Within the multi -objective (static) optimization field, various works related to the adaptive selection of genetic operators can be found. These include multiarmed bandit -based methods and probability -based methods. For dynamic multi -objective optimization, finding this type of work is very difficult. The main characteristic of dynamic multi -objective optimization is that its problems do not remain static over time; on the contrary, its objective functions and constraints change over time. Adaptive operator selection is responsible for selecting the best variation operator at a given time within a multi -objective evolutionary algorithm process. This work proposes incorporating a new adaptive operator selection method into a Dynamic Multiobjective Evolutionary Algorithm Based on Decomposition algorithm, which we call DMOEA/D-SL. This new adaptive operator selection method is based on a reinforcement learning algorithm called StateAction -Reward -State -Action Lambda or SARSA (A). SARSA Lambda trains an Agent in an environment to make sequential decisions and learn to maximize an accumulated reward over time; in this case, select the best operator at a given moment. Eight dynamic multiobjective benchmark problems have been used to evaluate algorithm performance as test instances. Each problem produces five Pareto fronts. Three metrics were used: Inverted Generational Distance, Generalized Spread, and Hypervolume. The non -parametric statistical test of Wilcoxon was applied with a statistical significance level of 5% to validate the results.

The multi-objective portfolio optimization problem with fuzzy trapezoidal parameters involves a search for a subset of projects that, within the given available resources, maximizes the benefits while reducing the uncer-tainty. Traditionally, evolutionary algorithms are used to solve this problem; however, they do not exploit the locality structure of a solution or compute the values of its objective functions using the full set of n decision variables. As a result, the number of evaluations that can be computed within a fixed amount of time decreases as the size of the instances increases, yielding poor performance. This work proposes a new non-evolutionary GRASP/& UDelta; algorithm that includes a novel local search with an efficient local computation strategy. The use of local computation reduces the number of operations required to compute the values of the objective functions from O(n) to O(1). Consequently, the increment in evaluations performed in the proposed approach increases the quality of the obtained solutions, particularly as the search space grows. An experiment conducted with instances of different sizes demonstrates the overall competitiveness of GRASP/& UDelta; compared to other state-of-the-art al-gorithms. Our results show, as expected, that the differences in performance become statistically more significant when dealing with instances defined by large search spaces. These results were validated using non-parametric statistical tests.

The Mixed No-Idle Permutation Flow Shop Scheduling Problem MNPFSSP pertains to the family of Regular Flow Scheduling Problems. In this problem, some machines allow idle time (existence of time between two consecutive jobs) and others do not; this is the general case of the No-Idle Permutation Flow Shop Scheduling Problem NPFSSP. The goal is to find the optimal sequence of jobs that minimizes the time interval in which all jobs to be scheduled are processed; known as makespan. This paper presents the performance of a Distribution Estimation Algorithm (EDA) based on the Generalized Mallows Model (GMM) in which an intuitive modification to the learning model is established by experimenting with changes in the position and dispersion parameters. To test the performance of the proposal 27 and 30 small and large instances respectively were generated from a particular instance selected from the site http://soa.iti.es . In the tests, the classical EDA-GMM algorithm taken from the state of the art and the one proposed were compared when applying the different combinations of the parameters in the learning model. The general experimental results show that when applying the change of position and dispersion parameters simultaneously, quality solutions are produced at the same or less time than those obtained in the rest of the cases. The results allow continuing the refinement and support of the proposal to be applied in a production model to the Mixed No-Idle Per-mutation Flow Shop Scheduling Problem.

One of the main conflicts in a car production plant is to deliver the orders received daily in a timely manner, which are not uniform and involve a large amount of human and material resources. The car sequencing problem is a NP-Hard problem that consists of finding the sequence of cars that minimizes the number of constraint violations in an assembly line. The problem can be approached from a mono-objective or multi-objective point of view. The objective of this paper is to treat a case study of this problem, presented at ROADEF 2005, from the multi-objective Pareto approach, taking the NSGAII algorithm as a basis for a proposal scheme and verifying its feasibility. A systematic and general improvement of the quality of the final Pareto fronts is verified, and the results of the implementation of a strategy scheme that consists of the initialization of the population guided by local search, and specialized crossover and mutation operators are reported. These results allow us to give continuity to the generation of an optimization proposal for the vehicle sequencing problem.

We present a device designed to provide assistance to an older adult in their home. Conceptually, this assistant is conceived as a multi-agent system designed to monitor and regulate itself according to the mental engagement of the user and contribute to help them maintain adequate levels. Physically, the assistant is realized as a desk, equipped with an automated pill dispenser and interconnected with other devices intended to provide security and comfort to the user. Through simulations it was possible to confirm that the system performs as desired for diverse scenarios. This conclusion encourages us to proceed with the next stage of development, where real data will be collected employing the prototype of the system described herein.

The main objective of an automobile production plant is to deliver on time and form the orders that are received daily. These orders are not homogeneous since they involve large quantities of cars that generally belong to different models and must be painted in different colors. The car sequencing problem that takes these characteristics into account was proposed by the Renault Company in 2005 as part of the ROADEF Challenge. This problem is NP-Hard and various techniques have been proposed to solve it, from exact methods to different heuristic algorithms. This work presents a feasibility study to apply two Distribution Estimation Algorithms (EDAs) to solve this problem. In addition, three important aspects are presented: the adaptation process of the algorithms, a technique for the execution of the algorithms called the "Stepped Approach with Discard" and a methodology that involves tolerance in the substitution of the individuals. The results obtained by the algorithms are also shown. The analysis of the results shows the algorithms adaptation process and the adjustments that can be made to improve their competence with the state of art.

Resumen.En el artículo se propone un algoritmo de Evolución Diferencial con Reparador Cromosómico (EDRC) aplicado a la secuenciación de vehículos, que consiste en encontrar una secuencia de producción de diferentes modelos de automóviles en una línea de ensamblaje.Este es un problema de satisfacción de restricciones multiobjetivo NP-Duro [1], en el que se busca violar la menor cantidad de restricciones.Para generar la población inicial se utiliza un operador de mutación basado en el cambio.Además, se propone un reparador cromosómico que toma en cuenta las características del problema y asegura la generación de individuos factibles.Las soluciones del EDRC fueron comparadas con los resultados reportados por el algoritmo de recocido simulado usado por la Renault [2] y el equipo que aplicó búsqueda tabú y búsqueda codiciosa greedy [3], mostrando competencia (39% de los casos), mejora (22% de los casos), no logrando competir en el 39% de los casos.En la etapa que dará continuidad al proyecto se analizarán operadores

Fuzzy logic systems provide a set of proven tools and methods to imitate or emulate human basic reasoning, that is, transform it into instructions that the computer can understand or transform into binary instructions. Based on the structure with multiple layers, subsystems and varied topologies that in previous research have shown that fuzzy hierarchical systems have been used to improve the interpretability, in this research work the objective is to design a fuzzy hierarchical system using fuzzy composite concepts artificial intelligence compounds to measure the efficiency of simulated scenarios. As a fundamental part of the present investigation, an analysis is made of the sensitivity of the results of the fuzzy system with respect to its inputs and with a set of membership functions, in a virtual scenario; which allows demonstrating the advantages obtained by applying a fuzzy hierarchical system to systems oriented to the area of health.

Smooth particle hydrodynamics (SPH) is a mesh free numerical method for solving hydrodynamical equations. For its functioning, the method uses; one integer-domain parameter (the total number of particles) and three real domain parameters (smoothing parameters and artificial viscosity). For a given problem (geometry and initial conditions) these parameters can be tuned to reduce the computational cost and improve the accuracy of the solutions. Optimized values of the SPH parameters using the evolutionary algorithms, Differential Evolution (DE) and Boltzmann Univariate Marginal Distribution Algorithm (BUMDA) are obtained for different Sod shock tube test problems. Comparison of the numerical solution of the physical variables with that of the exact solution shows that this optimization strategy can be used to make an initial guess of the SPH parameters based on the initial conditions of the simulation domain. The performance of the two algorithms are statistically compared.

Resumen.En el presente trabajo se muestran los resultados de los descriptores construidos mediante una metodología basada en Gramática Evolutiva.Se hace la comparación de dos parámetros de la propuesta, uno es el uso de los canales de color y la escala de grises en el proceso de construcción de los descriptores, lo cual se llevó a cabo utilizando dos gramáticas tipo Backus-Naur en el proceso evolutivo.La primera gramática usa los canales de color rojo, azul, verde y la escala de grises, mientras que la segunda gramática solo usa la escala de grises.El otro parámetro es la forma de representar los vectores de características de las imágenes procesadas con los descriptores construidos, comparando dos alternativas: el histograma y los estadísticos del histograma.Los descriptores construidos fueron aplicados a la clasicación de imágenes de texturas de piezas arqueológicas procedentes de la

Texture classification is a problem widely studied in computer vision, there exist two fundamental issues: how to describe texture images and how to define a similarity measure. The texture descriptors are mainly used to extract and represent the features of texture images and their performance is usually measured using a classification algorithm. In this paper, some of the most referenced texture descriptors, such as Gabor filter banks, Wavelets, and Local Binary Patterns, are compared using non-parametric statistical tests to know if there is a difference in performance. The descriptors are applied to five well-known texture image datasets, in order to be classified. Three classification algorithms, with a cross-validation scheme, are used to classify the described texture datasets. Finally, a Friedman test with multiple comparisons is used to compare the whole performance of the texture descriptors on a statistical basis. The statistical results suggest that for these tests there is a difference in performance, so it was possible to determine statistically, for the considered experimental settings, the best texture descriptor.

Evolutionary Artificial Neural Networks (EANNs) are a special case of Artificial Neural Networks (ANNs) for which Evolutionary Algorithms (EAs) are used to modify or create them. EANNs adapt their defining components ad hoc for solving a particular problem with little or no intervention of human expert. Grammatical Evolution (GE) is an EA that has been used to indirectly develop ANNs, among other design problems. This is achieved by means of three elements: a Context-Free Grammar (CFG) which includes the ANNs defining components, a search engine that drives the search process and a mapping process. The last component is a heuristic for transforming each GE's individual from its genotypic form into its phenotypic form (a functional ANN). Several heuristics have been proposed as mapping processes in the literature; each of them may transform a specific individual's genotypic form into a very different phenotypic form. In this paper, partially-connected ANNs are automatically developed by means of GE. A CFG is proposed to define the topologies, a Genetic Algorithm (GA) is the search engine and three mapping processes are tested for this task; six well-known pattern recognition benchmarks are used to statistically compare them. The aim of this work for using and comparing different mapping process is to analyze them for setting the basis of a generic framework to automatically create ANNs.

Grammatical Evolution (GE) is a grammar-based form of Genetic Programming. In GE, a Mapping Process (MP) and a Backus-Naur Form grammar (defined in the problem context) are used to transform each individual's genotype into its phenotype form (functional representation). There are several MPs proposed in the state-of-the-art, each of them defines how the individual's genes are used to build its phenotype form. This paper compares two MPs: the Depth-First standard map and the Position Independent Grammatical Evolution (pGE). The comparison was performed using as use case the problem of the selection and generation of features for pattern recognition problems. A Wilcoxon Rank-Sum test was used to compare and validate the results of the different approaches.

In the recent years, Grammatical Evolution (GE) has been used as a representation of Genetic Programming (GP). GE can use a diversity of search strategies including Swarm Intelligence (SI). Bee Swarm Optimization (BSO) is part of SI and it tries to solve the main problems of the Particle Swarm Optimization (PSO): the premature convergence and the poor diversity. In this paper we propose using BSO as part of GE as strategies to generate heuristics that solve the Bin Packing Problem (BPP). A comparison between BSO, PSO, and BPP heuristics is performed through the nonparametric Friedman test. The main contribution of this paper is to propose a way to implement different algorithms as search strategy in GE. In this paper, it is proposed that the BSO obtains better results than the ones obtained by PSO, also there is a grammar proposed to generate online and offline heuristics to improve the heuristics generated by other grammars and humans.

En este trabajo se muestra el comportamiento sinérgico que se produce en la implementación de una Hiperheurística de selección aplicada al problema del agente viajero (TSP, por sus siglas en inglés). Como órgano rector de la Hiperheurística se utilizó un Algoritmo Genético, y un conjunto de 5 heurísticas de bajo nivel. Para hacer las pruebas se utilizaron instancias de entrenamiento del estado del arte para TSP, y para el análisis de resultados, se hizo una comparación del mejor genotipo obtenido del entrenamiento de la combinación de las heurísticas, contra genotipos que contienen un solo tipo de heurística analizados desde un enfoque de optimización. En las pruebas estadísticas se utilizó como representante estadístico la mediana obtenida de dichos experimentos. Se presentan la explicación del entrenamiento fuera de línea de la Hiperheurística y los resultados que muestran que la hiperheurística es capaz de mejorar los resultados de las heurísticas aplicadas individualmente.

El Problema del agente viajero (TSP) es un problema de optimizacion combinatoria muy estudiado en el area de computacion cientifica y matematicas aplicadas. La importancia del TSP radica en que varios problemas de optimizacion combinatoria se pueden formular con base en el. Hasta la fecha no se ha encontrado un algoritmo deterministico que resuelva el TSP en un tiempo polinomial. En el estado del arte se han reportado soluciones factibles en tiempo polinomial, mediante el uso de algoritmos no deterministicos conocidos como Metaheuristicas. En este trabajo se implementaron: el Algoritmo Genetico (AG), Algoritmo Memetico (AM), y el Algoritmo de Sistema Inmune (ASI) para resolver el TSP Simetrico. Los algoritmos implementados pertenecen a una familia de Metaheuristicas conocida como Algoritmos Evolutivos los cuales estan inspirados en la evolucion natural. Para identificar el desempeno de los algoritmos seleccionados, se realizo una comparacion entre ellos haciendo uso de estadistica no parametrica para evidenciar el algoritmo con mejor desempeno para resolver el TSP.

In this paper, the NP-hard problem of minimizing power consumption in wireless communications systems is approached. In the literature, several metaheuristic approaches have been proposed to solve it. Currently a homogeneous cellular processing algorithm and a GRASP algorithm hybridized with path-relinking are considered the state of the art algorithms. The main contribution of this paper is the analysis of five main characteristics for a heterogeneous cellular processing algorithm, based on scatter search and GRASP. A series of computational experiments with standard instances were carried out to assess the impact of each one of these characteristics. Among the main analyses we found particularly interesting a time reduction by 74.24 %, produced by the stagnation detection characteristic. Also the communication characteristic improves the quality of the solutions by 24.73 %. The computational results show that our heterogeneous cellular processing algorithm is a good alternative for solving the problem. The proposed algorithm finds 34 new best known solutions, which is 27 % of the instances with unknown optimal values. A Friedman hypothesis test was carried out to validate that two state-of-the-art algorithms and the proposed algorithm are statistically equivalent.