PLASMATIC: Unsupervised Machine Learning Big Data technologies and techniques

Abstract

PLASMATIC (Advanced Predictive Maintenance for the Valencian plastic industrial sector) is a project funded by the Valencian Institute for Business Competitiveness (IVACE) and the European Union through the European Regional Development Fund (FEDER). The general objective of this project is to help the Valencian plastic sector companies to incorporate solutions from the so-called Factory 4.0, via knowledge and technologies in the fields of Big Data, Machine Learning and Business Intelligence. The main result will be an advanced predictive maintenance system to deal with: (i) anomalies detection; (ii) wear prediction; and (iii) maintenance planning optimization. This deliverable is part of the PT3 work package "Characterization and Exploratory Analysis", which aims to identify those variables and factors involved in the advanced predictive maintenance applied to plastic industries. This document reviews the usual steps in a predictive maintenance application, describing the most important state-of-the-art techniques and technologies. In particular, we highlight statistical analysis and machine learning techniques within a Big Data infrastructure, all of them crucial elements of the so-called Maintenance 4.0. In general, we emphasize those possible strategies when working in a non-supervised application. ; PLASMATIC. Project funded by the Valencian Institute of Business Competitiveness (IVACE) and European Union through the European Regional Development Fund (ERDF), within the public grant program adressed to Technological Institutes of the Valencian Community for the development of non-economic R&D projects carried out in cooperation with companies during 2017 with 87.210,96€. File number: IMDEEA/2017/114