A Procedural Method to Build Decision Support Systems for Effective Interventions in Manufacturing - A Predictive Maintenance Example from the Spring Industry

Predictive maintenance as one of the most prominent data-driven approaches enables companies to not only maximize the reliability of production processes but also to improve their efficiency. This is especially valuable in today's volatile environment. Nevertheless, companies still struggle to implement digital technologies to track and improve their manufacturing processes, which includes data driven decision support systems. Based on practitioner interviews we identified the lack of guidance as a root cause. Additionally, literature reveals a shortcoming of methods especially suited for the needs of the manufacturing industry. This study contributes to this field by answering the question of how a procedural method can look like to guide practitioners to build decision support systems for effective interventions in manufacturing. Applying a design science research approach, the manuscript presents a seven-step procedural method to build decision support systems in manufacturing. The approach was designed and field tested at the example of a predictive maintenance model for a spring production process. The findings indicate that the incorporation of all stakeholders and the uncovering and use of implicit process knowledge in humans is of utmost importance for success.