals that participated in the Health Foundation’s Safer Patients Initiative. Critical illness in hospitalised patients can be a predictable event preceded by observable physiological abnormalities, but research suggests that general wards may experience difficulty in detecting and responding to patient deterioration. As a result, growing numbers of acute hospitals are implementing early warning systems designed to detect and respond to early signs of patient deterioration. These systems involve track-and-trigger and rapid response mechanisms which seek to achieve accountability for standard risk management practices among doctors and nurses. The study found that accountability in relation to bedside observations was constituted through a combination of hierarchical accountability for fulfilling formal responsibilities, and horizontal accountability which encouraged sensible use of formal rules and responsiveness to calls for help and assistance. Although staff views on early warning systems were very positive, the findings also suggested that these systems may lead to undesirable practice and fail to manage certain aspects of risk. Problems identified with early warning systems included false reassurance, unnecessary alerts and ritualistic compliance, which could create unnecessary work and cause discomfort to patients. Among staff, reciprocal senses of obligation and responsibility helped to manage such problems, but could be obstructed by poor team work. The thesis suggests that focus on the alert system overshadowed accountability for the day-to-day management of early warning systems within teams. Managing the mundane may help both organisations and their staff to prevent and prepare for emergency situations, and reduce the fear of being implicated in poor management of risk.

A disease which is left untreated for a longer period is more likely to cause negative consequents for the patient. Even though the general practitioner is able to discover the disease quickly in most cases, there are patients who should have been discovered earlier. Electronic patient records store time-stamped health information about patients, recorded by the health personnel treating the patient. This makes it possible to do a retrospective analysis in order to determine whether there was sufficient information to give the diagnose earlier than the general practitioner actually did. Classification algorithms from the machine learning domain can utilise large collections of electronic patient records to build models which can predict whether a patient will get the disease or not. These models could be used to get more knowledge about these diseases and in a long-term perspective they could become a support for the general practitioner in daily practice. The purpose of this thesis is to design and implement a software system which can predict whether a patient will get a disease in the near future or not. The system should attempt to predict the disease before the general practitioner even suspects that the patient might have the disease. Further the objective is to use this system to identify warning signs which are used to make the predictions, and to analyse the usefulness of the predictions and the warning signs. The diseases asthma, diabetes 2 and hypothyroidism have been selected to be the test cases for our methodology. A set of suspicion-indicators which indicates that the general practitioner has suspected the disease are identified in an iterative process. These suspicion-indicators are subsequently used to limit the information available for the classification algorithms. This information is subsequently used to build prediction models, using different classification algoritms. The prediction models are evaluated in terms of various performance measures and the models themselves are analysed manually. Experiments are conducted in order to find favourable parameter values for the information extraction process. Because there are relatively few patients who have the disease test cases, the oversampling technique SMOTE is used to generate additional synthetical patients with the test cases. A set of suspicion-indicators has been identified in cooperation with domain experts. The availability of warning signs decreases as the information available for the classifier diminishes, while the performance of the classifiers is not affected to such a large degree. Applying the SMOTE oversampling technique improves the results for the prediction models. There is not much difference between the performance of the various classification algorithms. The improved problem formulation results in models which are more valid than before. A number of events which are used to predict the test cases have been identified, but their real-world importance remains to be evaluated by domain experts. The performance of the prediction models can be misguiding in terms of practical usefulness. SMOTE is a promising technique for generating additional data, but the evaluation techniques used here are not good enough to make any conclusions.

Nowadays, the trend towards globalization and internationalization of business has strong impact on companies’ strategy. When a company is thinking about strategy development, it needs to pay attention to variable aspects on micro- and macro-level. This study includes the overview of the factors which need to be considered while developing a strategy. Moreover, a significant emphasize is made on the contemporary deriving challenges at the market. The purpose of the study is to investigate a new approach to the strategic planning, named “Early warning system”. For that purpose, it was conducted qualitative research at the example of the Ukrainian telecommunication company – MTS Ukraine. I identified factors which influence the company and analyzed how company copes with deriving challenges. Moreover, it was found out the company’s strategic planning process is significantly influenced by specific market characteristics such as high level of bureaucracy, specific behavior of some competitors, growing market.

Thesis (M.S.)--Massachusetts Institute of Technology, Dept. of Civil and Environmental Engineering, 1995.
Includes bibliographical references (leaves 148-150).
by Achillefs Papadimitriou.
M.S.

Over one million high school students drop out of school each year in this country. Dropping out of school is a serious problem for the student, community, and the nation. Often dropouts are unable to compete in an increasingly technological society and face numerous consequences from their decision to leave school early including higher levels of poverty, unemployment, public assistance, incarceration, and poor health. Dropping out is a gradual process of school disengagement and related to individual, family, and school factors. In the past, it has been difficult to track individual student's progress through school and to determine accurate dropout and graduation rates. In 2005, the National Governors Association made a commitment to implement a uniform method to calculate and report graduates and dropouts as well as better data collections systems.This study intended to replicate aspects of other major studies around the county to determine the best early predictors of dropping out of school in this large school district in southern Arizona and use this information to build an early warning system. Student data were obtained from the district's Research and Accountability office for a cohort of students (n=6751) who began the ninth grade in fall 2006 and graduated or should have graduated in 2010. Data collected included general demographic information, academic data, number of schools attended, and school withdrawal codes.The intent of this research was to determine if there were statistically significant differences between dropouts and graduates in the variables collected and which variables yielded the highest effect sizes and should be included in the district's early warning system.Two analyses were used to determine significance differences between dropouts and graduates. Then four analyses were performed to determine the highest-yield variables for this district. Consistent with recent research in the field, the variables of ninth grade attendance, ninth grade English and Math grades, and GPA were the strongest predictors of student dropouts.Local educators can use this early warning information to help identify potential high school dropouts as early as possible and intervene more efficiently and effectively with these students.

Översvämningar förekommer regelbundet på stora delar av jordklotet. Utgående från de klimatförändringar vi ser idag med, bland annat, kraftigare och intensivare nederbörd är det troligt att det blir fler och större översvämningar framöver. Områden som inte tidigare varit översvämmade kan komma att bli översvämmande.   Varje land eller område med regelbundna översvämningar har någon form av varningssystem. Översikter över olika typer av system för att varna för översvämningar är svåra att finna.   Syftet med denna uppsats är att: Undersöka viktiga principer för effektiva varningssystem för översvämningar som orsakas av kraftig nederbörd och/eller vårflöden. Skapa en systematisk och strukturerad sammanställning över nuvarande system för att varna för översvämningar orsakade av kraftig nederbörd och/eller vårflöden. Ge praktiska exempel på några svenska kommuners system för att varna för översvämningar.   Litteratur- och internetsökningar kring varningssystem för översvämningar samt intervjuer med ”översvämningsansvariga” i Ödeshög och Vetlanda kommuner utgör grunden för informationen i denna uppsats.   Exempel på olika typer av varningssystem för översvämningar presenteras. Saker att tänka på vid val av, och vid drift och underhåll av, automatiska varningssystem belyses.   En indelning (Grust, 2006) av varningssystem i manuella, enkla automatiska och sofistikerade automatiska utökas till en matris med i matrisens andra dimension lokala, avrinningsområdes, nationella, multinationella och globala varningssystem. De i uppsatsen beskrivna exemplen på varningssystem positioneras in i matrisen.   Varningssystem för översvämningar i två svenska kommuner, Ödeshög och Vetlanda, beskrivs och likheter och skillnader mellan kommunernas system presenteras.
Flooding occurs all over the world on a regular basis. Based on the climate change ongoing today with heavier and more intense precipitation we can expect more and larger floods than we have seen before. Areas which have not been flooded earlier may now become flooded.   Every country or area which is flooded on a regular basis has some kind of warning system. Overviews of various types of flood warning systems are difficult to find.   The purpose of this thesis is to: Investigate important principles of effective flood warning systems caused by heavy precipitation and/or spring flood. Create a structured and systematical overview of current flood warning system for floods caused by heavy precipitation and/or spring flood. Describe two local Swedish community´s flood warning systems.   Literature and Internet search covering flood warning systems combined with interviews on site at Ödeshög and Vetlanda forms the basis for the information in this thesis.   Examples of various types of flood warning systems are presented. Issues regarding the selection of, and the operation and maintenance of, automatic warning systems are described.   A grouping (Grust, 2006) of warning systems in manual, simple automatic and sophisticated automatic is expanded to a matrix with the second dimension of the matrix being local, drainage area, national, multinational and global warning systems. The examples of warning systems covered in this thesis are placed in the matrix.   The similarities and differences between the flood warning systems in two local Swedish communities Ödeshög and Vetlanda are described.

The nursing culture of an inpatient pediatric unit was resistant to activating pediatric rapid response team (PRRT) alerts despite guidelines for activation. Nurses routinely assessed patients and assigned a pediatric early warning score (PEWS); however, the level of illness severity was not interpreted consistently among nurses and a PEWS action algorithm did not exist to guide nurses' minimal actions based on the PEWS score. Guided by 3 adult learning theories (Knowles, Kolb, and Bandura) and 1 evaluation model (Kirkpatrick), this staff education project sought to educate pediatric nurses on a PEWS action algorithm and determine whether this project improved nurses' knowledge, situational awareness, and attitude toward activating PRRT alerts. A convenience sample of 30 pediatric nurses completed a preeducation knowledge survey (EKS), attended an interactive PEWS education class, and completed a postEKS. After participating in the class, correct responses on the EKS increased from 43% to 82% and, using the Wilcoxon-signed rank test, a significant increase was noted in nurses' responses to questions related to self-efficacy, factual knowledge, and application. The overall increase in the nurses' self-efficacy and knowledge about the PEWS might enhance critical-thinking skills, foster identification of patients at risk for clinical deterioration, and empower nurses to follow the PEWS action algorithm including activation of PRRT alerts when indicated. This project has the potential to effect positive social change by supporting nurses' actions designed to improve pediatric patient outcomes.

Early Warning Systems (EWSs) for banking sectors are used to measure occurrence risks of banking crises, generally observed with a rundown of bank deposits and widespread failures of financial institutions. In countries with a small number of banks, for example Turkey with 48 banks (BDDK, 2011), every bank may be considered to have a systematic importance since the failure of any individual bank may carry a potential threat to lead to a banking crisis. Taking into account this fact the present study focuses on EWSs in Turkey. Since there is no single correct EWS to apply to all cases, in this study, 300 models were constructed and tested to find models as accurate as possible by using a trial-and-error process and by searching optimal feature subset or classifier methods. Empirical results indicate that prediction accuracy did not increase significantly while we got closer to the actual occurrence of bankruptcy. An important finding of the study was that trends of financial ratios were very useful in the prediction of bank failures. Instead of failures as a result of instant shocks, the banks'
failures followed through a path: first a downward movement affected the efficiency of the banks'
officers and the quality of management structure measured with "
Activity Ratios"
, then the profitability of the banks measured with "
Profit Ratios"
declined. At last, the performance and the stability of banks'
earnings stream measured with "
Income-Expenditure Structure Ratios"
and the level and quality of the banks'
capital base, the end line of defense, measured with "
Capital Ratios"
. At the end of study, we proposed an ensemble model which produced probability ratios for the success rates of the banks. The proposed model achieved a very high success rate for the banks we considered.

Background: Great strides have been made in combating malaria, however, the indicators in sub Saharan Africa still do not show promise for elimination in the near future as malaria infections still result in high morbidity and mortality among children. The abundance of the malaria-transmitting mosquito vectors in these regions are driven by climate suitability. In order to achieve malaria elimination by 2030, strengthening of surveillance systems have been advocated. Based on malaria surveillance and climate monitoring, forecasting models may be developed for early warnings. Therefore, in this thesis, we strived to illustrate the use malaria surveillance and climate data for policy and decision making by assessing the association between weather variability (from ground and remote sensing sources) and malaria mortality, and by building malaria admission forecasting models. We further propose an economic framework for integrating forecasts into operational surveillance system for evidence based decisionmaking and resource allocation.  Methods: The studies were based in Asembo, Gem and Karemo areas of the KEMRI/CDC Health and Demographic Surveillance System in Western Kenya. Lagged association of rainfall and temperature with malaria mortality was modeled using general additive models, while distributed lag non-linear models were used to explore relationship between remote sensing variables, land surface temperature(LST), normalized difference vegetation index(NDVI) and rainfall on weekly malaria mortality. General additive models, with and without boosting, were used to develop malaria admissions forecasting models for lead times one to three months. We developed a framework for incorporating forecast output into economic evaluation of response strategies at different lead times including uncertainties. The forecast output could either be an alert based on a threshold, or absolute predicted cases. In both situations, interventions at each lead time could be evaluated by the derived net benefit function and uncertainty incorporated by simulation.  Results: We found that the environmental factors correlated with malaria mortality with varying latencies. In the first paper, where we used ground weather data, the effect of mean temperature was significant from lag of 9 weeks, with risks higher for mean temperatures above 250C. The effect of cumulative precipitation was delayed and began from 5 weeks. Weekly total rainfall of more than 120 mm resulted in increased risk for mortality. In the second paper, using remotely sensed data, the effect of precipitation was consistent in the three areas, with increasing effect with weekly total rainfall of over 40 mm, and then declined at 80 mm of weekly rainfall. NDVI below 0.4 increased the risk of malaria mortality, while day LST above 350C increased the risk of malaria mortality with shorter lags for high LST weeks. The lag effect of precipitation was more delayed for precipitation values below 20 mm starting at week 5 while shorter lag effect for higher precipitation weeks. The effect of higher NDVI values above 0.4 were more delayed and protective while shorter lag effect for NDVI below 0.4. For all the lead times, in the malaria admissions forecasting modelling in the third paper, the boosted regression models provided better prediction accuracy. The economic framework in the fourth paper presented a probability function of the net benefit of response measures, where the best response at particular lead time corresponded to the one with the highest probability, and absolute value, of a net benefit surplus.  Conclusion: We have shown that lagged relationship between environmental variables and malaria health outcomes follow the expected biological mechanism, where presentation of cases follow the onset of specific weather conditions and climate variability. This relationship guided the development of predictive models showcased with the malaria admissions model. Further, we developed an economic framework connecting the forecasts to response measures in situations with considerable uncertainties. Thus, the thesis work has contributed to several important components of early warning systems including risk assessment; utilizing surveillance data for prediction; and a method to identifying cost-effective response strategies. We recommend economic evaluation becomes standard in implementation of early warning system to guide long-term sustainability of such health protection programs.

2014 - 2015
Landslide early warning systems are non-structural risk mitigation strategies aiming at dealing with intolerably high probabilities of landslide occurrence by reducing risk through the reduction of the exposed elements. The majority of landslide early warning systems deal with rainfall-induced landslides. The systems can be classified, as a function of the scale of analysis, into: “local” and “regional” systems. Several differences exists among these two different types of warning systems, such as: the actors involved in the process, the monitoring tools, the variables selected to define triggering thresholds, the way the warnings are issued and spread to the public. This work exclusively deals with regional landslide early warning systems (ReLEWSs). These systems are used to assess the probability of occurrence of landslides over appropriately-defined homogeneous alert zones of relevant extension, typically through the prediction and monitoring of meteorological variables, in order to give generalized warnings to administrators and the population. At first, a detailed review of the structure and the functioning of these systems is presented. The information has been gathered mainly from the literature, with the exception of the regional system operating in Campania region, Italy, the municipal system of Rio de Janeiro, Brazil, and the national Norwegian landslide early warning system. The functioning and the structure of the latter two systems have been analyzed in greater depth thanks to research periods spent, respectively, at the GEO-Rio foundation in Rio de Janeiro and at The Norwegian Water Resources and Energy Directorate (NVE) in Oslo. In literature, several authors provided a general description of the structure of a landslide early warning system. Starting from the analysis of these contributions, an original scheme and the main components of such systems for rainfall-induced landslides forecast is proposed. The scheme is based on a clear distinction among the following components: correlation laws, decisional algorithm and warning management. Subsequently, the functioning of the reviewed ReLEWSs has been described according to these components, with a special attention on how the performance of the various warning models was assessed. It is straightforward that a periodical assessment of the technical performance of a landslide early warning system, in terms of evaluation of the warning issued in relation to the landslides occurred, is a required task in order to continuously keep the system reliable. Nevertheless, no standard requirements exist for assessing the performance of regional warning models (ReWaMs) and, typically, this is evaluated by computing the joint frequency distribution of landslides and warnings, both considered as dichotomous variables. Herein, an original methodology to assess the performance of ReWaMs, called the “Event, Duration Matrix, Performance” (EDuMaP) method, is proposed. The performance is evaluated taking into account: the possible occurrence of multiple landslides in the warning zone; the duration of the warnings in relation to the time of occurrence of the landslides; the warning level issued in relation to the landslide spatial density in the warning zone; the relative importance system managers attribute to different types of errors. The applicability of EDuMaP method is tested considering three different ReLEWSs: the municipal early warning system operating in Rio de Janeiro (Brazil); the Norwegian landslide early warning system; the landslide early warning system for hydro-geological risk management of the Campania region, Italy. The main differences among these systems are discussed in great detail, mainly dealing with the functioning and the databases available for the three case studies. The LEWS operational in Rio de Janeiro is employed to issue a certain level of warning in four warning zones in which the municipality is divided. The warnings can be issued at any time during the day if the monitored rainfall exceeds pre-identified thresholds. The Norwegian landslide early warning system is employed to issue daily warnings adopting variable warning zones. In the LEWS of the Campania region each municipality has a reference rain gauge for which three different rainfall threshold are specified for the activation of 3 warning levels. The EDuMaP method was successfully employed to assess the performance for all these case studies, thus underlying the wide applicability of the method, which can be easily adopted to evaluate the performance of any regional landslide early warning systems for which landslides and warnings data are available. For the three case studies, sensitivity analyses are also conducted by varying some of the input parameters of the EDuMaP method. The results of these analyses indicate that the input parameters most affecting the performance of the warning models are: i) the landslide density criterion used to differentiate among the classes of landslide events; ii) the database on landslides considered in the simulations; iii) the time set xvii as the minimum time interval between landslide events; iv) the area of analysis; v) the time frame of the analysis. In conclusion, the analyses prove the applicability of the EDuMaP method in evaluating the performance of real case studies related to ReLWaMs characterized by different decisional algorithms, components and input parameters. The method can also be used as an effective tool to calibrate a warning model by back-analysing landslide and warning data in test area with the aim of defining the set of warning criteria which maximises the model performance. [edited by author]
XIV n.s.

A new zone was added to the existing Livestock Early Warning System (LEWS), which is a subproject of the USAID Global Livestock Collaborative Research Support Program. LEWS uses the PHYGROW model and satellite imagery of weather and vegetation to estimate the availability of forage to livestock and wildlife. Drought advisories are then distributed to governments, development organizations, and pastoralists via the Internet, satellite radios, and written reports. The Laikipia zone was established in 2001 to provide drought early warning for the arid pastoral rangelands of the Ewaso Ngiro ecosystem in the Laikipia and southern Samburu Districts, Kenya. Field verification of PHYGROW estimates of standing crop was conducted in 2002. In addition, research was conducted to determine the ability of the warning system to provide significant advance notice of emerging drought conditions. Results of this study indicate that LEWS is capable of providing accurate estimates of forage availability on East African rangelands. There is also evidence that the use of LEWS advisories could accelerate drought response by pastoralists as much as three to seven weeks.

By means of car2x communication technologies (car2x) driver warnings can be presented to drivers quite early. However, due to their early timing they could be misunderstood by drivers, distract or even disturb them. These problems arise if, at the moment of the warning, the safety–critical situation is not yet perceivable or critical. In order to examine, when drivers want to receive early warnings as a function of the situation criticality, a driving simulator study was conducted using the two early warning stages of a multi stage collision warning system (first stage: informing the driver; second stage: prewarning the driver). The optimum timing to activate these two early warning stages was derived by examining the drivers’ evaluation of these timings concerning their appropriateness and usefulness. As situational variation, drivers traveling at about 100 km/h were confronted with slow moving traffic either driving at 25 km/h or 50 km/h at the end of a rural road. In total, 24 participants were tested in a within-subjects design (12 female, 12 male; M = 26.6 years, SD = 7.2 years). For both stages, drivers preferred an earlier timing when approaching slow moving traffic traveling at 25 km/h (first stage: 447 m, second stage: 249 m ahead of the lead vehicle) compared to 50 km/h (first stage: 338 m, second stage: 186 m ahead of the lead vehicle). The drivers’ usefulness rating also varied with the timing, spanning a range of 8 s for driver-accepted timing variations and showed correspondence to the drivers’ appropriateness ratings. Based on these results and those of a previous study, a timing function for each of the two early warning stages depending on the speed difference between the safety–critical object and the host vehicle is presented. Indirectly, similar adaptations are already implemented in current collision warning systems, which use the time-to-collision to give drivers acute warnings in a later stage, when an immediate reaction of the driver may still prevent a collision. However, this study showed that drivers also favor this kind of adaptation for earlier warning stages (information and prewarning). Thus, adapting the timing according to the drivers’ preferences will contribute to a better acceptance of these collision warning systems.

One of the challenges that the petroleum industry faces is to ensure unimpeded flow of hydrocarbons. During production, transportation and processing, there can be free water in the produced fluid, and/or changes in temperature and pressure can lead to water condensation causing ice and/or hydrates formation. Gas hydrates pose serious flow assurance, economic and safety concerns. Chemical inhibitors are widely used to reduce the risks associated with hydrates. However, the upstream injection of hydrate inhibitors is generally based on thermodynamic model predictions and estimations of the worst conditions without much downstream measurements. This thesis presents a research work in which a number of techniques were investigated with the ultimate aim to mitigate hydrate risks during hydrocarbon recovery. Hydrate Monitoring Techniques are where the hydrate stability zone (HSZ) could be determined by testing downstream samples. Spectroscopy, dielectric permittivity, and freezing point depression methods were experimentally examined. A novel pseudo concentration approach was created as a result of this research work. This approach is more reliable and robust than the historically developed correlation, since it takes into account the pressure, hydrate structure and inhibitor type effects. Spectroscopy, dialectic permittivity set-ups and a freezing point prototype device based on Peltier heat pumps have been designed, built and tested. Hydrate Early Warning Techniques are where the hydrate formation could be detected based on water memory phenomenon. This phenomenon suggests that sampled fluid under specific conditions can carry remnant molecular structure related to hydrate formation if it had taken place. Spectroscopy, onset of ice formation and onset of hydrate formation were investigated. During this work, a multiple probe freezing apparatus and hydrate mini-rig prototypes have been designed, built, and tested. These techniques can provide technical measures for hydrate monitoring and early warning, helping to lessen the risk of pipeline blockages as well as to minimise the amount of chemicals required to inhibit any hydrate formation, hence improve the production economics and reduce the impact on the environment. Moreover, the investigated techniques show a potential to be deployed in Supervisory Control And Data Acquisition (SCADA) systems widely used in the petroleum industry for reservoir/production monitoring and management.

BACKGROUND: Late identification of deteriorating children undermines timely implementation of life-saving measures to prevent cardiopulmonary arrest (CPA) or death. The Pediatric Early Warning System (PEWS) has been validated for use in pediatric acute care settings for early identification of children at increased risk of physiologic deterioration, yet there is a dearth of evidence of the use of PEWS in primary care. Implementing the PEWS in primary care could guide rural primary care practitioners to early detection and prompt management of deteriorating children. This DNP project evaluated the attitudes and perceptions of rural practitioners towards the implementation of the PEWS scoring tool. METHODS: A cross-sectional descriptive design was conducted using an anonymous online survey via an email listserv. RESULTS: Seventeen practitioners responded to the survey, but only 14 participants met criteria for inclusion – 2 males and 11 females. The sex of one participant was not reported. Participants areas of specialization include 79% specialized in family practice, 79% pediatric specialists 14% and (7%) listed as "Other." Thirty-one percent of participants reported a travel distance of over 60 miles, while 39% reported a travel distance of over 60 miles lasting over 60 minutes via ground from a place of care to a hospital that specializes in the pediatric emergency care, and pediatric care respectively. Although 92% reported they have not heard of the PEWS tool prior to this survey, 54% strongly agree that the PEWS could help prevent cardiopulmonary arrest or death. Similarly, 54% of respondents reported they strongly agree that the PEWS can help identify deteriorating children, while 39% somewhat agree. Over 62% strongly agree that implementing the PEWS is appropriate in primary care, while 31% somewhat agree. Fifty-four percent of participants strongly agree they could use the PEWS tool in their practice. DISCUSSION: Participants have a positive view of the PEWS tool and perceive implementation of the PEWS to be a vital clinical decision support tool that could lead pediatric primary care providers to early detection of deteriorating children before the occurrence of an adverse event. Further study could determine the generalizability of implementing the PEWS in primary care.

This dissertation explores the potential contributions of the mechanisms for early warning and intelligence sharing to regional security in Africa. The Continental Early Warning System (CEWS) and the Committee on Intelligence and Security Services of Africa (CISSA) are centrally concerned with the dissemination of information to enable decision-making on continental security. The main focus of the dissertation is on the manner in which the information generated by the CEWS and CISSA can contribute to regional security. In order to analyse the potential contribution of the CEWS and CISSA to regional security, a sound theoretical framework is proposed so as to explore how and why states choose to cooperate, as well as addressing multifaceted cooperation and integration at inter-state, government department and nonstate levels. Constructivist interpretations of international cooperation are utilised to explore the role of ideas, meanings and understandings in shaping behaviour. The focus is placed on the manner in which interaction as provided for by the CEWS and CISSA can shape understandings of reality and potentially impact on the definition of actors&rsquo
interests. This is based on the assumption drawn from security community and epistemic community theory that, enabling the creation of shared meanings and shared knowledge there is the potential for both the CEWS and CISSA to have a positive influence on the choices that stakeholders take in favour of peaceful change.

A research report submitted to the Faculty of Humanities of the University of Witwatersrand in partial fulfilment of the requirements for the degree of Masters of Arts in International Relations, 2017
The international community's dynamics have undergone significant changes since 1990s. Famine, economic, security, and ethnic and religious animosities have greatly contributed to the emergence of conflicts globally. Preventive approaches, tools and structures in dealing with emerging conflicts rapidly became global trend. Early warning systems (EWS) were such preventive tools that bodies such as the African Union (AU) and the three regional economic communities (RECs) in Southern and West Africa and the Horn of Africa adopted to prevent conflicts on the African continent. [Abbreviated Abstract. Open document to view full version]
XL2018

碩士
國立臺北科技大學
電子工程系
106
This dissertation is a practical implementation of the wireless seismic system. The triaxial acceleration sensor to detect seismic waves. DC is removed through the AC capacitors. The temperature-dependent DC level is removed, and the signal is amplified to maximize the resolution of the analog-to-digital converter. The signal is converted to a digital signal by using a 16-bit analog to digital converter. The Raspberry Pi analyzes the data and uses the network to transfer to a computer server. After the Raspberry Pi analyzed the data, the Borland C++ Windows software was transmitted to the computer server using the Internet. The software can also be transmitted to the computer using Modbus RTU for data analysis and display of results. When the earthquake occurs, this design can detect the guided wave (P wave) before the earthquake, perform data analysis and achieve control of external circuits (turn off the gas and the power supply) to achieve disaster reduction effect.

碩士
國立臺灣海洋大學
河海工程學系
97
In this research, an early warning system was developed and proposed for drought management on the real-time reservoir operation. The system consists of three essential elements, namely, (a) drought watch, (b) water consumption measure, and (c) policy making. A Drought Alert Index (DAI) was used to characterize the alert level of drought severity. In addition, this study presents a risk-based decision model integrated into a drought early warning system (DEWS) for reservoir operation. Aspects of posterior risk, the corresponding options to given chances was provided to help decision makers to make better decisions. Decision makers can understand the inclination of attitude associated with any specific probability, and learn to adjust their attitudes in decision-making process. Besides, a user-friendly decision support system (DSS) for drought early warning system on DEWS has been introduced with an eye to practical use. As a pioneering experiment, the Tsengwen and Wushantou reservoirs in southern Taiwan were tested. The results show that the developed risk-based drought early warning system can really react to droughts occurred. The system can assist reservoir managers to work out applicable rule for real-time reservoir operation.

碩士
國立政治大學
行政管理碩士學程
98
Financial early warning system is a line both in the nature of financial management and operational evaluation of the dual function. To the financial crisis, With the role of prevention and warning system. The significance of means in accordance with relevant laws and regulations of the financial business and financial management principles. Certain number of selected variable set of a number of warning function, indication), cutoff or decimal value, discriminant model. According to the number of data, after testing with the accounting, it cause alarm or signal, so that the issue of the fail to meet the requirement, beyond the scope of the warning or to derail the number of abnormal conditions. To encourage the competent authorities or financial institutions (or banks) early attention to itself. By prevent and promptly correct and improve, to promote the sound management of the system. In recent years, financial institutions have become increasingly complex business and responsible for financial supervision authorities increasingly heavy responsibility, under the influence of the financial internationalization and liberalization. To solve this dilemma, how to make the best use of off-site monitoring tool, make up for lack of spot checks. It should be an effective program to strengthen the current financial supervision system. We all know that more and more attention to the financial supervisory authorities to the work of off-site monitoring. One of the most well-known and effective functioning is the “financial early warning system”. It could assess the performance of financial institutions, financial institutions and the issue of screening show that the functions of the police. For the financial supervision authorities, one of the important auxiliary tool, today. Keywords: financial early warning system, financial institutions, financial supervision

碩士
國立中興大學
土木工程學系所
101
Water is vital resource essential to life. Apart from droughts and floods struck in recent years, Taiwan has been affected by the other elements leading to unstable water supply such as industrial transformation and changes of water consumption structure. Unstable water supply would cause not only public panic but also social and economic losses. This study aims at developing scenario-based emergency planning for large-scale drought. Because drought is a chronic problem with slower progress than the other natural disasters, strict reservoir monitoring would be helpful in strengthening drought warning system. In this study, we use reservoir inflow and storage conditions as the basic parameters to strengthen warning mechanism, such as the current and history record of reservoir inflow, storage capacity, and water consumption plan. By calculation of the water balance, we estimate the storage capacity of reservoir within a certain period of time in the future. Furthermore, we also predict the possible changes resulting from different drought conditions as well as water supply and distribution strategies The research targets in this project are Feitsui Reservoir and Shimen Reservoir. The data from drought years are used to compare the operation of drought warning models in this study, and rule curve operation. Through this model we can find that, selected long times (90days) operation was too sensitive, caused nuisance. So we suggest that choosing the short times (60days) operation in this drought early warning system.

碩士
大同大學
工程管理碩士在職專班
104
DanShui River sewage systems (herein after referred to as the present system) large domestic market Basin counties of sewage systems, the plant stations are located in Taipei, New Taipei and Keelung, Taipei, including Taipei City and suburban sewage systems two parts, which covers the new Taipei City Luzhou、Sanchong、WuGu、XinZhuang、BanChiao、ShuLin、ZhongHe、YongHe、TuCheng、XinDian、ShenKeng、SanXia、TaiShan、XiZhi and BaLi other 15 districts and Keelung the QiDu、NuanNuan and AnLe other areas ，Taipei and suburbs excess sewage was collected together via facilities to transports.。 Such a huge and complex pipeline system, according to current experience shows that when rainfall events occur, even turned off all the sewage pumping station related upstream closure station, various sewage pumping station pumps are still much larger than the load condition when the sunny happen, it is necessary closure on stations and stations estimate the amount of water entering the dependency resolution and velocity of the system, the pumping station for the downstream flow of the estimated time as emergency decision making, to avoid the unexpected influx of large amounts of sewage, causing pumping stations flooded station and equipment damage and other disaster damage. In this paper, the current status of research, based on plant-based monitoring stations monitoring equipment in normal operation, the water level and flow rate data capture, calculate the measured flow value via Manning formula, but because of where the flow meter is located below the surface of sewage wastewater vulnerable covered or foreign body impact damage, resulting in distortion was measured flow data or zero, and can not use, so this research method is based on the current equipment is not damaged collected reliable water level and velocity values in order to obtain the flow rate value, and then based on the water level and flow to find the relationship is more than 0.6 above data regression analysis to find the right formula of linear regression, future until after the flow meter used for some time, often due to failure or distortion, the level of information which can be substituted into the regression equation to strike an appropriate the flow rate value to the correction period is missing or distorted information. At present, the research results have been applied to the monitoring station when the normal period of time, measured value and the regression equation calculation value is not much difference, but when the system is abnormal, the measured value is zero or negative, the regression equation calculation value remains stable the precise effect, then the system regression formula is not static, still need regular monitoring every year to the years when the instrument repair and maintenance, accounting together again this regression formula, just as the period of time after the test instrument calibration should be performed is the same reason。

碩士
國立臺灣海洋大學
河海工程學系
95
In this research, a color-coded early warning system is developed and proposed for drought management on the real-time reservoir operation. The system consists of three essential elements, namely, (a) drought watch, (b) water consumption measure, and (c) drought alert analysis. A new Drought Alert Index is used to characterize the alert level of drought severity. For demonstration, the drought warning procedures were effectively applied to the Ming-Te reservoir. The results show that the developed drought watch can really react to droughts occurred. The implementation of such a system proves that the decision support-like system can help the water authorities concerned take a timely action while confronting drought threats. Keywords: Drought alert index,Drought watch, Early warning system.

碩士
臺灣大學
財務金融學研究所
98
This paper aims to build a more predictable financial early warning system for Taiwan companies with Binary Logit model. We integrate traditional Expert system and Logit model, using five categories of variables including accounting variables, market variables, corporate governance variables, accountant variables, and macroeconomics variables, to build an integrating Logit model which conforms to expert system’s spirit that uses lots of information and does not lack in objectivity and theoretical background. We examine the one year later predict accuracy of corporate financial distress, and the changes in predict ability when continually adding each variable. We also examine the prior studies variables’ significance of affecting the predict ability. We find that after considering the five categories of variables, the expert system integrating model’s in-sample predict accuracy is 90.9%, slightly higher than accounting model’s 90.0%; and out-sample predict accuracy is 95.6%, also higher than accounting model’s 93.4%, especially the predict accuracy in financial distress companies growing from 65.9% to 84.4%. We conclude that all the five categories of variables have significant effect to financial distress prediction, and accounting variables have the best explanatory power, market variables are the second best one, but the other variables’ explanatory power are much weaker. In predict ability, we can get 90% predict accuracy when only using accounting variables, showing that accounting variables are still the most important factor. We also can improve the predict accuracy and explanatory power when adding other variables, and the market variables and macroeconomic variables especially have the higher contribution.

碩士
逢甲大學
資訊電機工程碩士在職專班
99
In Taiwan, debris flow disaster have been increasing in recent years due to various factors, such as complex earthquake faults, geographic layers of shattered rocks, high and steep mountains, concentrated rainfall, plus the overdevelopment of hillsides that result in soil erosion and water loss. Debris flows often bring a lot of rocks and mud causing a threat to the property and even the safety of the residents in the affected areas. With more and more extreme rainfall events as a result of climate change, the debris flow warning system is an effective notification mechanism for early-warning the disaster. However, the current system collects precipitation through Central Weather Bureau, which number of precipitation stations is not enough and rainfall situation is greatly affected by terrain cause the problems of imbalanced rainfall distribution. Therefore, the system can’t collect accurate rainfall data and cause accuracy of the system is not enough. This thesis proposes a watershed-based debris flow early-warning system which applies OGC SWE standard. The system uses the standardized data exchange mechanism to integrate and share heterogeneous monitor resources. In this way, the system can high density monitor the debris-flow-prone area. In addition, this thesis introduces a watershed-based sensing model and a rainfall warning decision method for the requirement of automatic debris flow early warning. Those mechanisms enable the system to collect sensing data under dependencies relationship and issue warning messages automatically. Moreover, SWE software provided by 52North is used to implement the proposed watershed-based debris flow early-warning system prototype. Finally, we use the historical rainfall data to compare and analyze the proposed rainfall warning decision method. According to experimental results, the proposed system can collect sensing data more accurate and advance the time of issue warning messages efficiently.

碩士
銘傳大學
經濟學系碩士班
99
During the period of a long market, there always can have easy targets for investment in the stock market. Once the optimistic mood emerges, irrational investors will certainly push the stock market to an even higher level. However, if co-movements of different industries occur and pass a threshold level, then the whole stock market can fall unexpectedly more than 200 points in the Taipei stock market in one day. By analyzing the WINDEX of the market and every individual industrial index,this study tries to reveal an early warning signal one day before a more than 200 point decline will happen. Basically, the phase transition of up-turn into down-turn co-movements of a majority of industrial indexes will be examined to serve as the early warning signal.

碩士
國立臺灣大學
土木工程學研究所
99
Taiwan is the island which is well known for its distinctive topography features. The central mountain range bisects the land from north to south resulting in most rivers have short courses and rapid streams. As a result, the scour effect could be seem as a critical problem for the bridge structure as the scale of natural disasters increase. The growing amount of bridge failure events in Taiwan due to typhoons and floods in recent years causing the public pay more and more attention towards the bridge safety under natural disaster, and the government and related organization also put a lot of efforts on such problem. This dissertation applies five kinds of sensing units which is based on the same sensing mechanism of developed scour monitoring system to laboratory experiments. Sensing units includes micro-electro-mechanical systems based pressure sensors, vibration-based detection sensors, floating ball, inner camera and handheld camera. The main purpose of this research is to verify the feasibility of developed monitoring system and to establish a warning benchmark by conducting the various test using reduced scale bridge structure model in laboratory. The feasibility of the developed monitoring system has been verified in laboratory by several experiments. The scour depth measurement by using accelerometers and the floating ball have been successfully applied to experiments, and it could be observed and confirmed that the scour depth can be regarded as the major parameter for the safety evaluation of caisson foundation bridge. Modes of scour failure for bridge structure are complex with many uncertainties, but the proposed sensing mechanisms for scour early warning could be achieved.

碩士
國立虎尾科技大學
經營管理研究所
98
Since 1980, Taiwan’s government relaxes restrictions of banks gradually, new banks establish induce the competition intense in bank-industry. And government strengthened banking supervision and practiced financial reform bills actively. Therefore, banks to face the sudden changes in the financial environment that management of risk is an important issue. This article using financial information among 33 domestic banks from first season in 2004 to the fourth season in 2008, using CAMELS ratings to assess the bank operating performance. The early warning model using Logit regression analysis to set up the non-risk model and risk model. That can be reference for financial inspection units and related financial institutions. The Empirical research results indicate as follows： (1) Select by factor analysis variables that consistent with CAMELS (Capital Adequacy、Asset Quality、Management、Earning、Liquidity、Sensitivity to market risk) that means this article can objectively evaluate the bank performance. Hence, that can reference for the relevant institutions and investors. (2) Use Logit regression analysis set up original sample models that prediction ability are：non-risk model A is 72.73% and risk model B is 78.79%. Predictive sample models of prediction ability：non-risk model A is 90.91% and risk model B is 96.97%. (3) The CAMELS of non-risk models has good predictive ability, the model add risk indicators that predictive ability and type I error work better than non-risk models.

碩士
淡江大學
國際貿易學系碩士在職專班
91
This study focuses on the building of risk warning system for the purposes of facilitating consumer banks’ credit management procedure. Using data consumers’ socioeconomic as well as their previous records concerning credit worthiness, two multivariate analysis method, the canonical discriminant analysis and the logistic discriminant analysis, are applied to construct a formula that can distinguish a good borrowers from the bad one, and that can be used for evaluating potential loan customers to effectively shorten the evaluation procedures. Samples of size 3,036 customers were drawn from the database of a specific bank in Taiwan. The sample was divided into two groups: one that consists of 2,581 (85%) observations, were used for model estimation; and the remaining 455 (15%) were acting as the test sample for model evaluation purpose. The results from the test sample simulations have shown that estimation based on the logistic discriminant analysis performs a little bit better than those conducted by the canonical discriminant analysis. However, the former was not able to discriminate the middle and the higher risk groups.

碩士
國立臺北科技大學
土木與防災研究所
99
Intelligent building system has the central management and control device functions, including remote monitoring, operation, control, recording and management. By building an integrated safety system platform, access control, surveillance, air conditioning, lighting, fire, electrical and other equipment used to control the management, operation and maintenance of monitoring equipment operation and maintenance records, in order to achieve safety, improve efficiency and save staff costs and energy conservation management purposes. In this study, design and development of intelligent building systems research, the first intelligent building systems in the current structure and function, a second analysis of the applicability of seismic instant alerts, and to explore the global business intelligent building system, building systems and functions framework, the survey found that the development of various business systems to an open integration platform, and IP-based architecture, Web browsers and wireless networking module. Then, by the survey result of the business intelligent building and the functional requirements, developing and designing the new intelligent building systems and a new functional module of earthquake early warning, choosing an existing building into the intelligent building systems and the function of earthquake early warning to create a complete intelligent building systems and earthquake early warning function. Intelligent building systems and building almost merged into one, commonly used in construction management, earthquake early warning proposed in the current building with intelligent building systems commonly used in the field of earthquake disaster prevention in order to reduce the losses caused by disasters.

碩士
國立臺灣海洋大學
河海工程學系
97
In this research, an early warning system was developed and proposed for drought management on the real-time reservoir operation. The system consists of three essential elements, namely, (a) drought watch, (b) water consumption measure, and (c) policy making. A Drought Alert Index (DAI) was used to characterize the alert level of drought severity. In addition, this study presents a risk-based decision model integrated into a drought early warning system (DEWS) for reservoir operation. Aspects of posterior risk, the corresponding options to given chances was provided to help decision makers to make better decisions. Decision makers can understand the inclination of attitude associated with any specific probability, and learn to adjust their attitudes in decision-making process. Besides, a user-friendly decision support system (DSS) for drought early warning system on DEWS has been introduced with an eye to practical use. As a pioneering experiment, Keelung Area was tested. The results show that the developed risk-based drought early warning system can really react to droughts occurred. The system can assist reservoir managers to work out applicable rule for real-time reservoir operation.

碩士
國立臺灣大學
國際企業學研究所
97
The main purpose of this thesis is to provide a measure of the probability of crises related to a aggregate index. Firstly we set a market index pressure to measure when a nation experienced currency crisis. Secondly, given a group of economic variables, we use Signal Approach suggested by Berg and Patillo (1999)、Goldstein (2000)、Kamisky and Reinhert (1998)、Edison to find which variables can have signals before the real crises happen. Finally, these workable variables would be used to set up an early warning system to predict the possibility of crises. Our warning system includes twelve nations: Argentina, Chile, Czech, Estonia, Hungary, Israel, Latvia, Lithuania, Poland, Romania, Slovakia, Turkey and also ten economic variables.

碩士
華梵大學
資訊管理學系碩士班
95
Financial Early warning system can not only help the management of the financial institutions but also diagnose their common operations. Since the early 1970s, many related researches have already made. However, most of them use traditional statistic ways to build the early warning system until recent years. Because of the vigorous development of the data mining techniques, many researches begin to apply those techniques to various fields also including early warning system. Data mining doesn’t need to satisfy many statistical antecedent assumptions and can transform enormous original data into meaningful and useful information. To build the early warning system model, the related financial laws, data, and operation management rules need to be taken into consideration. However, the number of features is too large and not all of them are helpful to prediction. Data sets with unimportant, noisy or high correlated features will significant decrease the classification accuracy rate. By removing these features, the efficiency and accuracy rate can obtain a better result. Back-propagation neural network (BPN), support vector machine (SVM) and decision tree (DT) are well-known data mining techniques, which can be applied to various fields and have higher classification ability. However, data mining techniques may suffer the problem of parameters settings. Bad parameter setting of data mining techniques will result worse accuracy rate. Therefore, this paper utilize one meta-heuristic, particle swarm optimization (PSO), to obtain suitable parameter optimization and select a subset of feature without degrade the classification accuracy rate. By the meta-heuristic global search characteristic, the parameters of BPN, SVM and DT can be optimized and the feature selection can be done at the same time to obtain the minimum set of features which can result in higher accuracy effectively. In order to evaluate the proposed approach, this research taken the report of the Taiwan Ratings to be the authority. The “Condition and Performance of Domestic Banks” from the Central bank of China, Republic of China (Taiwan) and the “Statistics of Financial Institutions” from the Financial Supervisory commission, Executive Yuan are planed to be the source data. Banks will be classified as one of three categories ( ”well”, ”average”, and ”risky”). In the experiment, although BPN and SVM have the high accuracy of forecast, the processes among them are black-box testing. Professionals can’t take these results into their future judgments. By the tree structure which was obtained from the proposed PSO+DT architecture, experts can obtain the best decision rules and thus make further evaluation and correction of our early warning system model. The experiment results shown that our proposed approaches can reduce unnecessarily features and improve classification accuracy significantly.

碩士
國立中正大學
會計學研究所
92
The traditional method of field supervision has become inadequate due to the continuing emergence of new banks, the openness of various banking business and the short of human resource for financial supervision. To effectively utilize the resource of financial supervision, there is a need for revolution on financial supervision method. A early banking warning system facilitates financial supervision working mainly in that it screens banks and list those to note with priority, so as to utilize the financial supervision effectively and find out problematic banks in time. This research, in the base of the CAMELS Rating System proposed by the FFIEC, studies the government-required report data from banks and the related codes to suggest the index of evaluation for early banking warning system. The researcher used the Web Service to obtain data needed, which can obtain data from the target banks anywhere and anytime. The researcher then, according the second version of the XBRL standard and with the technique of XPath, established a standard taxonomy package of early banking warning system, edited the reports of XBRL early banking warning system automatically with the practice of early banking warning analysis prototype, and collected all the reports from the target banks. The method of percentile ranking was used to elicit the risky or unqualified banks to cross-validate the taxonomy standard established

碩士
國立臺灣大學
財務金融學系
86
In the trend of liberalization in banking industry, sixteen newly-established banks were permitted to operate after 1991 and the operation risk were increasing. The purpose of this thesis is to focus on the newly-established and long existing banks to analyze their financial soundness and safety. Logit and Probit are used to build up an early warning system for Taiwan Financial Institutions and I also compare these two models. The sample data starts from 1992 to 1997. Some conclusions are drawn from my empirical study: (1) The newly-established banks, which are grouped into the worse-side, perform worse than old banks. (2) Probit model works better than Logit model in terms of the correct judgment, Type I Error, and Type II Error. (3) The number of banks on the worse-side increases from 1992 to 1997, which reflects the increasing risk on bank industry. (4) After Factor analysis, the remaining financial variables, which are still consistent with CAMEL model, accumulate enough power to fit the early warning system.

碩士
國立雲林科技大學
機械工程系
103
Firstly, this study fluid mechanics equations to establish mathematical model of pressure loss and capillary air sampling network and derive the mathematical model of the flow orifice, In order to make the results very early warning systems scada systems and reduce the actual experimental error, so we have to do the actual experiment, the results we got with Excel to establish a series of related samples and best wind speed of the capillary tube, then these best Winds database connected to the pattern control program allows pattern control program to achieve a more accurate calculation of output accuracy and more appropriate status, again derived using a mathematical model and then create a Execl as a pre-production step of writing diagram programming, Execl made available to verify that the design process, the program is correct at the time of writing code. Finally, these mathematical models can be pre-set with InduSoft write to know the actual detection of air information scada program that has allowed the engineer or operator knows the actual installation information to know, but cut the correct installation, improve warning effect. Keywords: very early warning systems、hydrodynamics、InduSoft、air scada program information

碩士
華夏技術學院
資產與物業管理研究所
100
Owing to the global climate changes in recent years, torrential rains always flood to cause serious disasters. For instance, 3,004.5 mm, 1,861 mm and 1,105,5mm rainfalls at Alishan, Chiayi (called as 88 Flood), Datong, Yilan and Manchou, Pingtong were caused by Typhoon Morakot, Magi and Nanmadol in 2009, 2010 and 2011; 610 Flood in 2012, followed by Typhoon Talim and Tembin, caused huge images. If such rainfall takes place in Taipei area, it will cause flood. Furthermore, there are a lot of high buildings in Taipei. The flood will cause the vehicles and electrical equipment on the basement as well as the public’s properties to seriously damage. Therefore, we study the installation of the Intelligence Early Warning System shortly before flooding, which automatically sends a warning message via mobile phone to tell owners and securities while the accumulated rainfalls and the ponding level reach the warning value, so that can immediately evacuate to reduce flood losses. The purpose of this study is to utilize the digital rain gauge and digital level gauge during the typhoon or torrential rain to monitor the rainfalls and the ponding depth of ground surface. With the warning rainfall from Disaster Prevention Information Website of Water Resources Agency, the Ministry of Economic Affairs and using the hydraulic analyses to investigate the characteristics of the regional drainage system, the warning rainfall and ponding level can be calculated. The Intelligence Early Warning System will automatically message owners and securities by PLC to acquire the data from the digital rainfall gauge and digital level gauge when reaching the warning value to evacuate immediately, so that can reduce property loss and avoid casualties. This study is to install the Intelligence Early Warning System at Chonghe as an example, follow the warning rainfall at the first and second grade from Disaster Prevention Information Website of Water Resources Agency, the Ministry of Economic Affairs, make use of reasonable formula to evaluate the flood flow for resident district and assume the community use two-year rainfall strength formula to design the drainage during the reconstruction from Construction and Planning Agency, Ministry of the Interior, to calculate the ponding depth caused by the warning flow corresponding to the warning rainfall at the first and second grade as the warning threshold under “normal drainage” and “completely no drainage”. This study tests the system when modeling the sand box and evaluates at Chonghe. Since rainfall is extended within one hour, the system will issue the second-grade warning when the ponding depth reaches 31.63 cm; when the ponding depth reaches 47.44 cm, the system will issue the first-grade warning. Since rainfall is extended within three hours, the system will issue the second-grade warning when the ponding depth reaches 63.25 cm; when the ponding depth reaches 86.97 cm, the system will issue the first-grade warning. Since rainfall is extended within six hours, the system will issue the second-grade warning when the ponding depth reaches 94.88 cm; when the ponding depth reaches 118.59 cm, the system will issue the first-grade warning. Since rainfall is extended within 12 hours, the system will issue the second-grade warning when the ponding depth reaches 126.50 cm; when the ponding depth reaches 158.13 cm, the system will issue the first-grade warning. Since rainfall is extended within 24 hours, the system will issue the second-grade warning when the ponding depth reaches 158.13 cm; when the ponding depth reaches 197.66 cm, the system will issue the first-grade warning. The Intelligence Early Warning System in this study not only monitors the rainfalls and the ponding depth immediately, but also automatically messages and proposes the early warning to evacuate so that can avoid property losses and make concrete contribution.

碩士
國立臺灣科技大學
電機工程系
106
The purpose of this thesis is to study and develop an early warning system of rolling blackout for Central or local Emergency Response Center and disturbed users. Taiwan government actively promotes “2025 non-nuclear homeland”, “green energy” and “sustainable energy” policies, coupled with the difficulty of new power plant erection and the resistance of environmental protection group, the system operating reserve deficient is not rare, and the risk of power shortage is greatly increased. On August 15, 2017, all 6 units of the Da-Tan Power Plant were tripped simultaneously due to the sudden interruption of natural gas supply. The unexpected event results in severe power blackout around entire Taiwan and affects the power supply of 5.92 million users. Nowadays, Taiwan Power Company still deficiencies in early warning mechanism for rolling blackout. It causes inconvenience, anxiety and economic loss to the local enterprise and the public and influences the people's livelihood as well. If there is an early warning mechanism, economic loss may be significantly reduced and the public's nervousness and hesitations will also be alleviated. In this thesis research, Visual C# is applied to develop an early warning system for Taiwan Power Company to provide advantageous information to Central or local government response center and affected users for emergency reaction . The proposed early warning system combined with geographic data can, therefore, provide following valuable information: (i) power blackout facts of traffic sign system, (ii) power blackout data for hospitals, schools, and public, and (iii) advanced power blackout notices to the households by mobile phone messages.