https://www.mdu.se/

Introduction: Modern production environments are characterized by increasing complexity, requiring effective tools to support decision-making. This thesis investigates the use of Dynamic Value Stream Mapping (DVSM) as a method for enhancing human-centered decision-making in manufacturing systems, in line with Industry 5.0 principles.

Methods: An Action Research approach was conducted, involving the development of a DVSM-based prototype in 4 cycles each containing 6 phases; diagnosing, action planning, action taking, evaluation and learning. Data was collected from various steel company stakeholders with a mix of quantitative and qualitative data, where the latter was analyzed using thematic analysis. Data was also sourced from production datasets and regular client feedback sessions. The DVSM was developed using model-driven engineering.

Results: The findings indicate that DVSM with a simplified VSM structured provides good overview for multiple stakeholders. Key contributions include enhanced system overview, support for focused information processing, and facilitation of collaborative decision-making. However, limitations were identified related to real-time data integration and the absence of deployment in a live production environment and as its data connectivity was semi-automatic and unidirectional, the DVSM was classified as a digital model but could also be considered a digital shadow depending on its definition.

Discussion: The results suggest that DVSM has potential as a decision-support tool in complex manufacturing settings and aligns with Industry 5.0 by supporting human-centered and collaborative processes. Nevertheless, further work is required to validate the approach in real-world environments and to address technical challenges related to scalability and data integration, especially considering the narrow scope of the study.