Sistema Regional de Información
en línea para Revistas Científicas de América Latina,
el Caribe, España y Portugal

In recent years, the Vendor Managed Inventory (VMI) system, in which the vendor manages its own and its retailers’ inventories, has been studied to improve the performance of two-echelon supply chains. However, most of these studies consider the pattern demands of the retailers as deterministic, which is very unlikely in practice where variability is significant. Thus, VMI systems based on deterministic demand patterns can lead to inefficient results, compromising the benefits of this system. Particularly within the pharmaceutical industry, an efficient supply chain through VMI is vital. The present work contributes within this context by proposing a multi-retailer VMI model to maximize the profits of a two-echelon supply chain in the presence of non-deterministic or uncertain demand. Due to the complexity of the model, a micro-genetic algorithm is developed to determine the lot size strategy to address the variable pattern of the non-deterministic demand within the profit function and reduce the stockout risk. The proposed model was validated through computer simulation, which is important to dynamically evaluate the performance of the model’s parameters. The dynamic evaluation showed that the proposed model is more efficient to reduce stockout events than models with consider deterministic demand patterns.  

Many real-world optimization problems involve changes related to the passage of time; this characteristic is known as dynamism. In this paper, we approach a dynamic multi-objective project portfolio selection problem with preferences. The objective of the general problem consists of determining the set of projects that optimize a vector of benefits considering the budget constraints. Both benefits and budgets are periodically changing, impacting the definition of the problem. Besides, the problem difficulty increases with the preferences of a decision-maker and more than one objective to satisfy. In this work, we present a new formulation of the described problem and a novel fuzzy method to incorporate the preferences of a decision maker. The method, called Fuzzy Filter (FF), uses fuzzy outranking relations to include controlled intensification and diversification to the solution process. For intensification, it keeps only non-dominated solutions that are in agreement with a decision maker. For diversification, it creates a nadir point from the filtered solutions and generates new solutions from this point. In order to provide an optimization benchmark of a real problem, instances with controlled difficulty were generated, and two algorithms of state of art were adapted to incorporate FF and dynamism. An analysis of extensive experimentation is presented as part of the benchmark.

Today it is crucial to have up-to-date information for companies to be more competitive in this business world. There are applications based on speech recognition that allows access to data stored in databases. However, the proper functioning of these applications lies in good pronunciation, a skill that most people do not have. In this paper, the architecture of an English mispronunciation detection module integrated into a chatbot is proposed. It allows users to enter the audio of the phrases in which they want to evaluate their pronunciation. The output is the mispronounced words, thus helping the user to practice their English language pronunciation. The proposed architecture consists of an Automatic Speech Recognizer (ASR) model based on a Transformer network that converts the audio signal to text and an algorithm for string alignment that identifies mispronounced words using the Levenshtein distance. The Transformer network was trained using the LibriSpeech and L2-ARTIC datasets. The module was evaluated using the Accuracy metrics, reaching 90%, and the Character Error Rate metric, reaching 9.5%. Additionally, its performance was evaluated on a group of real users, showing promising results.

The classification models of the states produced by the gelation tests of alkoxy benzoates require designing several corpora of data based on their characteristics. This work studies a series of alkoxybenzoates and 15 solvents characterized by Hansen Solubility Parameters and the number of carbons on the alkyl tail as a distinctive structural feature for the molecules. These properties were evaluated as attributes on the corpora on the kNN algorithm. Different configurations developed were analyzed, with three corpora designed varying their content according to their attributes. From this study, seem the relevance of some attributes over others on the performance prediction of the products class obtained. The significant samples correctly classified on corpora containing HSP and the number of carbons on the alkyl ether tail of alkoxybenzoates denote the influence of these properties on the classification. Also, the more suitable configurations on kNN, metric, k value, attribute weight is founded according to each corpus.

The use of hyper-heuristics to solve dynamic multi-objective optimization problems (DMOPs) that incorporate decision-maker's preferences is a recently addressed research area. This paper proposes the analysis and comparison of three hyper-heuristics to solve preferential DMOPs. The Dynamic Hyper-Heuristic with Plane Separation (DHH-PS), a previously proposed methodology using Plane Separation (PS), a reference-point-based preference incorporation method. This paper also proposes two versions of the Dynamic Population-Evolvability based Multi-objective Hyper-Heuristic (DPEM-HH), incorporating PS and different low-level heuristics sets. This work tests DHH-PS and both DPEM-HH-PS versions under multiple dynamic and preferential environments, seeking to extend the study of DHH-PS and analyze the capability of DPEM-HH-PS. DPEM-HH-PS exhibited suitability for type II DMOPs and randomly-changing instances. DHH-PS presented a better performance for tri-objective DMOPs. The combination of genetic algorithms and differential evolution in DPEM-HH-PS proved effective for solving preferential DMOPs. DHH-PS and DPEM-HH-PS were capable of adapting to different preferential and dynamic environments.

Globalization has changed the way companies do business in the world, always with the objective of having a greater presence in more countries through company purchases, strategic alliances, company acquisitions, investment projects. Companies also need to support decision-making to generate a strategy that considers the demands of consumers, the strategies of companies to consolidate their presence. The purpose of this study is to propose a system for decision making in presence of multiple criteria, to facilitate the generation of efficient strategic planning in the economic area of the company. The proposed approach is based on historical recognition of patterns and prediction of scenarios for multi-criteria ordinal classification using a recent classifier method

Smart Education is the influence of diverse technologies (Combinatorial Optimization, Machine Learning, Big Data, data visualisation, Internet of Education Things, Learning analytics, and others) to enhance the quality of education. In other words, Smart Education is the process of optimally managing human, economic and technological resources from educational institutions and research centres. Smart Education is part of the Smart Society (a component of a Smart City). The Smart Education components are the Internet of Education Things (Wireless Sensor Networks, RFID Technology), management of the education physical, infrastructure, smart classroom, smart campus, smart learning, Learning analytics, Smart Analysis, data science, Education Impact, and Educational Policy. This paper is a guide for understanding Smart Education components by presenting a survey of the characteristics, taxonomy (Education-Hard problems and Education-Soft problems), smart education indicators, history and future trends.

Academic failure is a situation that influences the terminal efficiency of students at a higher level. This is a consequence of various factors that affect the student. The objective is to identify factors that influence school failure in the face of the pandemic. A data collection model was applied for the analysis of factors that influence school failure based on the student's perception during the second year of the pandemic. The model considers technological, academic, and personal factors; The figure of the manager and the academic tutor are incorporated as elements that can influence school failure. The instrument was applied online, data analysis and reliability were performed with free software tools. The age range of the participants is 18 to 25 years old. 53.1% of the participants consider that the absence of the tutor in the face of failure problems influences failure. The factors that influence failure are the academic tutor, academic manager, internet access, the delivery of projects, the teacher's way of explaining, and health. The tutor and the academic manager have an important role as a guide in a situation of failure or academic performance.

The Timetabling Scheduling Problem (TTSP) is proposed as a schedule of a sequence of events between actors (teachers, students, workers, etc.) in a predefined period (typically hours), satisfying a set of constraints. TTSP has been traditionally considered in the operational research field and recently has been tackled with different Artificial Intelligence techniques. The proposed solutions to TTSP are in the range of traditional techniques (linear programming, whole programming, manual solution, network flow, etc.) and metaheuristic methods (simulations of human way, graph colouring, tabu search, genetic algorithms, simulated annealing, etc.). Job Shop Scheduling Problem (JSSP) is one of the best known combinatorial optimization NP-hard problems. There are many solutions to JSSP from a broad spectrum of researchers: management scientist, computational researcher, production experts, etc., from different individual areas and multidisciplinary areas. This article aims to model the TTSP in terms of JSSP in order to expand the possible solutions to this problem. We considered TTSP as JSSP because there are similarities at the mathematical model and the objective function. TTSP is modelled as JSSP where jobs represent the relation professor – signature – group and machines constitute the academic spaces.

A 3D color histogram is an image processing technique used to visualize the distribution of colors (Red-Blue-Green) in a picture. Because color distribution does not significantly change if a pictured object is translated or rotated, a 3D color histogram can be used as descriptor for automatic object recognition. However, this task requires cubes with high dimensionality. Within this context, the present work contributes with an approach to reduce the high dimensionality of the 3D color histogram and improve it as descriptor for object recognition. Tests performed with three databases (COIL-100, an own database, and CO3D) and three recognition systems corroborated its suitability for efficient object recognition, achieving overall recognition rates of 97.0% for objects with complex geometry and reflectance features. These results are more competitive when compared with other color descriptors as C-SIFT, RGB-SIFT, Color moments and RGB histograms.