Speaker
Description
The sequences of discrete events related to machine status can
characterize the operation of production systems. A model-based
methodology for Overall Equipment Effectiveness (OEE) improvement is
proposed based on the integrated analysis of the work-instructions,
the set of product-relevant expected manufacturing times, and
information about the competencies of the operators. The process model
is extracted from the log files of historical event data by process
mining algorithms.
The performance monitoring and conformance checking are based on the
comparison of the model and the work instructions with the help of
machine learning and process-flow simulation techniques.
The developed algorithms can detect manufacturing wastes and efficient
operating strategies. The applicability of the proposed methodology is
demonstrated in a cutting area optimization (CAO). The results confirm
the benefits of the proposed process mining technique in building
production monitoring relevant process models that can be incorporated
into the standard framework of OEE to identify the losses due to
quality problems, micro-stoppages, and product transitions.
