Colloquia

Samenvatting

Stork IMM, a manufacturer of plastic injection molding machines, wants to use data-driven maintenance to investigate the load on machines and to predict component failures. Current failures result in significant financial losses and downtime. By integrating data-driven maintenance, Stork IMM can minimize operational downtime and reduce warranty costs, increasing client satisfaction and strengthening the organization's competitive edge. However, Stork IMM is struggling with implementing data-driven maintenance and is not alone; implementing data-driven maintenance is a known hurdle for small and medium-sized enterprises (SMEs). In this project, we implement data-driven maintenance in Stork IMM, focusing on the implementation process to learn how to improve it for SMEs. The study's central question is: How to implement and leverage data-driven maintenance in SME Stork IMM?

During the implementation process, several hurdles for Stork IMM came to light. Bottlenecks were often related to the lack of system capabilities or skills required for data-driven applications, such as IT/OT convergence, organizational factors, and data completeness, consistency, and availability. Despite these hurdles, we implemented data-driven maintenance and were able to utilize the descriptive and diagnostic results valuably. Through data-driven analyses, we have diagnosed the degradation pattern of a critical component. The failure of this component appears to be visible a million cycles before failure, and these three months provide enough time to deliver a new part. We have also reduced the load on specific frames so that the cracks in these frames do not tear further and the frames do not collapse until the new frames arrive. Finally, we demonstrated how productivity can be increased with downtime information.

Technical and implementation frameworks from other researchers have been used and improved to implement data-driven maintenance. We have achieved unique insights by applying action research as our research method. By adopting a modular and iterative approach, we could progressively enhance the complexity of techniques and system capabilities. This allowed us to gradually develop the necessary skills and IT functionalities to achieve our required ambition level. A roadmap is created to help future implementers implement data-driven maintenance. The modular and iterative methodology, the coherence between the most important frameworks from literature, and improvements to these frameworks are incorporated in the roadmap.