Causal research optimises the production process: Industry 4.0 requirements necessitate prevention. "Up until now, preventive error avoidance mechanisms based on practice-related big data analytical methods have not been available for an MES. New developments, however, go far beyond processes such as traceability and process locking that have been used up until now," explains Dieter Meuser. "Future solutions will not only record and document errors, but will also detect the causes fully automatically. Specific and feedback-capable statements about process-specific problems are possible – machine processes will become more powerful,” he adds.
For example, a predictive maintenance (PdM) system can be set up which automatically parameterises the maintenance intervals for equipment, and thus significantly enhances both the qualitative and quantitative output of production plants. This will enable the overall plant effectiveness to be substantially improved.
Taking a proof of concept at the electronics factory of an automotive supplier as an example, Dieter Meuser demonstrates how big data analytical methods can be used to optimise processes by means of behavioural predictions. This is done by implementing a self-learning system that assures a corresponding flow of data and statistically evaluable statements. The aim is to formulate predictions for the required operating resources, servicing work and interventions in the processes of an SMT production line equipped with AOI systems. This will result in benefits such as more effective production planning and maintenance.
