Esta investigación ha permitido construir una Red Neuronal Artificial RNA con Función de Base Radial, y que utiliza la distancia de Mahalanobis RND, para la mejora de la calidad de diseño de procesos, obteniendo mejores resultados que los obtenidos con los análisis estadísticos tradicionales para los diseños experimentales y las RNA ya existentes, para los casos que se trabaje con varias variables dependientes e independientes y en los que sus relaciones no sean lineales. Asimismo, al RND permite obtener parámetros de entrada para lograr un nivel de calidad deseado; para ello se aplica una metodología que usa las RNA Inversa y Directa a la vez.

This article deals with the application of neuronal networks in multi-answer problems to improve quality. From a conceptual analysis, an application showing the efficiency of neuronal networks using the backpropagation network with training is shown through backpropagation-traingdn.

The article deals with the application of Neural Networks in parameters optimization problems for the improvement of processes in which the answers of quality are qualitative. We propose a methodology to improve the quality of processes using neural networks and present an application of neural networks in the solution of problems in which the answers of quality are qualitative.

The work has identified four dimensións related to service quality of the Library of the Universidad Nacional Mayor de San Marcos UNMSM. Three of them, which has been called “Tangible Aspects, Empathy and Reliability” can be considered equivalent to those originally obtained by Parasuraman, Zeithaml and Berry. The remaining dimensión is the integrated by Capacity of Response and Security has been called “Response-Security”. The dimensións found are consistent with the findings of other research in the university sector, but not included among those found by those authors. Also, has been tested the use of Neural Networks techniques and grouping with K Means, giving an equivalent result to factorial analysis. In the case of neural networks was used competitive learning with Learning Vector Quantitative LVQ.

The paper aims to review the existing methodologies for multiresponse optimization, integrate them into one and develop a new algorithm that allows to overcome the existing limitations. For this purpose we reviewed statistical optimization methodologies using the traditional response surface methodology with robust design, then reviewed the application of the bayesian approach to that obtained with traditional statistics, and finally reviewed artificial neural network applications to cases of optimization. After performing the analysis and discussion about the three methodologies were integrated into one, having developed a new algorithm to overcome the limitations and shortcomings of the previous methods. Also, we compared the results obtained with other methods with those obtained with the new method, with favorable outcome. Thus we have developed a multi-response optimization methodol...

The present article is about the analysis of systems for the solution of complex problems that presents in the organization in order to plan the operative design of the organization especially the system of work of high level.

This article aims at presenting a proposal to apply the Taguchi methods in improving production processes of small and medium enterprises. The theoretical framework for a methodology that has been simplified, elsewhere the application of case studies will be presented this opportunity presents

The article deals with the application of Neural Networks in parameters optimization problems for the improvement of processes in which the answers of quality are qualitative. We propose a methodology to improve the quality of processes using neural networks and present an application of neural networks in the solution of problems in which the answers of quality are qualitative.

The work has identified four dimensións related to service quality of the Library of the Universidad Nacional Mayor de San Marcos UNMSM. Three of them, which has been called “Tangible Aspects, Empathy and Reliability” can be considered equivalent to those originally obtained by Parasuraman, Zeithaml and Berry. The remaining dimensión is the integrated by Capacity of Response and Security has been called “Response-Security”. The dimensións found are consistent with the findings of other research in the university sector, but not included among those found by those authors. Also, has been tested the use of Neural Networks techniques and grouping with K Means, giving an equivalent result to factorial analysis. In the case of neural networks was used competitive learning with Learning Vector Quantitative LVQ.

The paper aims to review the existing methodologies for multiresponse optimization, integrate them into one and develop a new algorithm that allows to overcome the existing limitations. For this purpose we reviewed statistical optimization methodologies using the traditional response surface methodology with robust design, then reviewed the application of the bayesian approach to that obtained with traditional statistics, and finally reviewed artificial neural network applications to cases of optimization. After performing the analysis and discussion about the three methodologies were integrated into one, having developed a new algorithm to overcome the limitations and shortcomings of the previous methods. Also, we compared the results obtained with other methods with those obtained with the new method, with favorable outcome. Thus we have developed a multi-response optimization methodol...

The present article is about the analysis of systems for the solution of complex problems that presents in the organization in order to plan the operative design of the organization especially the system of work of high level.

This article aims at presenting a proposal to apply the Taguchi methods in improving production processes of small and medium enterprises. The theoretical framework for a methodology that has been simplified, elsewhere the application of case studies will be presented this opportunity presents

This research has led to construct an artificial neural network ARN with Radial Basis Function, and using Mahalanobis distance RND, for improving the quality of process design, which have performed better than those obtained with the for traditional statistical analysis of design of experiments and other RNA that already exist, for cases that are working with several independent and dependent variables in which its relations are not linear. It also allows with the RND obtain input parameters to achieve a desired level of quality, for it applies a methodology that uses RNAReverse and Direct at once.

This article deals with the application of neuronal networks in multi-answer problems to improve quality. From a conceptual analysis, an application showing the efficiency of neuronal networks using the backpropagation network with training is shown through backpropagation-traingdn.

La investigación ha desarrollado una metodología para medir la calidad y elaborar un modelo de calidad del servicio mediante el uso de ecuaciones estructurales. La investigación se aplicó en el comedor de estudiantes de la UNMSM. La metodología consiste en elaborar una encuesta según la técnica SERVQUAL de 22 preguntas y aplicarlas a una muestra representativa. Seguidamente se elabora el Análisis Factorial Exploratorio (AFE) con el objeto de determinar las dimensiones de la calidad del servicio, luego se realiza en Análisis Factorial Confirmatorio (AFC) con el objeto de cuantificar las relaciones entre las variables y las dimensiones; y finalmente, se realiza la determinación del Modelo de Ecuaciones Estructurales (MEE) que permite conocer la relación entre las dimensiones, lo cual constituye el modelo de la calidad del servicio. Para elaborar el AFE se utilizaron los software...

In this study we have developed a method for optimizing the parameters of quality of products that consists of five steps: 1) Determine the characteristics of product quality and process variables 2) Develop an experimental design with Taguchi Methods 3) Develop experiments with Response Surface Methodology. 4) Determine a neural network that represents the relationships between variables and quality characteristics. Using fuzzy variables if there is information not deterministic. 5) Optimize with the use of genetic algorithms. In this proposal, artificial neural networks ANN allow to estimate response functions; in the case of having the qualitative variables these are processed with fuzzy logic LD and in the optimization step genetic algorithms GA are used. An example of optimization with multiple responses is presented to verify the method.

Research has developed a methodology to measure the quality and develop a model of service quality using structural equations. The research was applied at the student’s cafeteria of San Marcos University. The methodology is to develop a survey technique according to SERVQUAL of 22 questions and apply them to a representative sample. Following the exploratory factor analysis (AFE) in order to determine the dimensions of service quality is made, then confirmatory factor analysis (AFC) is applied in order to quantify the relationships between variables and dimensions; and finally, is performed the structural equation model (MEE) to find out the relationship between the dimensions, which is the model of service quality. To prepare the AFE the Lisrel and SPSS software were used; and for AFC and MEE the Lisrel software was used.

In this study we have developed a method for optimizing the parameters of quality of products that consists of five steps: 1) Determine the characteristics of product quality and process variables 2) Develop an experimental design with Taguchi Methods 3) Develop experiments with Response Surface Methodology. 4) Determine a neural network that represents the relationships between variables and quality characteristics. Using fuzzy variables if there is information not deterministic. 5) Optimize with the use of genetic algorithms. In this proposal, artificial neural networks ANN allow to estimate response functions; in the case of having the qualitative variables these are processed with fuzzy logic LD and in the optimization step genetic algorithms GA are used. An example of optimization with multiple responses is presented to verify the method.

Aplica los métodos Taguchi a las pequeñas y microempresas, según metodología que contempla: plantear el problema, diseño del experimento, realización de pruebas, análisis de resultados y confirmación de resultados. Esta metodología es aplicada a casos específicos: elaboración de velas artesanales y de pop corn. En el caso de velas artesanales se llegó a la combinación óptima de los factores: A2 (90% parafina- 1 O% estearina), B2 (mecha delgada), C1 (molde de latón), D2 (con agitación) y E2 (llama media). En el caso del pop corn se llegó a la combinación óptima de los afctores: A1 (aceite de soya), B2 (50 cm^3 aceite), C2 (llama alta), 01 (sin precalentamiento), E2 (con agitación) y F2 (olla oval). Las principales ventajas de la metodología propuesta son la necesidad de menos pruebas, la elaboración del diseño es más simple, al igual que los cálculos y existe mayo...

This study proposes a model to improve identity registration service and correlates this service to the process quality and technology by means of structural equations using SPSS and AMOS. To reinforce the analysis of the information obtained from documents and records of the processes of identity registration, a 21 question survey was given to citizens coming to 1 of 16 regional headquarters, from which it was possible to demonstrate through structural equations, the existence of a strong relationship between the three variables: process quality, technology and service improvement.