Providing automated support for model selection is a significant research challenge in model management. Organizations maintain vast growing repositories of analytical models, typically in the form of spreadsheets. Effective reuse of these models could result in significant cost savings and improvements in productivity. However, in practice, model reuse is severely limited by two main challenges: (1) lack of relevant information about the models maintained in the repository, and (2) lack of end user knowledge that prevents them from selecting appropriate models for a given problem solving task. This study built on the existing model management literature to address these research challenges. First, this research captured the relevant meta-information about the models. Next, it identified the features based on which models are selected. Finally, it used Analytic Hierarchy Process (AHP) to select the most appropriate model for any specified problem. AHP is an established method for multi-criteria decision-making that is suitable for the model selection task. To evaluate the proposed method for automated model selection, this study developed a simulated prototype system that implemented this method and tested it in two realistic end-user model selection scenarios based on previously benchmarked test problems.

In this thesis, methods for using computer-based models as support tools for assessing Transport Telematic Services (TTSs) are studied. Such assessments provide one way to understand how TTSs can address problems caused by transportation, such as accidents, emissions, and energy consumption. TTSs are services based on telematic systems which are Intelligent Transport Systems (ITS) involving the integrated use of information and communication technologies in transport. The focus is on TTSs that are relevant for road freight transport, even though the suggested methods can easily be adapted for TTSs in other areas. We characterize TTSs, e.g., in terms of their functionalities, and apply computer-based modeling for pre-deployment assessment of various TTSs (from an ex-ante perspective). By analyzing information provided by the suggested computer-based models, it is possible to make an informed decision whether to (or not to) deploy a given TTS. A review of previous studies reveals information about relevant TTSs for freight transport in areas such as driver support, administration, safety, traffic management, parking, and goods handling. A hierarchical clustering algorithm and a k-minimum spanning tree algorithm were employed to analyze synergies of TTSs. Synergies can enable identification of sets of TTSs that can lead to cost savings if deployed on a common platform (cf. Multi-Service Architectures). An analytical model inspired by the net present value concept is used to estimate quantified societal benefits of TTSs. An optimization model is formulated and solved using a branch and bound method to determine an optimal combination of TTSs taking into consideration societal benefits, costs, dependencies, and synergies. The optimization model also addresses possible system architectures for achieving multiple TTSs. Dominance rough set approach is used to assess and compare benefit areas for TTSs specific to truck parking. The benefit areas are suggested with the help of conceptual modeling, which describes functional models of a system in terms of states, transitions among states, and actions performed in states. The main scientific contributions of the thesis are in suggesting new quantitative models, extending and applying existing models in the assessments of TTSs, and obtaining results that can help decision-makers select TTSs for medium-to long-term investments. Researchers can employ and build on the proposed methods when addressing different scenarios (geographic or organizational) involving similar TTSs. By studying a range of TTSs and possible Multi-Service Architecture concepts for such TTSs, the thesis contributes to achieving convergence of TTSs in a Multi-Service Architecture environment that will improve cost efficiency, minimize redundancies, and encourage the establishment of standards in the deployment of TTSs in road freight transport. TTSs implemented in such an environment can contribute to optimizing available capacity, accuracy, speed, and efficiency of road freight transport systems.

A importância de um bom sistema de medição de desempenho (SMD) para o sucesso de uma empresa é uma questão bastante estudada e discutida, tanto no meio científico quanto no próprio meio empresarial. No entanto, grande parte das empresas não possui um sistema de medição de desempenho que possa lhes trazer uma vantagem competitiva, sendo ele muitas vezes pontual e desligado da estratégia. Por isso, este trabalho é mais uma tentativa de aproximar a teoria e a prática dentro das empresas no que se refere à medição de desempenho organizacional. O objetivo da pesquisa é desenvolver um sistema informatizado para a gestão do desempenho e apresentar um método para sua implantação. Para a definição das características deste método e dos requisitos do sistema, empreendeu-se uma revisão bibliográfica e um estudo de caso em uma empresa de grande porte que já possui um SMD formalizado. O método foi aplicado em duas empresas de Sertãozinho para identificar seus objetivos estratégicos e indicadores de desempenho, o que resultou em uma primeira versão de um SMD formal para elas. Uma aplicação piloto do software foi feita em uma destas empresas para verificar seu funcionamento na prática. Após estes estudos, pôde-se concluir que o método e o sistema são úteis para implantar e acompanhar indicadores de desempenho nas empresas pesquisadas. No entanto, a implantação encontrou alguns obstáculos como a falta de tempo das empresas para se dedicarem à medição de desempenho e a resistência de alguns funcionários em usar o software. / The importance of a well-developed performance measurement system (PMS) for the success of a company is a subject which is widely studied and discussed either in the academic and business enviroments. However, many companies still do not have a performance measurement system able to bring them competitive advantage, because they are all too often ad-hoc solutions not concerned about the strategy. Hence, this work is yet another attempt to bring theory closer to practice inside the companies in the regard of business performance measurement. This research\'s objective is to develop an information system for performance measurement and present a method for its deployment. In order to define the characteristics of both the method and the information system, a literature review and a case study in a company which had a formal PMS were done. The method was then applied to two companies from Sertãozinho aiming to identify their strategic objectives and performance indicators, which resulted in a first version of their formal PMS. A pilot deployment of the software was made in one of these companies to verify how it would work in practice. After these studies, it was possible to conclude that the method and the information system were useful for deploying and keeping track of performance indicators in the researched companies. However, some dificulties were found during the deployment, such as the lack of time available for performance management and the resistance of some employees to the use of the software.

Introdução: O Sistema Manchester de Classificação de Risco (SMCR) estabelece um tempo para o atendimento médico conforme o grau de urgência dos pacientes, a partir de um processo de tomada de decisão por registro eletrônico ou manual. Objetivo: Avaliar o grau de confiabilidade, a acurácia e o tempo despendido para a realização do SMCR em registros eletrônico e manual. Método: Estudo exploratório-descritivo de abordagem quantitativa, realizado com todos 43 (100%) enfermeiros do Hospital Universitário da Universidade de São Paulo (HU-USP) que foram aprovados no curso de classificador do SMCR pelo Grupo Brasileiro de Classificação de Risco (GBCR). A casuística do estudo correspondeu ao número total de casos clínicos simulados (37) validados e fornecidos pelo GBCR e ao número de casos aplicados para cada participante da pesquisa (4). A amostra foi estimada, para que se obtivesse um coeficiente kappa de Cohen maior ou igual a 0,5, confiança de 95% e poder de teste de 80%, considerando o número de pacientes classificados em cada nível de prioridade clínica no HU-USP em 2016. A amostra foi composta por 10 participantes. A coleta de dados foi realizada, em duas fases, utilizando 20 casos clínicos simulados, os quais passaram por processo de avaliação junto a especialistas do GBCR. Na fase 1, foram entregues 4 casos para uso no registro manual e na fase 2, após aproximadamente, 4 semanas, os mesmos casos foram entregues para realização do SMCR por meio do registro eletrônico. A confiabilidade inter-avaliadores foi calculada pelo coeficiente kappa de Cohen e, em relação ao padrão ouro (acurácia), pela porcentagem de concordância. O tempo despendido foi analisado com o teste Wilcoxon-Mann-Whitney, considerando intervalo de confiança de 95% e valor de p menor que 0,05. Resultados: O grau de confiabilidade foi igual para a escolha dos fluxogramas e dos discriminadores com o uso de ambos os registros e apresentou diferenças na determinação da prioridade e dos sinais vitais. A acurácia apresentou diferença estatisticamente significante, apenas, em relação aos sinais vitais. O tempo despendido para o registro do SMCR foi menor com o uso eletrônico. Conclusão: O uso do registro eletrônico apresenta vantagens referentes à confiabilidade, acurácia e tempo despendido para a realização da classificação de risco, indicando a importância da adoção de tecnologias no processo de trabalho gerencial e assistencial nos serviços de saúde. / Introduction: The Manchester Triage System (MTS) defines clinical priority by determining the maximum allowed waiting time for medical care for the different levels of urgency based on decision support process by electronic or manual registration. Objective: Evaluate the degree of reliability, accuracy and time spent of the use of MTS with an electronic or manual registration. Method: This is an exploratory-descriptive study applied for all (43) nurses of the University Hospital of the University of São Paulo (HU-USP) approved in the MTS classifier course by Brazilian Group of Risk Classification. The casuistry of the study corresponded to the total number of simulated clinical cases (37) and the number of cases applied to each participant (4). The sample were calculated to obtain a Cohen kappa coefficient equal to 0.5 or greater than it, 95% confidence and 80% test power, considering the number of patients classified in each clinical priority level at the HU-USP in 2016. The sample had 10 participants. Data were collected from the triage of 20 clinical cases. This clinical cases were evaluated by Brazilian Group experts. In phase 1, 4 cases were given for use in the manual registration. In phase 2, after approximately 4 weeks, the same cases were given for use in the electronic registration. The inter-rater reliability was calculated by the Cohen kappa coefficient. The accuracy (the comparison with the golden standard) was calculated by the concordance percentage. The efficiency was calculated by the Wilcoxon-Mann-Whitney test, with a 95% confidence interval and p-value less than 0.05. Results: The degree of reliability was the same for the choice of flowcharts and discriminators with the use of electronic and manual registration. There were differences in reliability for determination of priority and the recording of vital signs. The accuracy presented a statistically significant difference only in relation to vital signs. The time spent with use the MTS was shorter with electronic registration. Conclusion: The use of the electronic registration had advantages related to the reliability, accuracy and time spent to use the MTS, showing how important it is to adopt technologies in the manage and assistencial care working process of the health services.

Cardiac auscultation, an important part of the physical examination, is difficult for many primary care providers. As a result, diagnoses are missed or auscultatory signs misinterpreted. A reliable, automated means of interpreting cardiac auscultation should be of benefit to both the primary care provider and to patients. This paper explores a novel approach to this problem and develops an algorithm that can be expanded to include all the necessary electronics and programming to develop such a device. The algorithm is explained and its shortcomings exposed. The potential for further development is also expounded. / Includes bibliography. / Thesis (M.S.)--Florida Atlantic University, 2016. / FAU Electronic Theses and Dissertations Collection

Personalized advertising represents an emerging trend in online advertising. Using enhanced data collection techniques, marketers can craft seemingly made to order advertisements tailored to specific individuals. In turn, this should lead to advertisements that are more relevant for consumers and more effective for marketers. Therefore, personalized advertising has the potential to benefit both consumers and firms alike. However, consumer acceptance of the technique remains a huge hurdle, as many consumers seem uncomfortable with the practice due in part to privacy concerns over the vast amounts of data collected and analyzed when generating personalized advertisements. Therefore, it is critical to garner a better understanding of consumers’ attitudes towards personalized advertising in order to be able to use those insights to alleviate consumer privacy concerns. The purpose of this research is to work towards developing a more thorough understanding of consumers’ attitudes towards personalized advertising by exploring the antecedents and outcomes of those attitudes. In particular, we examine what factors determine whether personalized advertising is perceived favorably vs. invasively by consumers and what effects those perceptions have on consumers’ attitudes and intentions. The research lends contributions to academicians, marketing practitioners, and consumers by helping to achieve an increased understanding of personalized advertising’s impact on consumers’ perceptions. The empirical study employed in this research utilizes a conceptual framework that integrates privacy calculus theory with previous research on invasiveness, advertising acceptance, and innovation adoption. In addition, this research contributes to the marketing and information privacy literatures by making a theoretical connection between perceived invasiveness and its relationship with privacy concerns, as well as its impact on consumers’ attitudes and behavioral intentions. The results from the empirical research reveal that a number of constructs, such as perceived invasiveness, privacy concerns, perceived usefulness, and consumer innovativeness demonstrate significant relationships with consumers attitudes and behavioral intentions in the context of personalized advertising. Implications for managers, researchers, and consumers are discussed. / Includes bibliography. / Dissertation (Ph.D.)--Florida Atlantic University, 2014. / FAU Electronic Theses and Dissertations Collection

Esta pesquisa descreve e analisa a padronização de informações gerenciais em sistemas de informações de uma empresa multinacional. A estratégia de pesquisa adotada foi um estudo de caso único que contou com a triangulação de diversas fontes de evidência: arquivos, documentos, entrevistas e observação direta e participante. O referencial teórico aborda estudos antecedentes ao tema, conceitos sobre empresas internacionalizadas, informações gerenciais, Sistemas Integrados de Gestão (ERP) e Business Intelligence (BI). A migração para um sistema global está contextualizada num cenário de globalização em que está presente a intensificação das transações entre países em termos de investimentos, mercadorias e serviços, entre outras características. Neste cenário, observa-se maior demanda por informações padronizadas e comparáveis tanto para usuários externos quanto internos às organizações. Os avanços na tecnologia de informação, por exemplo, em bancos de dados e comunicação em redes viabilizou as iniciativas das empresas para implementação de sistemas compartilhados por diversos países. Assim como outras empresas na década de 90, as subsidiárias da empresa em estudo adotaram soluções locais de ERP. Já no ano 2000, durante o planejamento de longo prazo, decidiu-se que seria melhor partir para uma solução global de ERP e de BI em vez de prosseguir com melhorias em cada um dos sistemas locais. Em 2001, iniciou-se o projeto global com a primeira implementação e em 2011 o sistema é utilizado em aproximadamente 25 países. O suporte da alta gestão e participação dos usuários mostraram-se relevantes no que se refere a \"implementar com\" em vez de \"implementar para\" os usuários. Com relação à utilização das informações, os relatórios mais acessados no BI são relacionados a vendas e margens, controladoria, contabilidade e ordens de produção. No entanto, os relatórios de vendas e margens não foram os mais padronizados, pois embora as fontes dos dados sejam globais, os relatórios e consultas às informações são predominantemente locais devido à ausência de uma estrutura organizacional de vendas em nível mundial e ao fato de que a maioria dos usuários somente tem permissão de acesso aos dados de suas respectivas regiões. Já os relatórios de controladoria, contabilidade e compras apresentaram maior aderência aos relatórios globais, uma vez que estas áreas possuem responsáveis em nível global que fomentaram o uso e desenvolvimento de informações mundialmente padronizadas. No caso da empresa deste estudo, observa-se que, enquanto o ERP foi mais direcionado à consistência dos processos e com poucos relatórios pré-formatados, o BI foi direcionado aos relatórios e consultas de informações com a flexibilidade da análise multidimensional. Os benefícios identificados do sistema de BI global foram: consolidação de informações das subsidiárias; comparabilidade das informações entre subsidiárias; flexibilidade do BI para conversão de moedas; rastreabilidade e maior controle sobre a qualidade dos dados. Já as principais limitações do BI global foram: adaptação ao fuso horário dos países; equilibrar a necessidade de rapidez nas consultas com o nível de detalhe e o volume de dados armazenado; pouca flexibilidade para modificações das estruturas de dados já criadas; dificuldade de manter um catálogo de relatórios globais para todos os usuários e concentração da utilização dos relatórios para análises de informações passadas em vez de análises preditivas. A partir das evidências empíricas desta pesquisa, conclui-se que não há o cenário em que o ideal seria tudo sob um padrão mundial ou deixar tudo local, pois as necessidades em nível global e local são ajustadas de forma dinâmica, conforme as características da organização e perspectivas de negócio. / This research describes and analyzes the management information standardization in the information systems of a multinational company. The research strategy adopted was a single case study which involved multiple sources of evidence including: files, documents, interviews, and direct and participant observation. The literature review covers prior studies related to this subject, concepts about international companies, management information, Enterprise Resource Planning (ERP) and Business Intelligence (BI) systems. The migration to a global system is part of the overall globalization scenario with an ever increasing number of transactions between countries in terms of both investments and goods and services. In this context, there is a higher demand for standardized information to external and internal users. The breakthroughs in information technology, for example data bases and network communications, make possible corporate initiatives to have systems shared by subsidiaries located in several countries. Like other companies in the nineties, the subsidiaries of the studied company adopted their local ERP solutions. In the year 2000, during the five-year plan, it was determined that a global ERP and BI solution would be better than to continue improving each of the local systems. 2001 was the first roll-out toward the global system and now in 2011 the system is used in approximately 25 countries. The support from top management and involvement from business users was relevant to the success in implementing a global system (\"implement with\" instead of \"implement to\" the users). An analysis of information usage in the BI system showed that, the most accessed reports were related to sales and margins, controlling, accounting and production. However, reports about sales and margins are not standardized across the globe despite sharing the same data sources. These reports are mostly local due to the fact that there is no organization structure in charge of global sales and most importantly that users only have security authorization to their local information. The reports about controlling, accounting and purchasing presented a higher standardization once these areas had a global management structure that fostered the use and development of global reports. In this studied company, it was noticed that the ERP and BI complement each other. While the ERP system was focused on providing and enforcing consistent processes with only a few pre-formatted reports, the BI was focused on giving users the flexibility to retrieve information from reports. The benefits identified from a global BI were: consolidation of subsidiaries data, comparable information between subsidiaries, flexibility to handle currency conversions and better tracking and control over data quality. On the other hand, the main limitations identified were: inability to adapt to different time zones, a challenge in balancing between quicker response time and level of detail and volume of data necessary, little flexibility for changes in the current data storage models and in the creation of new ones, difficulties in the maintenance of a global catalogue of reports available for all users, and that reports are mostly used for analysis of past data and not predictive analysis. From the empirical evidences of this research, the conclusion indicates that there is not a scenario in which the ideal is all management information under a global standard or under a local requirement, because the needs at global and local levels are adjusted dynamically according to the company characteristics and business perspectives.

Este trabalho apresenta a construção de um sistema especialista aplicado ao diagnóstico de doenças cardíacas, usando como ferramenta computacional redes Bayesianas. O trabalho envolveu a interação entre diferentes áreas do conhecimento, engenharia e medicina, com maior foco na metodologia da construção de sistemas especialistas. São apresentados os processos de definição do problema, modelagem qualitativa e quantitativa, e avaliação. Neste trabalho, os processos de modelagem e avaliação foram realizados com o auxílio de um especialista médico e de dados bibliográficos. São apresentados como resultados a rede Bayesiana construída e um software para manipulação de redes Bayesianas denominado iBNetz. / This work presents the construction of an expert system applied to the diagnosis of heart diseases, using Bayesian networks as a modeling tool. The work involved interactions between two different fields, engineering and medicine, with special emphasis on the methodology of building expert systems. The processes of problem definition, qualitative and quantitative modeling, and evaluation are presented here. In this work, the modeling and evaluation processes have been conducted with the aid of a medical expert and bibliographic sources. The work has produced a Bayesian network for diagnosis and a software, called iBNetz, for creating and manipulating Bayesian networks.

by Tam Chi-Fai. / Thesis (M.Phil.)--Chinese University of Hong Kong, 1999. / Includes bibliographical references (leaves 87-88). / abstract --- p.I / table of content --- p.II / list of figures --- p.V / acknowledgments --- p.VII / Chapter chapter 1 --- introduction --- p.1 / Chapter 1.1 --- Two Stage Operations Planning --- p.1 / Chapter 1.2 --- Iterative Activities in the Two Stage Planning Approach --- p.3 / Chapter 1.3 --- Expert Decision Support System for Two Stage Planning --- p.4 / Chapter 1.4 --- Scope of the Study --- p.5 / Chapter 1.5 --- Organization of the Thesis --- p.6 / Chapter chapter 2 --- literature review --- p.7 / Chapter 2.1 --- Network Design for Air Express Service --- p.7 / Chapter 2.2 --- Integrative Use of Optimization and Simulation Model --- p.8 / Chapter 2.3 --- Expert System & Decision Support System --- p.11 / Chapter 2.3.1 --- Expert System --- p.11 / Chapter 2.3.2 --- Decision Support System --- p.13 / Chapter 2.3.3 --- ES / DSS Integration --- p.14 / Chapter chapter 3 --- research methodology --- p.19 / Chapter 3.1 --- Review on DSS / ES Integration --- p.19 / Chapter 3.2 --- System Design --- p.20 / Chapter 3.3 --- Prototyping --- p.22 / Chapter 3.4 --- Analysis and Evaluation --- p.23 / Chapter chapter 4 --- system architecture and knowledge modeling --- p.24 / Chapter 4.1 --- Architecture Overview --- p.24 / Chapter 4.1.1 --- System Architecture and Interactions --- p.26 / Chapter 4.1.2 --- Decision Support System --- p.27 / Chapter 4.1.3 --- Expert System --- p.32 / Chapter 4.2 --- System Operations --- p.35 / Chapter 4.2.1 --- Operations Flow --- p.35 / Chapter chapter 5 --- case study and prototyping --- p.38 / Chapter 5.1 --- Case Background --- p.38 / Chapter 5.1.1 --- The Service Network --- p.38 / Chapter 5.1.2 --- Objectives of the Project --- p.40 / Chapter 5.1.3 --- Network Design Methodology --- p.41 / Chapter 5.2 --- Iterative Network Planning --- p.49 / Chapter 5.2.1 --- Multi-period Network Planning Feedback --- p.50 / Chapter 5.2.2 --- Feedback in Validation and Evaluation --- p.51 / Chapter 5.3 --- The System Prototype --- p.57 / Chapter 5.3.1 --- Data Management and Model Manipulation --- p.57 / Chapter 5.3.2 --- Intelligent Guidance for the Iterations --- p.65 / Chapter chapter 6 --- evaluation and analysis --- p.75 / Chapter 6.1 --- Test Scenario for Network Planning --- p.75 / Chapter 6.1.1 --- Consultation Process --- p.75 / Chapter 6.1.2 --- Consultation Results --- p.78 / Chapter 6.2 --- Effectiveness of EDSS in Network Planning --- p.81 / Chapter 6.3 --- Generalized Advancement and Limitation --- p.82 / Chapter chapter 7 --- conclusion --- p.85 / bibliography --- p.87 / appendices --- p.89

Decision support systems will be an asset dealing with the complexity involved in many decision situations for companies, organizations and societies by integrating different aspects into a holistic pattern. That creates a close relationship to systems science since systems thinking promote holism as a profitable way to handle complexity. The ideal decision support system should not be used to make automatic decisions but to assist a human being in the decision process. That process is sometimes described as a model consisting of the phases, intelligence, design and choice. Intelligence is needed to understand the situation and find the information that is needed to continue the process. Design means designing different alternatives and in the last phase, choice, the alternatives are evaluated and the best alternative is chosen. A good decision support system should give the user assistance through the whole process. The main purpose of our research is identifying the process of building an efficient Decision Support System. The target groups are the people who are working with multinational companies that are specialized in constructing and delivering decision support systems to end users. The number of target companies involved in this study is only two and is limited Indian Multinational companies. The theoretical study helps in identifying the basic characteristics of a decision support system, exploring the types of decision support systems used in current organizations, resulting if there is any particular standard for constructing DSS today and signifying approach for constructing a user friendly decision support system by analyzing the existing literature related to DSS. At the same time, empirical study advances the research problem from a practical angle. The conclusion for this research is a comprehensive report in relation to the varieties of Decision Support Systems used in today’s organizations, qualities that a decision support system ought to possess and suggested process to be implemented for building an efficient decision support system. / Program: Magisterutbildning i informatik