The diploma thesis deals with fuzzy logic and its use in business activities of a distribution company. Two models using fuzzy set theory are presented. These models are able to evaluate quality of customers through their attributes. Based on these calculations, they are classified into several price categories which differ in margin.

The diploma thesis deals with the use of fuzzy logic in the evaluation of customers in the business company SPINA Trade, s.r.o. MS Excel and MATLAB are used to program two models. Based on the input attributes regarding the customer, the model evaluates the recommended approach to the customer and his priority.

In this thesis a novel approach to compare airborne observations of spectral upward and downward irradiances with along-track radiative transfer simulations (RTS) are presented. The RTS are performed with the ecRad radiation scheme of the Integrated Forecast System (IFS) operated by the European Centre for Medium Range Weather Forecast (ECMWF) and the library for Radiative transfer (libRadtran) on basis of hourly 0.1° IFS analysis data (IFS AD). The comparison aims to investigate the general capability of the utilized models to reproduce the observed radiation field. Simultaneous utilization of ecRad and libRadtran, driven by the same IFS AD, and comparison with observations enables to separate for potential errors in the applied IFS AD and ecRad.

Previous authors have developed many different types of water pipe condition and failure models using the various methodologies available.  Contrary, current utilities are struggling to maintain their current water infrastructure system, due to the lack of effective prediction tools at hand.  The gap between the methodologies available in academic research and the tools available to current water utilities needs to be addressed.  This paper presents a fuzzy inference prediction model used to forecast the performance rating of individual drinking water pipeline sections (node to node) in which utilities can easily apply to their drinking water infrastructure system. Prior to the development of a prediction model, a through literature and current practice review is completed detailing and summarizing all the available mathematical models.  Following, an infrastructure overview is presented detailing the various pipe materials, lifecycle and failure modes and mechanisms.  A data structure is also detailed which lists all parameters that affect the condition and/or performance of a pipeline.  All of these tools are successfully used to develop a fuzzy inference performance model. The fuzzy inference performance model is considered novel in that it considers close to 30 pipe parameters.  Moreover, the performance model is applied using the Western Virginia Water Authority (WVWA) and the Washington Suburban Sanitary Commission (WSSC) databases to evaluate and verify the predicting results.  Lab testing of several pipe samples is also used to evaluate the model.  The testing consists of a ring bearing test which is used to calculate the rupture modulus of the pipe.  Comparing the original vs. the current rupture modulus can determine the remaining strength of the pipe.  The remaining strength can then be used to assess the performance results predicted by the fuzzy inference model. Further a framework is set forth which utilizes the model's predicted performance ratings to develop deterioration curves which can be used as a tool to forecast and plan future inspection, repair, rehabilitation and replacement of water pipelines.  The deterioration model is made up of a Markov chain approach coupled with a non-optimization technique. / Ph. D.

The ability to predict the weather carries great societal benefit. To ascertain reliable predictions the numerical weather prediction models used need evaluation with particular attention paid to their representations of atmospheric processes. This type of process-based model evaluation is performed through comparison of large datasets of observational data and  model results. The Year Of Polar Prediction Project site Model Intercomparison Project (YOPPsiteMIP) works to further process-based model evaluation in polar regions and has, to this end, created extensive datasets for these types of analysis. However, the lack of standardized tools to visualize the analyses carries with it significant limitations for both the usability of the data as well as the standardization of the visualizations created from it. To amend this, a set of visualization tools have been created. They were evaluated in the context of visualization quality and source code maintainability. They were found satisfactory in all cases except for the runtime. These tools further the ability to perform process-based analysis with the YOPPsiteMIP datasets in standardized formats. They are limited to the project specific MDF file type. However, with the continued spread of the MDF file type the tools become increasingly useful in furthering model evaluation on larger scales than YOPPsiteMIP.

Haptic-based Virtual Reality (VR) applications have many merits. What is still obscure, from the designer’s perspective of these applications, is the experience the users will undergo when they use the VR system. Quality of Experience (QoE) is an evaluation metric from the user’s perspective that unfortunately has received limited attention from the research community. Assessing the QoE of VR applications reflects the amount of overall satisfaction and benefits gained from the application in addition to laying the foundation for ideal user-centric design in the future. In this thesis, we address certain issues and concerns regarding QoE of virtual environments. In essence, we propose a taxonomy for the evaluation of the QoE for multimedia applications and in particular VR applications. The taxonomy classifies QoE related parameters into groups. The groups’ organization is generated from the definition we have adopted for the QoE which is the Quality of Service (QoS) plus the user experience (UX). We model this taxonomy using first mathematical modeling based on weighted averages and then a Fuzzy logic Inference System (FIS) to quantitatively measure the QoE of haptic virtual environments. We test both models conducting user study analysis to evaluate the QoE of a VR application. These models serve as engines that facilitate the calculation of QoE with minimal amount of users. We specifically attend to the issue of the new media, haptics, within the context of increasing the QoE of virtual environments (VE). This special attention is important for comparing the effect of tactile and kinesthetic feedback on the QoE. In accordance, we investigate a particular topic that seems to have a colossal effect on QoE throughout our analysis, which is fatigue. Our analysis involved users' studies since the main focus is on the user. The QoE for virtual environments is in its primary stages. This thesis tackles issues that are vital in dealing with and understanding the importance of QoE. The various results suggest a positive user's disposition toward haptics and virtual environments, yet there will always be obstacles and challenges such as fatigue that if minimized will enhance the QoE of haptic-based applications.

Submitted by Ingryd Palloma Teodósio da Nóbrega (palloma_nobrega@hotmail.com) on 2018-04-19T12:42:22Z No. of bitstreams: 1 Dissertação de Mestrado_definitivo.pdf: 1192753 bytes, checksum: 780733e53b843a769b459dd932674e60 (MD5) / Approved for entry into archive by Alexandra Maria Donadon Lusser Segali null (alexmar@fcav.unesp.br) on 2018-04-19T12:57:20Z (GMT) No. of bitstreams: 1 nobrega_ipt_me_jabo.pdf: 1192753 bytes, checksum: 780733e53b843a769b459dd932674e60 (MD5) / Made available in DSpace on 2018-04-19T12:57:20Z (GMT). No. of bitstreams: 1 nobrega_ipt_me_jabo.pdf: 1192753 bytes, checksum: 780733e53b843a769b459dd932674e60 (MD5) Previous issue date: 2018-02-23 / Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) / Compreender o metabolismo energético das aves e como a energia é utilizada para mantença, ganho de peso e produção de ovos, permite a elaboração e avaliação de modelos que estimam a exigência nutricional considerando as diferenças de peso corporal (P), ganho de peso (GP) e massa de ovo (MO) na avicultura industrial. Objetivou-se com esta pesquisa analisar os coeficientes que representam a partição da energia ingerida por codornas japonesas na fase de produção de ovos, a partir de um estudo dose-resposta. Foram utilizadas 70 codornas japonesas da linhagem VICAMI®, com 24 semanas de idade, durante 8 semanas, alojadas em galpão convencional. Duas dietas foram formuladas, uma com alto (3.600 kcal/kg) e a outra com baixo (2.100 kcal/kg) teor de energia. Para modificar a energia retida pelas aves, foi empregada a técnica da diluição, obtendo os níveis crescente de energia metabolizável da dieta. Foram utilizados sete tratamentos distribuídos inteiramente ao acaso, com dez repetições e uma codorna por unidade experimental. Os tratamentos foram 2.118; 2.381; 2.557; 2.776; 2.908; 3.171 e 3.435 kcal/kg, com base na composição analisada das dietas determinadas em ensaio de metabolismo. As variáveis analisadas foram ingestão de energia metabolizável (IEM), produção de calor (PC) e energia retida (ER) expressas em kcal/kg0,67. A energia metabolizável para mantença (EMm) foi obtida a partir da relação entre ER e IEM, considerando a condição ER = 0. A exigência metabolizável para ganho de peso (EMg) foi estimada pela relação entre eficiência de utilização de energia (k) e energia líquida para ganho (ELg). A exigência de energia metabolizável para massa de ovo (EMo), foi obtida por meio da energia retida no ovo, dividida pela eficiência de utilização de energia para massa de ovo (ko). Os valores estimados para EMm, EMg e EMo foram 155,60 kcal/kg P0,67; 5,89 kcal/g e 2,74 kcal/g, respectivamente. Os modelos que predizem a IEM baseados nos parâmetros que representam exigência de energia de acordo com seu fracionamento, foram avaliados por meio da decomposição linear do erro (observado - predito), em erro escalar e viés de predição, obtidos por regressão linear entre os erros e valores preditos. O modelo obtido foi IEM = 155,60 × P0,67 + 5,89 × GP + 2,74 × MO, a composição do erro analisado (3,79 kcal/ave.dia) mostrou que o modelo é imparcial, com 93% de precisão nas estimativas, portanto, capaz de predizer o conjunto de dados analisados, o que valida seu uso. / – To understand the energy metabolism of birds and how energy is used for maintenance, weight gain and egg production, allow the elaboration and evaluation of models that estimate the nutritional requirement considering the differences of body weight (BW), body weight gain (BWG) and egg mass (EO) in industrial poultry. The objective of this research was to analyze the coefficients that represent the partition of the energy ingested by Japanese quails in the egg production phase, from a doseresponse study. In a conventional shed it was housed 70 Japanese quails of VICAMI® line, at 24 weeks old, during 8 weeks. Based on the dilution technique it was formulated two energy levels diets, one with high (3,600 kcal/kg) and other with low (2,100 kcal/kg). Seven treatments were randomly distributed, with ten replicates and one quail per experimental unit. Treatments were levels of metabolizable energy in the diet being: 2,118; 2,381; 2,557; 2,776; 2,908; 3,171 and 3,435 kcal/kg. The variables analyzed were metabolizable energy intake (MEI), heat production (HP) and retained energy (RE) expressed in kcal/kg0.67. The metabolizable energy for maintenance (MEm) was obtained from the relation between RE and MEI, solving the equation RE = 0. The metabolizable requirement for weight gain (MEg) was estimated by the relation between use efficiency of energy (k) and net energy for gain (NEg). The metabolizable energy requirement for egg mass (MEe) was obtained by the relation between energy retained in the egg, divided by the energy utilization efficiency for egg mass (ke). The estimated values for MEm, MEg, MEe were 155.60 kcal/kg BW0.67 , 5.89 kcal/g and 2.74 kcal/g; respectively. The models that predict MEI based on the parameters that represent the energy requirement according to its fractionation were evaluated by linear error decomposition (observed - predicted), in scalar error and prediction bias obtained by linear regression between errors and predicted values. The obtained model was MEI = 155.60 × BW0.67 + 5.89 × BWG + 2.74 × EO, the analyzed error composition (3.79 kcal/bird.day), of display showing the model is unbiased, with 93% accuracy in the estimates, therefore, able to predict the analyzed data set, which validates its use.

Haptic-based Virtual Reality (VR) applications have many merits. What is still obscure, from the designer’s perspective of these applications, is the experience the users will undergo when they use the VR system. Quality of Experience (QoE) is an evaluation metric from the user’s perspective that unfortunately has received limited attention from the research community. Assessing the QoE of VR applications reflects the amount of overall satisfaction and benefits gained from the application in addition to laying the foundation for ideal user-centric design in the future. In this thesis, we address certain issues and concerns regarding QoE of virtual environments. In essence, we propose a taxonomy for the evaluation of the QoE for multimedia applications and in particular VR applications. The taxonomy classifies QoE related parameters into groups. The groups’ organization is generated from the definition we have adopted for the QoE which is the Quality of Service (QoS) plus the user experience (UX). We model this taxonomy using first mathematical modeling based on weighted averages and then a Fuzzy logic Inference System (FIS) to quantitatively measure the QoE of haptic virtual environments. We test both models conducting user study analysis to evaluate the QoE of a VR application. These models serve as engines that facilitate the calculation of QoE with minimal amount of users. We specifically attend to the issue of the new media, haptics, within the context of increasing the QoE of virtual environments (VE). This special attention is important for comparing the effect of tactile and kinesthetic feedback on the QoE. In accordance, we investigate a particular topic that seems to have a colossal effect on QoE throughout our analysis, which is fatigue. Our analysis involved users' studies since the main focus is on the user. The QoE for virtual environments is in its primary stages. This thesis tackles issues that are vital in dealing with and understanding the importance of QoE. The various results suggest a positive user's disposition toward haptics and virtual environments, yet there will always be obstacles and challenges such as fatigue that if minimized will enhance the QoE of haptic-based applications.

In this work spectral airborne measurements of upward irradiance and a novel remote sensing technique for the cloud droplet number concentration are used to evaluate the representation of clouds in current operational weather prediction models.

Artificial intelligence (AI) and machine learning (ML) present some application opportunities and challenges that can be framed as learning problems. The performance of machine learning models depends on algorithms and the data. Moreover, learning algorithms create a model of reality through learning and testing with data processes, and their performance shows an agreement degree of their assumed model with reality. ML algorithms have been successfully used in numerous classification problems. With the developing popularity of using ML models for many purposes in different domains, the validation of such predictive models is currently required more formally. Traditionally, there are many studies related to model evaluation, robustness, reliability, and the quality of the data and the data-driven models. However, those studies do not consider the concept of the applicability domain (AD) yet. The issue is that the AD is not often well defined, or it is not defined at all in many fields. This work investigates the robustness of ML classification models from the applicability domain perspective. A standard definition of applicability domain regards the spaces in which the model provides results with specific reliability. The main aim of this study is to investigate the connection between the applicability domain approach and the classification model performance. We are examining the usefulness of assessing the AD for the classification model, i.e. reliability, reuse, robustness of classifiers. The work is implemented using three approaches, and these approaches are conducted in three various attempts: firstly, assessing the applicability domain for the classification model; secondly, investigating the robustness of the classification model based on the applicability domain approach; thirdly, selecting an optimal model using Pareto optimality. The experiments in this work are illustrated by considering different machine learning algorithms for binary and multi-class classifications for healthcare datasets from public benchmark data repositories. In the first approach, the decision trees algorithm (DT) is used for the classification of data in the classification stage. The feature selection method is applied to choose features for classification. The obtained classifiers are used in the third approach for selection of models using Pareto optimality. The second approach is implemented using three steps; namely, building classification model; generating synthetic data; and evaluating the obtained results. The results obtained from the study provide an understanding of how the proposed approach can help to define the model’s robustness and the applicability domain, for providing reliable outputs. These approaches open opportunities for classification data and model management. The proposed algorithms are implemented through a set of experiments on classification accuracy of instances, which fall in the domain of the model. For the first approach, by considering all the features, the highest accuracy obtained is 0.98, with thresholds average of 0.34 for Breast cancer dataset. After applying recursive feature elimination (RFE) method, the accuracy is 0.96% with 0.27 thresholds average. For the robustness of the classification model based on the applicability domain approach, the minimum accuracy is 0.62% for Indian Liver Patient data at r=0.10, and the maximum accuracy is 0.99% for Thyroid dataset at r=0.10. For the selection of an optimal model using Pareto optimality, the optimally selected classifier gives the accuracy of 0.94% with 0.35 thresholds average. This research investigates critical aspects of the applicability domain as related to the robustness of classification ML algorithms. However, the performance of machine learning techniques depends on the degree of reliable predictions of the model. In the literature, the robustness of the ML model can be defined as the ability of the model to provide the testing error close to the training error. Moreover, the properties can describe the stability of the model performance when being tested on the new datasets. Concluding, this thesis introduced the concept of applicability domain for classifiers and tested the use of this concept with some case studies on health-related public benchmark datasets. / Ministry of Higher Education in Libya