Engineering knowledge is relevant to the field in manufacturing and logistics processes. Every change and high-level decision making depend on this. To support engineering knowledge and to keep it up-to-date, the data-driven process model discovering is proposed.
Process mining is a family of process management that support the analysis of business processes based on event logs. During process mining, specialized data mining algorithms are applied to event log data to identify trends, patterns and details contained in event logs recorded by the manufacturing execution system (MES). Process mining aims to improve process efficiency and understanding of processes.
Process simulation technique
The huge amount of data are generated from MES, and it becomes harder to understand the overall processes using the data analysis techniques. Process mining can extract process-oriented knowledge from event logs extracted through MES, so it can exploit the big data and provide an accurate view of the manufacturing process. Process mining provides a manufacturing process model, which is valuable to provide an insight into actual manufacturing processes. Based on this model, the discrete-event process simulation is proposed to process improvement and to answer the “what if…” questions.
Simulation in manufacturing systems is the use of software to analyze them and thereby obtain important information. It is easy to understand the difference made by changes in the local system, but it is very difficult or impossible to assess the impact of this change in the overall system. The simulation gives us some measure of this impact.
The Sunstone system provide the following data for the process-simulation software:
- Parts produced per unit time
- Time spent in system by parts
- Time spent by parts in queue
- Time spent during transportation from one place to another
- In time deliveries made
- Build up of the inventory
- Inventory in process
- Percent utilization of machines and workers.