Data only reveals its actual value once it is processed and visualized in an appropriate way. Manufacturing Analytics includes a large range of methods, functions and applications to analyze data and to derive information or even real knowledge. The more complex these analyses become, the more important innovative technologies like Artificial Intelligence (AI), Machine Learning or Deep Learning are.
But Manufacturing Analytics already starts with simple evaluations or the calculation of important key figures (KPI) for the shop floor. Our Analytics applications consist of the following four pillars:
Kennzahlen / KPI
KPIs are important instruments providing a quick overview. KPIs are calculated by aggregating the collected data. Typical KPIs of the manufacturing environment are the OEE (Overall Equipment Effectiveness), the utilization efficiency or the scrap rate. Due to the integrated data acquisition and storage, MES applications are particularly suitable for calculating and displaying key figures.
Self Service Analytics
Self-Service Analytics are flexible applications for the demand-oriented evaluation of large amounts of data (Big Data). The typical functions like pivot tables, filters or drill down allow versatile approaches and provide useful insights. In addition to the HYDRA applications, MPDV offers its own product for Self-Service Analytics – the MES Cockpit Applications.
The browser-based application, MES Cockpit, visualizes detailed data from MES systems like HYDRA or other applications in formats that can be designed individually. General managers, controllers, production managers and shop floor staff benefit from an increased transparency in production, which is ensured by target-oriented data evaluations. Additionally, reliable KPIs support the continuous improvement process.
The MES Cockpit applications provide efficient graphic functions enabling correlative considerations of different areas and measurements. They also show trends making sure people in charge can respond and take countermeasures even before critical situations occur or specified limit values are violated.
If required, MES-Cockpit applications assist in setting up KPI systems based on the definitions illustrated in the VDMA standard sheet 66412-1 (ISO 22400).
If Analytics is combined with Artificial Intelligence, large amounts of data can be processed in real time. Model-based analyses are often used in this context. Ideally, the insights gained are used to derive predictions for the future. Predictive Quality is a first example of such an application.
To predict quality, one must be aware that scrap or rework can also occur if all process parameters are within the valid tolerances. This is due to complex connections and interactions, which are often down to the actual production technology. Predictive Quality takes these interrelationships into account and gives employees in production the opportunity to see immediately whether the part currently being made is a pass or fail.