Abstract

In this work, data-driven soft sensors are developed for the debutanizer column for online monitoring of butane content in the debutanizer column bottom product. The data set consists of data for seven process inputs and one process output. The total process data were equally divided into a training set and a validation set using the Kennard–Stone maximal intra distance criterion. The training set was used to develop multiple linear regression, principal component regression and back propagation neural network models for the debutanizer column. Performances of the developed models were assessed by simulation with the validation data set. Results show that the neural network model designed using Levenberg–Marquardt algorithm is capable of estimating the product quality with nearly 95% accuracy. The performance of the neural network model reported in this article is found to be better than the performances of least square support vector regression and standard support vector regression models reported in the literature earlier.

Abstract

This work provides an unsupervised learning approach based on a single-valued performance indicator to monitor the global behavior of critical components in a viaduct, such as bearings. We propose an outlier detection method for longitudinal displacements to assess the behavior of a singular asymmetric prestressed concrete structure with a 120 m high central pier acting as a fixed point. We first show that the available long-term horizontal displacement measurements recorded during the undamaged state exhibit strong correlations at the different locations of the bearings. Thus, we combine measurements from four sensors to design a robust performance indicator that is only weakly affected by temperature variations after the application of principal component analysis. We validate the method and show its efficiency against false positives and negatives using several metrics: accuracy, precision, recall, and F1 score. Due to its unsupervised learning scope, the proposed technique is intended to serve as a real-time supervision tool that complements maintenance inspections. It aims to provide support for the prioritization and postponement of maintenance actions in bridge management. Authors would like to acknowledge the discussions with Marcos Pantaleón from APIA XXI, Ambher Monitoring Systems and Banobras S.N.C. This work has received funding from the European’s Union Horizon 2020 research and innovation program under the grant agreement No 690660 (RAGTIME Project) and No 769373 (FORESEE Project). This paper reflects only the author’s views. The European Commission and INEA are not responsible for any use that may be made of the information contained therein. David Pardo has received funding from the European Union’s Horizon 2020 research and innovation program under the Marie Sklodowska-Curie grant agreement No 777778 (MATHROCKS), the European POCTEFA 2014-2020 Project PIXIL (EFA362/19) by the European Regional Development Fund (ERDF) through the Interreg V-A Spain-France-Andorra program, the Project of the Spanish Ministry of Science and Innovation with reference PID2019-108111RBI00 (FEDER/AEI), the BCAM “Severo Ochoa” accreditation of excellence (SEV-2017-0718), and the Basque Government through the BERC 2018-2021 program, the two Elkartek projects 3KIA (KK-2020/00049) and MATHEO (KK-2019-00085), the grant "Artificial Intelligence in BCAM number EXP. 2019/00432", and the Consolidated Research Group MATHMODE (IT1294-19) given by the Department of Education. Document type: Article

Abstract

The current labor market demands new qualities and knowledge from the recent graduates of universities. This has been called ‘21st century digital skills’ and there is not enough research on the self-perception of students in this regard. The objective of this research was to measure student self-perception about their own 21st century digital skills related to the use of information and communication technologies (ICT) in Higher Education. A questionnaire was generated and applied to 356 students with the stratified random sampling technique. A principal component analysis was carried out, supported by adequate values of the Kaiser-Meyer-Olkin coefficient and the Bartlett sphericity test. The data indicate that the students use digital technology in academic projects, primarily; they have a lot of skill when using ICT for information management, to develop critical thinking and to solve problems, as well as to manage mobile devices. However, their self-perception in the use of ICT in teaching classes is low. The results suggest that the students do not believe that the use of ICT in the classroom is useful to develop this kind of emerging digital skills. Instead, they think that carrying out academic projects does strengthen the acquisition and development of such skills in relation to the use of ICT.

Abstract

International audience; In Functional Data Analysis, the underlying structure of a raw observation is functional and data are assumed to be sample paths from a single stochastic process. When data considered are functional in nature thus infinite-dimensional, like curves or images, the multivariate statistical procedures have to be generalized to the infinite-dimensional case. By approximating random functions by a finite number of random score vectors, the Principal Component Analysis approach appears as a dimension reduction technique and offers a visual tool to assess the dominant modes of variation, pattern of interest, clusters in the data and outlier detection. A functional statistics approach is applied to univariate and multivariate aircraft trajectories.

Abstract

International audience; This paper presents an original method to evaluate air traffic complexity metrics. In previous works, we applied a principal component analysis (PCA) to find the correlations among a set of 27 complexity indicators found in the literature. Neural networks were then used to find a relationship between the components and the actual airspace sector configurations. Assuming that the decisions to group or split sectors are somewhat related to the controllers workload, this method allowed us to identify which components were significantly related to the actual workload. We now focus on the subset of complexity indicators issued from these components, and use neural networks to find a simple relationship between these indicators and the sector status.

Abstract

Governments worldwide face the difficult challenge of deciding which infrastructure projects to prioritize and select for implementation, given the limits of available funding and the need to attain their developmental goals. The key objective of this report is to conduct a comparative exercise between the World Bank's Infrastructure Prioritization Framework, a multicriteria analysis–based methodology to project prioritization, and a more complex cost-benefit analysis–based approach. The report focuses on Chile, which has a well-institutionalized evaluation process that uses cost-benefit analysis to assess projects on their quality and ability to generate value for money. The analysis compares the results of the Infrastructure Prioritization Framework alongside Chile's current cost-benefit analysis–based and multicriteria analysis approaches to the same subsets of projects in the road transport and water reservoir subsectors, respectively. The results show that the Infrastructure Prioritization Framework has application beyond its original proposition and can complement a traditional cost-benefit analysis by directly considering social and environmental policy goals that are otherwise difficult to quantify in a cost-benefit analysis. The analysis also finds that in Chile there is a discrepancy between the stated goals and objectives of the appraisal system and the actual implementation. In the case of transport sector projects, there is an evident deviation between cost-benefit analysis–based selection policy and actual decisions made for project implementation. In the case of water catchment selection, there is a bias toward projects with higher financial-economic performance as compared to social-environmental performance, despite policy intentions to afford consideration to environmental and social development goals. Document type: Book

Abstract

Increased rail competitiveness has been the objective of many countries around the world, including member states of the EU. Although railway market liberalization has always been accompanied by high expectations of increased efficiency and competitiveness, the overall impact of such decisions can be considered controversial. This paper aims to contribute to the scientific debate by conducting a factor analysis of some East-Central European countries from the aspect of rail freight competitiveness. Since many highly correlated factors influence competitiveness, its mathematical&ndash statistical representation and analysis is difficult due to the high number of dimensions of the factor space. Moreover, competitiveness cannot be measured directly only as a latent variable which is a feature of Principal Component Analysis (PCA). The introduced PCA, model by way of reducing the number of dimensions, can highlight the relations among the attributes and determine the most crucial issues capable of increasing rail competitiveness in the given countries and also of clustering those national railway markets. Recommendations for structural changes in national rail freight markets of the region are also supplied. Our results show that international rail competitiveness depends rather on market efficiency than on market liberalization due to the fact that the Global Competitiveness Index and Export/Import attributes did not significantly correlate with market concentration. As for the larger domestic rail freight sectors, small freight forwarders&mdash spawned by liberalization&mdash are shown to play a significant role in increasing competitiveness. Document type: Article