Data-driven process improvement in production : A Master Thesis carried out at Scania CV

Abstract: The manufacturing industry is currently in its fourth paradigm shift, commonly referred to as Industry 4.0. As a part of this shift, the importance of collecting production data is ever increasing, due to its potential value. Scania is a heavy vehicle manufacturer, that is undergoing this transformation while simultaneously expanding their global production capacity. The purpose of the study is to assist Scania in this transformation by forming a recommendation for production engineers, to assist them in specifying production line equipment with good data collection capabilities. This was done partly by conducting a literature review, to gain a good understanding of the topic of research. Furthermore, a global benchmark consisting of interviews of employees within the organization was done. The interviews focused on: (1) how different production lines/units collect data, (2) different data collection and analysis software used by some units and (3) experts in relevant fields from within the organization. The results showed that Scania’s production data collection varies, depending on the production unit and their respective needs. Some production units use specialized production data software, that enables production traceability, visualisation, and analysis capabilities. The results further found that an existing framework within Scania for ensuring machine connection when specifying production equipment, exists but varies in usage. The results also included data scientist’s perspectives on how to enable data driven process improvements by highlighting the importance of data quality, good documentation, and data accessibility. The results led to three recommendations: (1) ensure machine connection by using a framework that exists within Scania, however, is currently not widely adopted, (2) design a structured database storage solution for the production data and tools to access and analyse the data, (3) encourage production engineers to increase their knowledge within data driven process improvement and encourage cross-functional collaboration with data scientists.