N91467 Data Analytics

This course introduces the students to the role of data analytics, one of the Industry 4.0 pillars, and its role for supporting the processes within manufacturing companies. The students learn the basics of descriptive, predictive, and prescriptive data analytics, supported by real examples of data analytics applications for the shop-floor management through theoretical classes, company visits and empirical seminars.

The course addressed, in particular, predictive data analytics and machine learning applications with the language R. Through R studio and R language the students practice on supervised and unsupervised machine learning with R and practically apply this knowledge to a real industrial data set with a project team-work. The project is also presented to the class at the end of the course, to make the students practice on public speaking and effective oral presentations on technical and scientific topics.

Moreover, some lessons are dedicated to the study of complex networks, i.e. networks with many nodes and links, through Social Network Analysis measures. Students learn to measure the importance of nodes and to identify the topology of the networks with the aim of characterizing their functioning.

The course consists of two modules.

Module 1: Data analytics and machine learning

- Introduction to data analytics: descriptive, predictive and prescriptive data analytics

- Predictive data analytics and Machine learning techniques (supervised and unsupervised)

- R studio and R language

- Machine learning with R

- Company visits and seminars on data analytics applications for shop-floor management

- Exercises and Project work development

Module 2: Network analysis

- Social networks

- Local measures of networks

- Global measures of networks

- Clustering on networks

The course is delivered through traditional frontal lessons but also innovative teaching methods, such as team working, brainstorming, company visits and company seminars.

The course evaluation consists of two parts.

Module 1 (Data analytics and machine learning), which weights 80% of the final grade, is evaluated based on a project work on the topics addressed during classes, developed by students in team and delivered as a tool developed on R studio and ppt presentation, and discussed with the lecturers during an oral presentation at the end of the course, with a final Q&A session.

Module 2 (Network analysis), which weights 20% of the final grade, is evaluated with an oral exam at the end of the course.