Digital Twin and Data Analytics for Product Quality Monitoring and Root-Cause Analysis

The complexity of the product development process has been increasing continuously over the last decades. The reasons are progressively complex supply chains as well as distributed production sites and an increasing individualization of products. Furthermore, the development of cybertronic products connected to the internet as well as the software and services associated with these products present new challenges and require fundamental innovations in product development.

In addition to the challenges for product developers, the increased complexity is relevant for quality management. Highly connected systems and processes result in problems that cannot be fully controlled with preventive measures. For this reason, product failures cannot be completely avoided. In case of a problem, such vast amounts of data are available that the search for the technical cause of the error - the root-cause - is often time-consuming and difficult.

Given these challenges, product developers and quality engineers require an efficient management of the available data about each individual product instance and software that supports their root-cause analysis through (partial) automation.

This dissertation presents a Digital Twin for quality management, as well as data analysis methods for root-cause analysis.