Aplicación de técnicas de inteligencia artificial a la predicción de contaminantes atmosféricos

Abstract Air pollution is a big threat and a phenomenon that has a specific impact on human health, in addition, changes that occur in the chemical composition of the atmosphere can change the weather and cause acid rain or ozone destruction. Those are phenomena of global importance. The World Health Organization (WHO) considerates air pollution as one of the most important global priorities. Salamanca, Gto., Mexico has been ranked as one of the most polluted cities in this country. The industry of the area led to a major economic development and rapid population growth in the second half of the twentieth century. The impact in the air quality is important and significant efforts have been made to measure the concentrations of pollutants. The main pollution sources are locally based plants in the chemical and power generation sectors. The registered concerning pollutants are Sulphur Dioxide (SO2) and particles on the order of ∼10 micrometers or less (PM10). The prediction in the concentration of those pollutants can be a powerful tool in order to take preventive measures such as the reduction of emissions and alerting the affected population. In this PhD thesis we propose a model to predict concentrations of pollutants SO2 and PM10 for each monitoring booth in the Atmospheric Monitoring Network Salamanca (REDMAS - for its spanish acronym). The proposed models consider the use of meteorological variables as factors influencing the concentration of pollutants. The information used along this work is the current real data from REDMAS. In the proposed model, Artificial Neural Networks (ANN) combined with clustering algorithms are used. The type of ANN used is the Multilayer Perceptron with a hidden layer, using separate structures for the prediction of each pollutant. The meteorological variables used for prediction were: Wind Direction (WD), wind speed (WS), Temperature (T) and relative humidity (RH). Clustering algorithms, K-means and Fuzzy C-means, are used to find relationships between air pollutants and weather variables under consideration, which are added as input of the RNA. Those relationships provide information to the ANN in order to obtain the prediction of the pollutants. The results of the model proposed in this work are compared with the results of a multivariate linear regression and multilayer perceptron neural network. The evaluation of the prediction is calculated with the mean absolute error, the root mean square error, the correlation coefficient and the index of agreement. The results show the importance of meteorological variables in the prediction of the concentration of the pollutants SO2 and PM10 in the city of Salamanca, Gto., Mexico. The results show that the proposed model perform better than multivariate linear regression and multilayer perceptron neural network. The models implemented for each monitoring booth have the ability to make predictions of air quality that can be used in a system of real-time forecasting and human health impact analysis. Among the main results of the development of this thesis we can cite: A model based on artificial neural network combined with clustering algorithms for prediction with a hour ahead of the concentration of each pollutant (SO2 and PM10) is proposed. A different model was designed for each pollutant and for each of the three monitoring booths of the REDMAS. A model to predict the average of pollutant concentration in the next 24 hours of pollutants SO2 and PM10 is proposed, based on artificial neural network combined with clustering algorithms. Model was designed for each booth of the REDMAS and each pollutant separately. Resumen La contaminación atmosférica es una amenaza aguda, constituye un fenómeno que tiene particular incidencia sobre la salud del hombre. Los cambios que se producen en la composición química de la atmósfera pueden cambiar el clima, producir lluvia ácida o destruir el ozono, fenómenos todos ellos de una gran importancia global. La Organización Mundial de la Salud (OMS) considera la contaminación atmosférica como una de las más importantes prioridades mundiales. Salamanca, Gto., México; ha sido catalogada como una de las ciudades más contaminadas en este país. La industria de la zona propició un importante desarrollo económico y un crecimiento acelerado de la población en la segunda mitad del siglo XX. Las afectaciones en el aire son graves y se han hecho importantes esfuerzos por medir las concentraciones de los contaminantes. Las principales fuentes de contaminación son fuentes fijas como industrias químicas y de generación eléctrica. Los contaminantes que se han registrado como preocupantes son el Bióxido de Azufre (SO2) y las Partículas Menores a 10 micrómetros (PM10). La predicción de las concentraciones de estos contaminantes puede ser una potente herramienta que permita tomar medidas preventivas como reducción de emisiones a la atmósfera y alertar a la población afectada. En la presente tesis doctoral se propone un modelo de predicción de concentraci ón de los contaminantes más críticos SO2 y PM10 para cada caseta de monitorización de la Red de Monitorización Atmosférica de Salamanca (REDMAS). Los modelos propuestos plantean el uso de las variables meteorol ógicas como factores que influyen en la concentración de los contaminantes. La información utilizada durante el desarrollo de este trabajo corresponde a datos reales obtenidos de la REDMAS. En el Modelo Propuesto (MP) se aplican Redes Neuronales Artificiales (RNA) combinadas con algoritmos de agrupamiento. La RNA utilizada es el Perceptrón Multicapa con una capa oculta, utilizando estructuras independientes para la predicción de cada contaminante. Las variables meteorológicas disponibles para realizar la predicción fueron: Dirección de Viento (DV), Velocidad de Viento (VV), Temperatura (T) y Humedad Relativa (HR). Los algoritmos de agrupamiento K-means y Fuzzy C-means son utilizados para encontrar relaciones existentes entre los contaminantes atmosféricos en estudio y las variables meteorológicas. Dichas relaciones aportan información a las RNA para obtener la predicción de los contaminantes, la cual es agregada como entrada de las RNA. Los resultados del modelo propuesto en este trabajo son comparados con los resultados de una Regresión Lineal Multivariable (RLM) y un Perceptrón Multicapa (MLP). La evaluación de la predicción se realiza con el Error Medio Absoluto, la Raíz del Error Cuadrático Medio, el coeficiente de correlación y el índice de acuerdo. Los resultados obtenidos muestran la importancia de las variables meteorológicas en la predicción de la concentración de los contaminantes SO2 y PM10 en la ciudad de Salamanca, Gto., México. Los resultados muestran que el MP predice mejor la concentración de los contaminantes SO2 y PM10 que los modelos RLM y MLP. Los modelos implementados para cada caseta de monitorizaci ón tienen la capacidad para realizar predicciones de calidad del aire, estos modelos pueden ser implementados en un sistema que permita realizar la predicción en tiempo real y analizar el impacto en la salud de la población. Entre los principales resultados obtenidos del desarrollo de esta tesis podemos citar: Se propone un modelo basado en una red neuronal artificial combinado con algoritmos de agrupamiento para la predicción con una hora de anticipaci ón de la concentración de cada contaminante (SO2 y PM10). Se diseñó un modelo diferente para cada contaminante y para cada una de las tres casetas de monitorización de la REDMAS. Se propone un modelo de predicción del promedio de la concentración de las próximas 24 horas de los contaminantes SO2 y PM10, basado en una red neuronal artificial combinado con algoritmos de agrupamiento. Se diseñó un modelo para cada caseta de monitorización de la REDMAS y para cada contaminante por separado.

A combined measure for quantifying and qualifying the topology preservation of growing self-organizing maps

The Self-OrganizingMap (SOM) is a neural network model that performs an ordered projection of a high dimensional input space in a low-dimensional topological structure. The process in which such mapping is formed is defined by the SOM algorithm, which is a competitive, unsupervised and nonparametric method, since it does not make any assumption about the input data distribution. The feature maps provided by this algorithm have been successfully applied for vector quantization, clustering and high dimensional data visualization processes. However, the initialization of the network topology and the selection of the SOM training parameters are two difficult tasks caused by the unknown distribution of the input signals. A misconfiguration of these parameters can generate a feature map of low-quality, so it is necessary to have some measure of the degree of adaptation of the SOM network to the input data model. The topologypreservation is the most common concept used to implement this measure. Several qualitative and quantitative methods have been proposed for measuring the degree of SOM topologypreservation, particularly using Kohonen's model. In this work, two methods for measuring the topologypreservation of the Growing Cell Structures (GCSs) model are proposed: the topographic function and the topology preserving map

Computing the Hessenberg matrix associated with a self-similar measure

We introduce in this paper a method to calculate the Hessenberg matrix of a sum of measures from the Hessenberg matrices of the component measures. Our method extends the spectral techniques used by G. Mantica to calculate the Jacobi matrix associated with a sum of measures from the Jacobi matrices of each of the measures. We apply this method to approximate the Hessenberg matrix associated with a self-similar measure and compare it with the result obtained by a former method for self-similar measures which uses a fixed point theorem for moment matrices. Results are given for a series of classical examples of self-similar measures. Finally, we also apply the method introduced in this paper to some examples of sums of (not self-similar) measures obtaining the exact value of the sections of the Hessenberg matrix.

A collaborative filtering similarity measure based on singularities.

Recommender systems play an important role in reducing the negative impact of informa- tion overload on those websites where users have the possibility of voting for their prefer- ences on items. The most normal technique for dealing with the recommendation mechanism is to use collaborative filtering, in which it is essential to discover the most similar users to whom you desire to make recommendations. The hypothesis of this paper is that the results obtained by applying traditional similarities measures can be improved by taking contextual information, drawn from the entire body of users, and using it to cal- culate the singularity which exists, for each item, in the votes cast by each pair of users that you wish to compare. As such, the greater the measure of singularity result between the votes cast by two given users, the greater the impact this will have on the similarity. The results, tested on the Movielens, Netflix and FilmAffinity databases, corroborate the excellent behaviour of the singularity measure proposed.

A Balanced Memory-Based Collaborative Filtering Similarity Measure.

Collaborative filtering recommender systems contribute to alleviating the problem of information overload that exists on the Internet as a result of the mass use of Web 2.0 applications. The use of an adequate similarity measure becomes a determining factor in the quality of the prediction and recommendation results of the recommender system, as well as in its performance. In this paper, we present a memory-based collaborative filtering similarity measure that provides extremely high-quality and balanced results; these results are complemented with a low processing time (high performance), similar to the one required to execute traditional similarity metrics. The experiments have been carried out on the MovieLens and Netflix databases, using a representative set of information retrieval quality measures.

Obtaining contradiction measure on intuitionistic fuzzy sets from fuzzy connectives

In a previous paper, we proposed an axiomatic model for measuring self-contradiction in the framework of Atanassov fuzzy sets. This way, contradiction measures that are semicontinuous and completely semicontinuous, from both below and above, were defined. Although some examples were given, the problem of finding families of functions satisfying the different axioms remained open. The purpose of this paper is to construct some families of contradiction measures firstly using continuous t-norms and t-conorms, and secondly by means of strong negations. In both cases, we study the properties that they satisfy. These families are then classified according the different kinds of measures presented in the above paper.

A methodology to measure the interface shear strength by means of the fiber push-in test

A methodology is presented to measure the fiber/matrix interface shear strength in composites. The strategy is based on performing a fiber push-in test at the central fiber of highly-packed fiber clusters with hexagonal symmetry which are often found in unidirectional composites with a high volume fraction of fibers. The mechanics of this test was analyzed in detail by means of three-dimensional finite element simulations. In particular, the influence of different parameters (interface shear strength, toughness and friction as well as fiber longitudinal elastic modulus and curing stresses) on the critical load at the onset of debonding was established. From the results of the numerical simulations, a simple relationship between the critical load and the interface shear strength is proposed. The methodology was validated in an unidirectional C/epoxy composite and the advantages and limitations of the proposed methodology are indicated.

Topological measure locating the effective crossover between segregation and integration in a modular network

We introduce an easily computable topological measure which locates the effective crossover between segregation and integration in a modular network. Segregation corresponds to the degree of network modularity, while integration is expressed in terms of the algebraic connectivity of an associated hypergraph. The rigorous treatment of the simplified case of cliques of equal size that are gradually rewired until they become completely merged, allows us to show that this topological crossover can be made to coincide with a dynamical crossover from cluster to global synchronization of a system of coupled phase oscillators. The dynamical crossover is signaled by a peak in the product of the measures of intracluster and global synchronization, which we propose as a dynamical measure of complexity. This quantity is much easier to compute than the entropy (of the average frequencies of the oscillators), and displays a behavior which closely mimics that of the dynamical complexity index based on the latter. The proposed topological measure simultaneously provides information on the dynamical behavior, sheds light on the interplay between modularity and total integration, and shows how this affects the capability of the network to perform both local and distributed dynamical tasks.

Dominance intensity measure within fuzzy weight oriented MAUT: an application

We introduce a dominance intensity measuring method to derive a ranking of alternatives to deal with incomplete information in multi-criteria decision-making problems on the basis of multi-attribute utility theory (MAUT) and fuzzy sets theory. We consider the situation where there is imprecision concerning decision-makers’ preferences, and imprecise weights are represented by trapezoidal fuzzy weights.The proposed method is based on the dominance values between pairs of alternatives. These values can be computed by linear programming, as an additive multi-attribute utility model is used to rate the alternatives. Dominance values are then transformed into dominance intensity measures, used to rank the alternatives under consideration. Distances between fuzzy numbers based on the generalization of the left and right fuzzy numbers are utilized to account for fuzzy weights. An example concerning the selection of intervention strategies to restore an aquatic ecosystem contaminated by radionuclides illustrates the approach. Monte Carlo simulation techniques have been used to show that the proposed method performs well for different imprecision levels in terms of a hit ratio and a rank-order correlation measure.

Safety and health in forest harvesting operations. Diagnosis and preventive actions. A review

Aim of study: to review the present state of the art in relation to the main labour risks and the most relevant results of recent studies evaluating the safety and health conditions of the forest harvesting work and better ways to reduce accidents. Area of study: It focuses mainly on developed Countries, where the general concern about work risks prevention, together with the complex idiosyncrasy of forest work in forest harvesting operations, has led to a growing interest from the forest scientific and technical community. Material and Methods: The main bibliographic and Internet references have been identified using common reference analysis tools. Their conclusions and recommendations have been comprehensively summarized. Main results: Collection of the principal references and their most important conclusions relating to the main accident risk factors, their causes and consequences, the means used towards their prevention, both instrumental as well as in the aspects of training and business management, besides the influence of the growing mechanization of logging operations on those risks. Research highlights: Accident risk is higher in forest harvesting than in most other work sectors, and the main risk factors such as experience, age, seasonality, training, protective equipment, mechanization degree, etc. have been identified and studied. The paper summarizes some relevant results, one of the principal being that the proper entrepreneurial risk management is a key factor leading to the success in minimizing labour risks..

Estudio del impacto del usuario final en los procesos de desarrollo de sistemas de información

Durante décadas y aun en la actualidad muchas organizaciones, a nivel mundial, continúan afrontando pérdidas significativas debido a fracasos parciales y totales respecto a sus inversiones en sistemas de información (SI), planteando serios retos a los niveles gerenciales y los profesionales de SI. Estadísticas alarmantes y décadas de experiencia en la praxis en el área de SI en diversas organizaciones llevan al autor a poner el énfasis en los usuarios finales internos (UF) que son designados como representantes (UFR) de sus pares en los proyectos de desarrollo de SI (PDSI) por considerarlos como factores influyentes de manera significativa en el problema. Particularmente, con enfoque en ciertos factores de los UFR críticos para el éxito de los PDSI, con dimensiones analizadas de forma aislada o incompleta en otros estudios empíricos, a la fecha. No se encontraron estudios en Latinoamérica ni en otras latitudes que abordasen el fenómeno del éxito/fracaso de los SI desde el punto de vista adoptado en esta tesis. Por ello, esta investigación empírica ha evaluado en qué grado estos factores pudiesen influenciar los resultados durante el desarrollo e implementación de SI y su posible impacto en la satisfacción de los UF, siendo esta última aceptada por variados autores como la principal medida del éxito de los SI. Este estudio fue realizado en América Latina en las cuatro grandes empresas industriales que integran verticalmente el sector aluminio de Venezuela, sometidas a un macro PDSI para instalar el paquete, de tipo ERP, SAP/R3. Experimentados profesionales fueron encuestados o entrevistados, tales como altos ejecutivos, desarrolladores, líderes de proyecto y líderes de los UF. Un enfoque metodológico de triangulación permitió combinar un análisis cuantitativo con un análisis cualitativo interpretativo del tipo hermenéutico/dialéctico, hallándose resultados convergentes y complementarios. Un análisis estadístico, utilizando Partial Least Squares (PLS), seguido de un análisis hermenéutico/dialéctico. Los resultados confirmaron un hecho importante: en los casos problemáticos, paradójicamente, los orígenes de las razones de rechazo de esos SI argumentadas por los UF, en alto grado, apuntaron a los UFR o a ellos mismos. Los resultados también confirmaron la prevalencia de factores de orden cognitivo, conductual y político en estas organizaciones sobre los tecnológicos, al igual que el alto riesgo de dar por sentado la presencia y calidad de los factores requeridos de los UFR y de los otros factores estudiados. La validación estadística del modelo propuesto reveló al constructo conocimientos de los UFR como la principal variable latente, con los variables indicadoras que componen este constructo ejerciendo la mayor influencia sobre la calidad y el éxito de los SI. Un hallazgo contrario al de otros estudios, mostró que los conocimientos sobre las tecnologías de la información (TI) fueron los menos relevantes. Los SI de nómina y de administración de los RRHH fueron los más problemáticos, como suele ser el caso, por su complejidad en organizaciones grandes. Las conclusiones principales confirman el decisivo rol de los UF para el éxito de los PDSI y su relación con la creciente problemática planteada, la cual amerita más investigación y de las organizaciones una mayor atención y preparación. Descuidar los factores humanos y sociales así como una efectiva planificación y gestión de los mismos en preparación para estos proyectos origina serios riesgos. No obstante las limitaciones de este trabajo, la problemática analizada suele influir en los PDSI en diversas organizaciones, indistintamente de su tamaño o tipo de SI, estimándose, por tanto, que los resultados, conclusiones y recomendaciones de esta investigación tienen un alto grado de generalización. Una relación de indicadores claves es suministrada con fines preventivos. Finalmente, los factores evaluados pueden usarse para ampliar el modelo reconocido de DeLone y McLean (2003), conectándolos como variables latentes de sus variables independientes calidad de la información y calidad del SI. ABSTRACT For decades, many organizations worldwide have been enduring heavy losses due to partial and total failures regarding their investments in information systems (IS), posing serious challenges to all management levels and IS practitioners. Alarming statistics in this regard and decades of practice in the IS area lead the author to place an emphasis on the end users (EU) who are appointed in representation of their peers (EUR) to IS development projects (ISDP), considering them as highly influential factors on the problem. Especially, focusing on certain EUR success factors, and their dimensions, deemed critical to any IS development and implementation, omitted or not thoroughly analyzed neither in the theory nor in the empirical research on the subject, so far. No studies were found in Latin America or elsewhere addressing the phenomenon of IS success/failure from the perspective presented herein. Hence, this empirical research has assessed to what degree such factors can influence the outcomes of an ISDP and their feasible impact on the EU´s satisfaction, being the latter accepted by several authors as the main measure of IS success. This study was performed in Latin America embracing four major industrial enterprises, which vertically integrate the aluminum sector of Venezuela, subjected to a macro ISDP to install the ERP-type package SAP/R3. The field work included surveying and interviewing experienced professionals such as IS executives, IS developers, IS project leaders and end-user project leaders. A triangulation methodological approach allowed combining quantitative and interpretive analyses, obtaining convergent and complementing results. A statistical analysis, using Partial Least Squares (PLS), was carried out followed by a hermeneutical/dialectical analysis. Results confirmed a major finding: in problematic cases, paradoxically, the origins of IS rejection reasons argued by the EU, at a high degree, were usually traceable to the EUR and themselves. The results also confirmed the prevalence of cognitive, behavioral and political factors in these organizations as well as the high risk of taking for granted the presence and quality of those factors demanded from the EUR. The statistical validation of the proposed model revealed the construct EUR knowledge as the main latent variable, with its items exerting a major influence on IS quality and success. Another finding, in contradiction with that of other studies, proved knowledge of information technology (IT) aspects to be irrelevant. The payroll and the human resources administration IS were the most problematic, as is usually the case in large companies. The main conclusions confirm the EU´s decisive role for IS success and their relationship with the problem, which continues, demanding more research and, from organizations, more attention and preparation. Neglecting human and social factors in organizations as well as their effective planning and management in preparation for ISDP poses serious risks. Despite the limitations of this work, the analyzed problem tends to influence ISDP in a wide range of organizations; regardless of their size or type of IS, thus showing a high degree of generalization. Therefore it is believed that the results, conclusions and suggestions of this research have a high degree of generalization. A detailed checklist comprising key measures is provided for preventive actions. Finally, the factors evaluated can be used to expand the well-known model of DeLone & McLean (2003), by connecting them as latent variables of its independent variables information quality and IS quality.

Categorical Structures in Non-Commutative Measure Theory.

In this paper, we investigate effect algebras and base normed spaces from the categorical point of view. We prove that the category of effect algebras is complete and cocomplete as well as the category of base normed spaces is complete, and discuss the contravariant functor from the category of effect algebras to the category of base normed spaces.

Methodologies to measure and to manage the decarbonisation of road transport: the case of Spain

El sector del transporte por carretera es uno de los principales contribuyentes de consumo de combustible y emisiones de España. Por lo tanto, la evaluación de los impactos ambientales del tráfico rodado es esencial para los programas de mitigación del cambio climático y la eficiencia energética. Sin embargo, uno de los retos en la planificación del transporte y el diseño de políticas consiste en la aplicación de metodologías de evaluación de emisiones consistentes, el diseño de estrategias y la evaluación de su eficacia. Las metodologías existentes de evaluación de las emisiones del transporte por carretera, utilizan diferentes niveles de análisis y períodos. Sin embargo, estos análisis son puntuales y no existe una continuidad en el análisis de diferentes estrategias o políticas. Esta tesis doctoral proporciona conocimientos y herramientas para el análisis de las políticas destinadas a reducir las emisiones de tráfico, tomando España como caso de estudio. La investigación se estructura en dos partes: i) el desarrollo y aplicación de metodologías para el análisis de factores y políticas que contribuyen en la evolución de las emisiones GEI del transporte por carretera en España; desde una perspectiva nacional; y ii) el desarrollo y aplicación de un marco metodológico para estimar las emisiones del tráfico interurbano y de evaluar estrategias centradas en la operación del tráfico y en la infraestructura. En resumen, esta tesis demuestra la idoneidad de utilizar diferentes herramientas para analizar las emisiones de tráfico desde diferentes puntos de vista. Desde el diseño de políticas de mitigación y eficiencia energética a nivel nacional, a estrategias centradas en la operación del tráfico interurbano y la infraestructura. Road transport is one of the major contributors to fuel consumption and emissions in Spain. Consequently, assessing the environmental impacts of road traffic is essential for climate change mitigation and energy efficiency programs. However, one of the key challenges of policy makers and transport planners consists of implementing consistent assessment emissions methodologies, applying mitigation strategies, and knowing their effectiveness. Current state-of-the-art emissions assessment methodologies estimate emissions from different levels and periods, using different approaches. Nevertheless, these studies are timely and they usually take different methodologies for analysing different strategies or policies, regardless of the assessment as a whole. This doctoral thesis provides knowledge and methodologies for analysing policies designed to reduce road traffic emissions, using the case study of Spain. The research procedure consists of two main scopes: i) the development and application of methodologies for analysing key factors and policies driving the GHG emissions of road transport in Spain; from a national perspective; and ii) the development and application of a road traffic emissions model for assessing operational and infrastructure strategies of the interurban road network at segment level. In summary, this thesis demonstrates the appropriateness to use different tools to analyse road traffic emissions at different levels: from appropriate nationwide mitigation and energy efficiency policies, to strategies focused on the operation of interurban traffic and infrastructure.

A CO2-saving-based methodology to measure the impact of the SUMP in European Cities: Application to the City of Burgos

Urban mobility in Europe is always a responsibility of the municipalities which propose measures to reduce CO2 emissions in terms of mobility aimed at reducing individual private transport (car). The European Commission's Action Plan on Urban Mobility calls for an increase in the take-up of Sustainable Urban Mobility Plans in Europe. SUMPs aim to create a sustainable urban transport system. Europe has got some long term initiatives and has been using some evaluation procedures, many of them through European projects. Nevertheless, the weak point with the SUMPs in Spain, has been the lack of concern about the evaluation and the effectiveness of the measures implemented in a SUMP. For this reason, it is difficult to know exactly whether or not the SUMPs have positively influenced in the modal split of the cities, and its contribution to reduce CO2 levels. The case of the City of Burgos is a very illustrative example as it developed a CiViTAS project during the years 2005-2009, with a total investment of 6M?. The results have been considered as ?very successful? even at European level. The modal split has changed considerably for better, The cost-effectiveness ratio of the SUMP in the city can be measured with the CO2 ton saved, specifically 36 ? per CO2 ton saved, which is fully satisfactory and in line with calculations from other European researchers. Additionally, the authors propose a single formula to measure the effectiveness of the activities developed under the umbrella of a SUMP.

Designing new cereal cultivars as an adaptation measure using crop model ensembles

To date, crop models have been little used for characterising the types of cultivars suited to a changed climate, though simulations of altered management (e.g. sowing) are often reported. However, in neither case are model uncertainties evaluated at the same time.