Development of a Predictive Process Monitoring Methodology in a Self-organized Manufacturing System

This paper presents a methodology for the introduction of a predictive model in a flexible, self-organized manufacturing system, allowing the system to make improved decisions. Taking into consideration the high efficiency required for product manufacturing, predictive models provide valuable insights for accurate decision-making to optimize ongoing processes for key indicators such as process execution time. Furthermore, these models can be used to generate an appropriate response to possible perturbations in the system. Data analysis tools such as process mining allow developing a methodology that enables predictive process modelling in flexible self-organized manufacturing systems. The simulated system in this study is based on the manufacturing cell AIP-PRIMECA, which is located at Polytechnic University Hauts de France in Valenciennes. The process mining tools Apromore, Nirdizati, and ProM were used for the development of the proposed methodology and its implementation in the simulated system. It is expected that applying the proposed methodology will make the manufacturing system more efficient.