Skip to main content

Short Course: Machine Learning for Predictive Maintenance


Continuous training programs/short-courses aimed at expanding the current offer of HEIs in terms of lifelong learning courses and providing innovative training paths to adult learners, interested in re/upskilling their knowledge and skills in topics related to industry 4.0 (same as listed above), better equipping them to undertake or keep pace with organizational/job-related changes. These courses will include a practical component, comprising hands-on projects.

Short Course Title

Machine Learning and Data Analytics to Predictive Maintenance

Short Course Date

February 2020 31.1, 1.2, 7.2, 8.2, 14.2, 15.2, 21.2, 22.2

Friday: From 18 p.m. to 21 p.m.

Saturday: From 10 p.m.

Purpose And Motivation

Competitiveness is more crucial than ever in the current economic landscape, affecting a company's ability to deliver quality products at low prices. Due to their potential to substantially reduce production costs and increase productivity, proactive maintenance practices, such as predictive maintenance, have a profound impact on business competitiveness. With predictive maintenance, maintenance interventions are performed only when necessary and before equipment failure occurs, avoiding potentially catastrophic breakdowns and unnecessary interruptions of production. Thanks to the current trend of automation and data exchange in industrial environments, assisted by the rise of the Internet of Things, a large amount of operational data is now available, or can be acquired with relative ease. Machine Learning and Data Mining techniques can be used to extract knowledge from this data and support a company's decision making to improve its operations and competitiveness.


At the end of this course, trainees should have a deeper understanding of Predictive Maintenance and its advantages, as well as knowledge of the process and tools required for its implementation.

Learning Outcomes

Trainees are expected to:

  • Be able of recognizing the added value of Predictive Maintenance;

  • Understand the basics of Machine Learning;

  • Be familiar with the process of predicting failures and with some of the algorithms that can be used;

  • Know some of the most commonly used intelligent data analysis tools.


This course is destinated to directors of Maintenance/manufacturing.

Short Course Description

Session 1 - Introduction / Data Analysis with R (3h)

  • Course objectives

  • Introduction to Predictive Maintenance

  • Installation and configuration of the R development environment

Session 2 - Introduction / Data Analysis with R (3h)

  • Installation and configuration of the R development environment (continued)

  • Essential R concepts

Session 3 – Machine Learning / Data Exploration(3h)

  • Introduction to Machine Learning

  • Exploratory analysis of labelled data

Session 4 - Supervised Learning (3h)

  • Supervised learning algorithms in the context of Predictive Maintenance

Session 5 - Supervised Learning (3h)

  • Supervised learning algorithms in the context of Predictive Maintenance (continued)

Session 6 – Data Exploration (3h)

  • Exploratory analysis of unlabelled data

Session 7 - Unsupervised Learning (3h)

  • Unsupervised learning algorithms in the context of Predictive Maintenance

Session 8 - Unsupervised Learning (3h)

  • Unsupervised learning algorithms in the context of Predictive Maintenance (continued)

Course Staff

Course Staff Image #1

Marta Fernandes

Researcher (GECAD – ISEP/P.Porto)

Assistant Professor (ISEP/P.Porto)

PhD Student of Computer Engineering – Intelligent Systems (Universidad de Salamanca; GECAD – ISEP/P.Porto)

BSc and MSc in Informatics Engineering (ISEP – P.Porto)

Course Staff Image #2

Goreti Marreiros

Goreti Marreiros is Coordinator professor with Habilitation of the Institute of Engineering – Polytechnic of Porto (ISEP), Director of GECAD (Research Group on Intelligent Engineering and Computing for Advanced Development and Innovation), a research unit ranked with Excellent by the Portuguese Science and Technology Foundation, Director of the post-graduation in BigData&Decision Making in ISEP and Member of ISEP Scientific Council. She has PhD and habilitation in Informatics by Minho University and in the last years she participated as Principal investigator in 15 research projects: 8 European and 7 nationals. She is currently working in applying Artificial intelligence techniques (e.g. multi-agent systems, knowledge based systems, machine learning techniques) to the creation of smart environments. Since 2002 she authored more than 160 scientific publications, 100 of which in international journals and books. She is also vice-president of APPIA – Portuguese Association of Artificial Intelligence. E-mail: Author’s ORCID: