Search

Building footprint maps are of high importance nowadays since a wide range of services relies on them to work. However, activities to keep these maps up-to-date are costly and time-consuming due to the great deal of human intervention required. Several automation attempts have been carried out in the last decade aiming at fully automatizing them. However, taking into account the complexity of the task and the current limitations of semantic segmentation deep learning models, the vast majority of approaches rely on aerial imagery (

This paper develops a Failure Mode, Effects and Criticality Analysis (FMECA) for a heating, ventilation and air conditioning (HVAC) system in railway. HVAC is a safety critical system which must ensure emergency ventilation in case of fire and in case of loss of primary ventilation functions. A study of the HVAC’s critical areas is mandatory to optimize its reliability and availability and consequently to guarantee a low operation and maintenance cost. The first part of the paper describes the FMECA which is performed and reported to highlight the main criticalities of the HVAC system under analysis. Secondly, the paper deals with the problem of the evaluation of a threshold risk value, which can distinguish negligible and critical failure modes. Literature barely considers the problem of an objective risk threshold estimation. Therefore, a new analytical method based on finite difference is introduced to find a univocal risk threshold value. The method is then tested on two Risk Priority Number datasets related to the same HVAC. The threshold obtained in both cases is a good tradeoff between the risk mitigation and the cost investment for the corrective actions required to mitigate the risk level. Finally, the threshold obtained with the proposed method is compared with the methods available in literature. The comparison shows that the proposed finite difference method is a well-structured technique, with a low computational cost. Furthermore, the proposed approach provides results in line with the literature, but it completely deletes the problem of subjectivity.

Nowadays, road maps play a key role in our society. Therefore, keeping those maps up-to-date is highly important. The extraction of road networks from satellite imagery is a complex problem, not only because of occlusions, shadows produced by non-road objects, but also due to the limited spatial resolution of the imagery used. The feasibility to detect a road depends on its width, which can reach sub-pixel size in some satellite products. In the last decade, many attempts have been carried out to automatize this labour. However, the vast majority of methods rely on aerial imagery, whose costs are not yet affordable for maintaining up-to-date maps. This work demonstrates that it is also possible to accurately detect roads using freely available Sentinel-2 imagery, regardless of their width. For that purpose, a new deep learning architecture which combines semantic segmentation and super-resolution techniques is proposed. As a result, fine-grained road network maps at 2.5 m are generated from 10 m imagery taken as input. To evaluate this proposal a data-set composed of 20 cities spread across the Spanish territory is used.

Learning good-performing classifiers from data with easily separable classes is not usually a difficult task for most of the algorithms. However, problems affecting classifier performance may arise when samples from different classes share similar characteristics or are overlapped, since the boundaries of each class may not be clearly defined. In order to address this problem, the majority of existing works in the literature propose to either adapt well-known algorithms to reduce the negative impact of overlapping or modify the original data by introducing/removing features which decrease the overlapping region. However, these approaches may present some drawbacks: the changes in specific algorithms may not be useful for other methods and modifying the original data can produce variable results depending on data characteristics and the technique used later. An unexplored and interesting research line to deal with the overlapping phenomenon consists of decomposing the problem into several binary subproblems to reduce its complexity, diminishing the negative effects of overlapping. Based on this novel idea in the field of overlapping data, this paper proposes the usage of the One-vs-One (OVO) strategy to alleviate the presence of overlapping, without modifying existing algorithms or data conformations as suggested by previous works. To test the suitability of the OVO approach with overlapping data, and due to the lack of proposals in the specialized literature, this research also introduces a novel scheme to artificially induce overlapping in real-world datasets, which enables us to simulate different types and levels of overlapping among the classes. The results obtained show that the methods using the OVO achieve better performances when considering data with overlapped classes than those dealing with all classes at the same time. The work of J. A. Sáez was supported by the Juan de la Cierva-formación Fellowship from the Spanish Ministry of Economy, Industry and Competitiveness under Grant FJCI-2015-25547. The work of M. Galar was supported in part by the Spanish Ministry of Economy, Industry and Competitiveness (MINECO), in part by the AEI/FEDER, and in part by the UE under Project TIN2016-77356-P. The work of B. Krawczyk was supported by the Polish National Science Center under Grant UMO-2015/19/B/ST6/01597.

Rail industry is rapidly developing, and rail becomes ever more viable in a wide range of regions. Therefore, the passenger experience and comfort has become a major concern for operators in the world. Heating, ventilation and air conditioning systems are used in railways to provide passengers thermal comfort and proper air motion. The ventilation system is one of the main elements of the system. Its components include both mechanical and electronic devices. All components are subjected to stress, and this tends to reduce their life cycle; the reliability of the ventilation system must be evaluated to plan and schedule appropriate maintenance activities. The paper evaluates reliability of the ventilation system using fault tree analysis and a reliability block diagram. Both techniques analyse data qualitatively; moreover, with specific algorithms they also provide quantitative results in term of reliability and probability of system failure. The paper compares the two reliability evaluation methods to verify their accuracy.

In the last years we have seen huge advancements in the area of Machine Learning, specially with the use of Deep Neural Networks. One of the most relevant examples is in image classification, where convolutional neural networks have shown to be a vital tool, hard to replace with any other techniques. Although aggregation functions, such as OWA operators, have been previously used on top of neural networks, usually to aggregate the outputs of different networks or systems (ensembles), in this paper we propose and explore a new way of using OWA aggregations in deep learning. We implement OWA aggregations as a new layer inside a convolutional neural network. These layers are used to learn additional order-based information from the feature maps of a certain layer, and then the newly generated information is used as a complement input for the following layers. We carry out several tests introducing the new layer in a VGG13-based reference network and show that this layer introduces new knowledge into the network without substantially increasing training times. This work was partially supported by the Public University of Navarre under the projects PJUPNA13 and PJUPNA1926 .

Data preprocessing is a key stage in data mining that allows machine learning algorithms to obtain meaningful insights. Many preprocessing problems such as feature selection or instance selection can be modelled as optimisation/search problems. Evolutionary algorithms have traditionally excelled in this task when dealing with data of a moderate size. However, their application to large datasets typically involves very high computational costs. In this work, we propose a hybrid surrogate model for evolutionary undersampling in imbalanced classification problems. These are characterised by having a highly skewed distribution of classes in which evolutionary algorithms aim to balance the training data by selecting only the most relevant data. The proposed technique combines a two-stage clustering-based surrogate method with a windowing approach to quickly approximate fitness values of the chromosomes and accelerate the search. The experiments carried out in 44 standard imbalanced datasets show that the proposed hybrid surrogate model highly reduces the computational cost of the evolutionary algorithm without a considerable loss of performance.

The tool-path problem has been extensively studied in manufacturing technologies, as it has a considerable impact on production time. Additive manufacturing is one of these technologies; it takes time to fabricate parts, so the selection of optimal tool-paths is critical. This research analyzes the tool-path problem in the direct energy deposition technology; it introduces the main processes, and analyzes the characteristics of tool-path problem. It explains the approaches applied in the literature to solve the problem; as these are mainly geometric approximations, they are far from optimal. Based on this analysis, this paper introduces a mathematical framework for direct energy deposition and a novel problem called sequence strategy generation. Finally, it solves the problem using a benchmark for several different parts. The results reveal that the approach can be applied to parts with different characteristics, and the solution to the sequence strategy problem can be used to generate tool-paths. This work was supported in part by the Project HARITIVE under Grant HAZITEK 2017 and in part by the Project ADDISEND under Grant ELKARTEK 2018 through Basque Government, and in part by the European Union Horizon 2020 Research and Innovation Programme under Grant 822064. The work of Roberto Santana was supported in part by IT-1244-19, in part by the ELKARTEK Programmes through Basque Government, and in part by the Spanish Ministry of Economy, Industry and Competitiveness under Grant TIN2016-78365-R.

The design of an ensemble of classifiers involves the definition of an aggregation mechanism that produces a single response obtained from the information provided by the classifiers. A specific aggregation methodology that has been studied in the literature is the use of fuzzy integrals, such as the Choquet or the Sugeno integral, where the associated fuzzy measure tries to represent the interaction existing between the classifiers of the ensemble. However, defining the big number of coefficients of a fuzzy measure is not a trivial task and therefore, many different algorithms have been proposed. These can be split into supervised and unsupervised, each class having different learning mechanisms and particularities. Since there is no clear knowledge about the correct method to be used, in this work we propose an experimental study for comparing the performance of eight different learning algorithms under the same framework of imbalanced dataset. Moreover, we also compare the specific fuzzy integral (Choquet or Sugeno) and their synergies with the different fuzzy measure construction methods. This work was partially supported by the Public University of Navarre under the projects PJUPNA13 and PJUPNA1926.

In Machine Learning an ensemble refers to the combination of several classifiers with the objective of improving the performance of every one of its counterparts. To design an ensemble two main aspects must be considered: how to create a diverse set of classifiers and how to combine their outputs. This work focuses on the latter task. More specifically, we focus on the usage of aggregation functions based on fuzzy measures, such as the Sugeno and Choquet integrals, since they allow to model the coalitions and interactions among the members of the ensemble. In this scenario the challenge is how to construct a fuzzy measure that models the relations among the members of the ensemble. We focus on unsupervised methods for fuzzy measure construction, review existing alternatives and categorize them depending on their features. Furthermore, we intend to address the weaknesses of previous alternatives by proposing a new construction method that obtains the fuzzy measure directly evaluating the performance of each possible subset of classifiers, which can be efficiently computed. To test the usefulness of the proposed fuzzy measure, we focus on the application of ensembles for imbalanced datasets. We consider a set of 66 imbalanced datasets and develop a complete experimental study comparing the reviewed methods and our proposal. This work was supported in part by the Spanish Ministry of Science and Technology (AEI/FEDER, UE) under Project TIN2016-77356-P, and in part by the Public University of Navarra under Project PJUPNA13.

Reliability analysis is widely used in many industrial fields to predict the remaining life of complex systems by assessing their current health status. This paper deals with one of the best-known techniques for reliability analysis: the reliability block diagram. This method models the reliability of the system based on the system’s architecture and the reliability of its components. The work analyses the control system of a 2MW wind turbine, proposing two different reliability models. The first draws on a standard control system architecture. The second introduces a cold standby redundancy architecture for the data acquisition subsystem and a warm standby redundancy architecture for the power supply subsystem. With these configurations, it is possible to improve the system reliability by neglecting some failure modes because one of the branches of the redundant configuration will be either inactive or partially active.

All devices are fabricated from materials that degrade with time. Degradation will continue until some critical device parameter can no longer meet the required specification for proper device functionality. Reliability estimation can assess the current health of a system and predict its remaining life. This kind of analysis is critical to improve safety, optimize scheduled maintenance, reduce life-cycle cost and minimize down time. This work analyses the reliability of a 2MW wind turbine using the Reliability Block Diagram method. The paper compares the reliability estimated in standard environmental conditions and the reliability evaluated considering the real temperature and humidity values acquired using a SCADA system.

The Era of Big Data has forced researchers to explore new distributed solutions for building fuzzy classifiers, which often introduce approximation errors or make strong assumptions to reduce computational and memory requirements. As a result, Big Data classifiers might be expected to be inferior to those designed for standard classification tasks (Small Data) in terms of accuracy and model complexity. To our knowledge, however, there is no empirical evidence to confirm such a conjecture yet. Here, we investigate the extent to which state-of-the-art fuzzy classifiers for Big Data sacrifice performance in favor of scalability. To this end, we carry out an empirical study that compares these classifiers with some of the best performing algorithms for Small Data. Assuming the latter were generally designed for maximizing performance without considering scalability issues, the results of this study provide some intuition around the tradeoff between performance and scalability achieved by current Big Data solutions. Our findings show that, although slightly inferior, Big Data classifiers are gradually catching up with state-of-the-art classifiers for Small data, suggesting that a unified learning algorithm for Big and Small Data might be possible., To appear in 2019 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE 2019)

Survival prediction systems are used among emergency services at hospitals in order to measure their quality objectively. In order to do so, the estimated mortality rate given by a prediction model is compared with the real rate of the hospital. Hence, the accuracy of the prediction system is a key factor as more reliable estimations can be obtained. Survival prediction systems are aimed at scoring the severity of patients' injuries. Afterward, this score is used to estimate whether the patient will survive or not. Luckily, the number of patients who survive their injuries is greater than that of those who die. However, this degree of imbalance implies a greater difficulty in learning the prediction models. The aim of this paper is to develop a new prediction system for the Hospital of Navarre with the goal of improving the prediction capabilities of the currently used models since it would imply having a more reliable measurement of its quality. In order to do so, we propose a new strategy to conform an ensemble of classifiers using an evolutionary under sampling process in the bagging methodology. The experimental study is carried out over 462 patients who were treated at the Hospital of Navarre. Our new ensemble approach is an appropriate tool to deal with this problem as it is able to outperform the currently used models by the staff of the hospital as well as several state-of-the-art ensemble approaches designed for imbalanced domains. This work was supported in part by the Spanish Ministry of Science and Technology under Project TIN2016-77356-P (AEI/FEDER, UE), in part by the Network Project under Grant TIN2014-56381-REDT, and in part by the Health Department of the Government of Navarre under Project PI-019/11.

Fuzzy Rule Based Classification Systems have the benefit of making possible to understand the decision of the classifier. Additionally, they have shown to be robust to solve complex problems. When these capabilities are applied to the context of Big Data, the benefits get multiplied.Therefore, to achieve the highest advantages of Fuzzy Rule Based Classification Systems, the output model must be both interpretable and accurate. The former is achieved by using fuzzy linguistic labels, that are related to human understanding. The latter is achieved by means of robust fuzzy rules, which are identified by means of a component known as "fuzzy rule weight". However, obtaining these rule weights is computationally expensive, resulting on a bottle-neck when applied in Big Data problems.In this work, we propose Chi-BD-SF, which stands for Chi Big Data Support Filtering. It comprises a scalable yet accurate fuzzy rule learning algorithm. It is based on the well-known Chi et al., exchanging the rule weight computation by a support metric in order to solve the conflicts between different consequent rules. In order to show the goodness of this proposal, we analyze several performance metrics, such as the quality of classification, the robustness of the rule base generated and the runtimes of the usage of traditional weights and the support of the rule. The results of our novel Chi-BD-SF approach, in contrast to related Big Data fuzzy classifiers, show that this proposal is able to out-speed the usage of rule weights also obtaining more accurate results.

In this work we study the possibility of learning fuzzy measures from classifiers’ performance for improving the standard aggregation methods in classifier ensembles. Fuzzy measures are set-valued functions, which are not necessarily additive, and they are the basis for constructing non-linear fuzzy integrals, such as Choquet or Sugeno integral. These integrals have shown to be very useful in the aggregation of interacting criteria, since this interaction can be well modeled by a fuzzy measure. Classifier ensembles are composed of several classifiers and are aimed at improving the performance of every one of their counterparts. There are two main aspects about ensembles, first, how to build them, and second, how to combine the outputs of all their members. In this work, we focus on the second part, which is a key factor to obtain a successful ensemble. More specifically, we focus on the usage of fuzzy measures for the aggregation phase aiming at taking into account the coalitions and interactions among the members of the ensemble. Our hypothesis is that taking such information into account can lead to better performance. Moreover, we propose to directly obtain the fuzzy measure from data by considering the performance of each subset of classifiers in the ensemble. This way, one needs not include any additional learning for the fuzzy measure that can easily lead to overfitting. In order to test the usefulness of the proposed fuzzy measure, we will consider a set of 33 highly imbalanced datasets and we will develop a complete experimental study comparing the proposed combination scheme with other approaches commonly considered in the literature.

Designing efficient learning models capable of dealing with tons of data has become a reality in the era of big data. However, the amount of available data is too much for traditional data mining techniques to be applicable. This issue is even more serious when evolutionary algorithms are a key part of the learning algorithm. In this scenario, one typical approach is to follow a divide-and-conquer strategy, where data is divided into different chunks that are individually and independently addressed. Afterwards, the partial knowledge obtained from each chunk of data is combined in order to give a solution to the problem. Nevertheless, these kinds of local approaches do not look at data as a whole, missing a global view of the problem, which may result in less accurate models that also depend on how data is split. In this work, we focus on evolutionary feature selection algorithms. A divide-and-conquer approach to handle evolutionary feature selection in big data was already developed. We aim at designing its global counterpart, which looks at the feature selection problem from a global perspective, making use of the data as a whole to select the most appropriate features. In order to do so, we consider Apache Spark as a big data technology where our algorithm is implemented. We design a genetic algorithm capable of dealing with big datasets by selecting the proper parameters for our base algorithm (the well-known CHC) and adapting the evaluation procedure to take all the distributed data into account. Several preliminary results are discussed to study the feasibility of global evolutionary feature selection methods for big datasets.

The main drawback of Fuzzy Rule-Based Classification Systems (FRBCSs) when they are applied in Big Data problems is the lack of scalability. Previously proposed approaches consist in concurrently fitting several Chi et al. FRBCSs whose rule bases are then aggregated to obtain the final model. This methodology is seriously affected by the degree of parallelism used for the execution of the algorithm, showing a significant decrease in classification performance as the degree of parallelism increases. This work focuses on the design of a new FRBCS for Big Data classification problems (CHI-BD) that generates exactly the same rule base regardless of the degree of parallelism. Our approach recovers the model that would be built by the original Chi et al. algorithm if it was able to deal with Big Data problems.

Performance-based Logistics (PBL) contracts require metrics and methodologies set in a systemic way to provide readiness for the warfighter at reasonable costs. There must be a clever and verifiable connection of metrics in order to access and analyze data and to deliver sound and consistent inputs for the entire support system to accomplish its goals. A system of incentives and penalties usually takes place and suggests functions to be maximized and/or minimized in a multi-criterion and highly integrated environment. This study deals with the optimization of the entire setup of Performance-Based Logistics contracts and is limited to a preliminary study in a simple scenario that has showed a promises results. A mixed method, using both the £-Constraint and Goal Programming is proposed to model the case. Results indicate that metrics used for reliability, maintainability, availability and supportability (RAMS) and costs can be optimized simultaneously for clever contracts.

Decomposition strategies have been shown to be a successful methodology to tackle multi-class classification problems. Among them, One-vs-One approach is a commonly used technique that consists in dividing the original multi-class problem into easier-to-solve binary sub-problems considering each possible pair of classes. However, this methodology is limited to those classifiers returning a single real value for each prediction. In this work, we present a new One-vs-One approach that is able to deal with interval-valued outputs. In order to achieve this goal, we propose applying a normalization method for intervals along with the corresponding extension of three different aggregation strategies: voting, weighted voting, and WinWV. The experimental results show the suitability of the normalization method and the improvement obtained by One-vs-One with respect to a state-of-the-art interval-valued Fuzzy Rule-Based Classification System (IVTURS).

In this work we present an optimization of the only Fuzzy Rule-Based Classification System that is able to face Big Data classification problems to date, i.e., Chi-FRBCS-BigDataCS. The aim of this optimization is to speed up the learning process of the algorithm without affecting the model obtained. Our proposal is based on the usage of Look-Up-Tables to pre-compute the membership degrees of the different linguistic labels, avoiding millions of computations. Furthermore, we propose a new data flow that: 1) takes advantage of the sorting and merging phases of MapReduce to group duplicated/conflicting rules and avoids an exhaustive search over the rule base; 2) removes the bottleneck caused by the execution of a single reducer by supporting as many reducers as needed. These adaptations have shown a significant improvement in terms of runtime without altering the model obtained by Chi-FRBCS-BigDataCS.

In avionics application one of the most important competition factors is the reliability, given that the failure occurrence may leads to a critical state for the functioning of the aircraft. Different maintenance, prognostics and diagnostics approaches are possible with the final aim to optimize both system's availability and safety. Aircraft engines represent a safety critical part of the airplane. For this reason it is a key issue to allocate the proper amount of maintenance at each individual maintenance event. In this paper a mathematical replacement model is proposed to guarantee that the correct amount of maintenance is performed.