Component for the „Smart Factory Elements“ model
Manufacturing Analytics or Analytics overall is an important element in the control loop of the "Smart Factory Elements" model. According to this control loop, production is planned (Planning & Scheduling) based on specifications from different sources and this plan is then implemented or executed (Execution). The data collected is analyzed (Analytics) in order to make predictions which, together with other findings, can then be included in planning. The Industrial Internet of Things supports this cycle by collecting and providing data.
Analytics as a basis for optimizations
The functions and applications of the "Analytics" element use both conventional and innovative methods to prepare data recorded in the shop floor for KPIs and reports. In addition to classic dashboards, MPDV's Manufacturing Execution System (MES) HYDRA also provides extensive long-term analysis and big data evaluation as well as flexible self-service analytics applications. The latter are particularly useful if a great deal of data from different sources is being correlated and evaluated using different criteria. Based on this, optimization measures can then be initiated in the shop floor.
Future analytics applications will increasingly rely on Artificial Intelligence (AI) methods. Machine Learning in particular has proven to be a suitable technology for detecting irregularities in the production process. A first example of such so-called Advanced Analytics is Predictive Quality from MPDV. This allows the quality of a product to be predicted on the basis of recorded process values. Further application scenarios are the analysis and optimization of maintenance processes (Predictive Maintenance) and the early detection of deviations in everyday production.
Ultimately, MES HYDRA already meets many of the requirements that the "Smart Factory Elements" model places on a modern IT tool for Manufacturing Analytics. The increased use of Artificial Intelligence will gradually lead to new applications.
(approx. 2,200 characters)