Process Mining in Smart Manufacturing
Especially in Europe, manufacturing enterprises get drilled to develop strategies to create a digital CO2 product passport for produced products. In addition to that comes the necessity of analysis and optimization tools, with a particular focus on circular economy. To achieve this goal, the supply chain is in focus both horizontally and vertically. The horizontal view addresses both the value stream within company boundaries and the entire supply chain. The product carbon footprint (PCF) is primarily determined using data provided by the manufacturing execution system (MES) and the enterprise resource planning (ERP) system. These systems are linked via standardized interfaces. Compared to the vertical view, this involves a low level of effort in digitalization. One research question here is to examine the trade-off between the required accuracy and the necessary digitization effort. In the vertical approach, individual processes in the company are analyzed and enabled to calculate a PCF in real-time. For this detailed determination of the PCF at the machine level, extensive monitoring of the corresponding production systems is required, which is why these are equipped with appropriate sensor technology. To determine the PCF, the previously used MES and ERP data are expanded to include machine data in order to implement a cause-based determination of the PCF.
However, the collection of production data, for example on resource consumption during production and its link to CO2 emissions, is only of limited value, as it only relates to the respective machines. This means that general statements can be made about capacity utilization and consumption at certain points in time via plants or plant components, but only to a limited extent about the consumption and thus also emissions for a specific manufactured product, as this is not recorded or assigned individually in a dedicated manner.
| Category | Data Analytics |
| Current Version | 1.x |
| Language | Multilingual |
| Status | Prototype |
| Authors | Software AG Research team |
A new approach here is the additional, largely automatic, monitoring of products to be produced along the entire production line. This involves recording exactly when a workpiece generally goes into production and leaves the production line, as well as which workstation is passed, when, and for how long. At the same time, all consumption data is collected throughout the day for the digitized production line and management systems. Process mining (we use ARIS) can now be used to merge both sets of data. This makes it possible to determine which resources were used for a specific product to be manufactured for each work step at a workstation and to use the mining of a process to precisely determine the exact machining sequence or machining process, including any reworking. Using the corresponding analysis tool, not only can accurate CO2 product passports be issued, but even in the case of serious production errors a reduction in waste output could be achieved, or at least for the post-processing cycles, the consumption can be reduced or in some cases even specific product-related causes can be identified and later avoided.
