Etemadi Fuzzy Linear Regression (EFLR)

Modeling and forecasting are among the most powerful and widely-used tools in decision support systems. The Fuzzy Linear Regression (FLR) is the most fundamental method in the fuzzy modeling area in which the uncertain relationship between the target and explanatory variables is estimated and has been frequently used in a broad range of real-world applications efficaciously. The operation logic in this method is to minimize the vagueness of the model, defined as the sum of individual spreads of the fuzzy coefficients. Although this process is coherent and can obtain the narrowest α-cut interval and exceptionally the most accurate results in the training data sets, it can not guarantee to achieve the desired level of generalization. While the quality of made managerial decisions in the modeling-based field is dependent on the generalization ability of the used method. On the other hand, the generalizability of a method is generally dependent on the precision as well as reliability of results, simultaneously. In this paper, a novel methodology is presented for the fuzzy linear regression modeling; in which in contrast to conventional methods, the constructed models' reliability is maximized instead of minimizing the vagueness. In the proposed model, fuzzy parameters are estimated in such a way that the variety of the ambiguity of the model is minimized in different data conditions. In other words, the weighted variance of different ambiguities in each validation data situation is minimized in order to estimate the unknown fuzzy parameters. To comprehensively assess the proposed method's performance, 74 benchmark datasets are regarded from the UCI. Empirical outcomes show that, in 64.86% of case studies, the proposed method has better generalizability, i.e., narrower α-cut interval as well as more accurate results in the interval and point estimation, than classic versions. It is obviously demonstrated the importance of the outcomes' reliability in addition to the precision that is not considered in the traditional FLR modeling processes. Hence, the presented EFLR method can be considered as a suitable alternative in fuzzy modeling fields, especially when more generalization is favorable.

MODELAGEM NEURO-FUZZY DA EVAPOTRANSPIRAÇÃO DE REFERÊNCIA BASEADA NO MÉTODO DE CAMARGO

MODELAGEM NEURO-FUZZY DA EVAPOTRANSPIRAÇÃO DE REFERÊNCIA BASEADA NO MÉTODO DE CAMARGO LUANA POSSARI MAZIERO1, STEPHANIE LESCHOT FREDERICK2, CAMILA PIRES CREMASCO2, FERNANDO FERRARI PUTTI2 E LUÍS ROBERTO ALMEIDA GABRIEL FILHO3 1Departamento de Engenharia Rural, Faculdade de Ciências Agronômicas, Universidade Estadual Paulista, Avenida Universitária, 3780, CEP 18610-034, Altos do Paraíso, Botucatu, São Paulo, Brasil, [email protected]. 2Departamento de Engenharia de Biossistemas, Faculdade de Ciências e Engenharia, Universidade Estadual Paulista, Rua Domingos da Costa Lopes, nº 780, Jardim Itaipu, 17602-496, Tupã, São Paulo, Brasil, [email protected], [email protected], [email protected]. 3Departamento de Gestão, Desenvolvimento e Tecnologia, Faculdade de Ciências e Engenharia, Universidade Estadual Paulista, Rua Domingos da Costa Lopes, nº 780, Jardim Itaipu, 17602-496, Tupã, São Paulo, Brasil, [email protected]. 1 RESUMO O conhecimento sobre a evapotranspiração é fundamental para determinar o balanço hídrico de uma determinada região, pois pode afetar a política de gestão hídrica da bacia. Nesse contexto, o uso de modelagem matemática com abordagem difusa, como a modelagem fuzzy, cuja origem se deu justamente devido ao desafio de se trabalhar com incertezas, pode auxiliar na determinação da evapotranspiração, auxiliando no processo de tomada de decisão. Desta forma, no presente artigo, desenvolveu-se um modelo neuro-fuzzy (baseado em lógica fuzzy e redes neurais) para determinar a evapotranspiração de referência pelo método de Camargo. Definiu‑se como variáveis de entrada a temperatura e radiação solar, ambas coletadas pelo Instituto Nacional de Meteorologia (INMET) na estação de Tupã, os dados foram considerados pelo período de um ano. Tal sistema, possibilita ao produtor a obtenção instantânea do valor da evapotranspiração de referência, além da classificação qualitativa em classes. A partir dos processos realizados neste trabalho, o método computacional estabelecido, mostrou-se capaz de calcular instantaneamente a evapotranspiração de referência pela equação de Camargo, a partir das variáveis radiação solar e temperatura, relatando que quanto menor os valores de temperatura e radiação solar, menor será o valor da evapotranspiração de referência. Palavras-chave: lógica fuzzy, redes neurais, irrigação, balanço hídrico. MAZIERO, L. P.; FREDERICK, S. L.; CREMASCO, C. P.; PUTTI, F. F.; GABRIEL FILHO, L. R. A. NEURO-FUZZY MODELING OF REFERENCE EVAPOTRANSPIRATION BASED ON THE CAMARGO METHOD 2 ABSTRACT Knowledge about evapotranspiration is essential to determine the water balance of a given region, since it can affect the basin's water management policy. In this context, the use of mathematical modeling with diffuse approach as fuzzy modeling, in which its origin was rightly due to the challenge of working with uncertainties, it can assist in the determination of evapotranspiration, helping in the decision-making process. Thus, in this article, he developed a neuro-fuzzy model (based on fuzzy logic and neural networks) to determine the reference evapotranspiration by the Camargo method. The input variables were temperature and solar radiation, both collected at the National Meteorology Institute (INMET) at the Tupã station, the data were considered for a period of one year. Such a system allows the producer to instantly obtain the reference evapotranspiration value, in addition to the qualitative classification in classes. Based on the processes conducted in this work, the established computational method could instantly calculate the reference evapotranspiration from the Camargo equation, based on solar radiation and temperature variables, reporting that the lower the values of temperature and solar radiation, the lower will be the reference evapotranspiration value. Keywords: diffuse logic, neural networks, irrigation, water balance.

Fuzzy modelling of the transportation logistics processes

The relevance of mechanisms and fuzzy modeling methods used in transport and logistics processes is justified. The logic of the transport and logistics chains study by their decomposition into separate economic entities with subsequent synthesis, with taking into account their conflicting interests, is presented. The task of managing transport and logistics processes is set within the framework of the conceptual positions of the fuzzy sets theory: the use of linguistic variables, the concept of a fuzzy set, fuzzy inference, implemented on the basis of fuzzy sets operations. In this study, the fuzzy modeling on the bottom-up conclusions from the premises to the conclusion was used. Various methods of implication and defuzzification have been analyzed. An iterative procedure for managing transport and logistics chains and their links has been proposed, which ensures the adaptation of the process to the specified performance indicators. A calculated example is given for a marshalling yard as a link in the transport and logistics chain.