DISA

• When? May 5th 14.00 – 16.00
• Where? D1172 – Växjö (link will provided for those who wants to attend online)
• Registration: We would like to know how many that will attend onsite/online in order to get some fika for those onsite. So please register by May 3rd https://forms.gle/R1GuWXGXjiDYGWaQ6

Agenda
14.00-14.10 Welcome and practical information from Welf Löwe
14.10-14.55 Presentation and discussion: Title Data intensive applications and service development at Volvo CE – Joel Cramsky, Industry PhD-student Volvo CE
14.55 – 15.05 Coffee break
15.05 – 15.50 Presentation and discussion – Designing an Intelligent Predictive Maintenance Framework for Cyber-Physical Systems using Machine Learning and Digital Twin Technology – Mehdi Saman Azari, PhD student at LNU
15.50 -16.00 Sum up and plan for our next seminar on June 2nd

Abstracts

Data intensive applications and service development at Volvo CE – Joel Cramsky, Industry PhD-student Volvo CE

Volvo CE is aiming at becoming a leading ESG Tech company. An important part of this journey is to become more data driven and applying new business models focusing on services. That could mean leasing equipment instead of selling it or providing uptime maintenance contracts. Either way the high price here is to provide predictive maintenance for the customers. There are many other interesting services as well such as driver coaching, map features, safety features and more.

How to do this is ongoing research at VCE. Several methods have been explored so far. Digital twin concepts and pure data driven approaches using data and machine learning. Interesting questions and problems are; what methods to use?, what data is needed?, what complexities has to be considered?

Designing an Intelligent Predictive Maintenance Framework for Cyber-Physical Systems using Machine Learning and Digital Twin Technology – Mehdi Saman Azari, PhD student at LNU

Cyber-Physical Systems (CPSs) are essential for enabling Industry 4.0 by integrating physical processes with virtual equivalents to enhance flexibility, performance, and reliability. However, CPSs can be vulnerable to various physical threats due to their complexity and heterogeneity. The consequences of such threats can range from financial loss to loss of human lives. To reduce the risk of malfunctions, physical parts of these systems need to be carefully monitored, possibly through the use of intelligent predictive analytics techniques, such as predictive maintenance (PdM). One of the primary goals of PdM is to detect and identify the root cause of potential asset failures in a timely manner.

Data-driven diagnostics, which involve collecting sensor data and training fault diagnosis algorithms with relevant features, have gained widespread popularity over the last decade due to the increasing amounts of data produced by smart sensors and the rapid development of machine learning (ML) algorithms for predictive analytics. However, the success of data-driven PdM models largely depends on the availability of significant labeled data to train machine learning-based PdM models. Gathering such data, especially faulty data, involves significant costs and is impractical for machines in industrial applications due to safety and economic reasons.

The purpose of this seminar is to provide an overview of the latest developments in the application of Machine Learning (ML) to Data-driven Diagnostic Strategies (DDS). The presentation will highlight the preliminary results achieved so far, as well as the limitations of these approaches and possible solutions to overcome them. Particular attention will be given to current investigations of transfer learning and digital twin approaches to address the limitations of traditional ML and deep learning algorithms.