Root cause analysis for resilient production systems through Industry 4.0 technologies

Doktorsavhandling, 2023

Creating and developing resilient production systems is critical if manufacturing companies are to thrive in a globally competitive market. Being flexible and agile, resilient systems can avoid, withstand, adapt to and recover from disturbances. A crucial ability is learning from experienced disturbances so they can be avoided in future. This is commonly done in manufacturing companies by performing a root cause analysis. However, the current practice
of root cause analysis lacks efficiency and effectiveness, which contributes to the high reoccurrence of disturbances encountered daily by manufacturing companies. Fortunately, with the introduction of Industry 4.0 technologies, the process of root cause analysis is expected to change greatly. With the aim of supporting practitioners in improving their root cause analysis processes, this research focuses on: (1) describing the current challenges; (2) describing the requirements for new technological solutions; and (3) identifying and designing new technological solutions, given the context of Industry 4.0. To do so, a qualitative approach was adopted, inspired by design science research (DSR) and based on six studies involving manufacturing companies and technology providers.
Regarding the main challenges, the results of this research indicate that manufacturing companies are still performing unstructured root cause analysis, relying on experts to identify root causes and struggling to know how to analyse and integrate relevant data effectively. Furthermore, regarding requirements, the results of this research indicate that technological solutions for root cause analysis should be data-driven and easy to use. They should integrate different data sources, allow secure collaboration and support employee learning. Based on the requirements, the results of this research indicate that the leading technological solutions involve such things as data analytics, the development of thesauruses of disturbances and their causes, the design of specific data architectures and systems for root cause analysis and the design of platforms for stronger collaboration. Finally, in this research, specific high-level designs are proposed for an application to support root cause analysis of machine stops; and a collaborative platform for root cause analysis at the value-chain level. This research has practical and theoretical implications. Its results may be used directly by practitioners to gain
insight into potential improvements to their practices and as input for developing specific root cause analysis applications. The results of this research also advance knowledge in the field of root cause analysis by providing empirical evidence of challenges, requirements and solutions.