Data lake – basis for
analyses and innovations

Mission transparency in manufacturing: Two prerequisites must be met for the digital image of a machine or a control system. Firstly, the comprehensive connectivity of all machines, and secondly, all signals and data storage.

The data lake in the FORCAM FORCE EDGE connectivity solution makes it easy to store all the data generated in manufacturing – at the signal level, interpretation level, event level, write operations, configuration changes, and transmitted NC files. The data can be made available to other applications from the data lake through the EDGE API interface – the latest AI algorithms, visualization tools, or audit requirements.

A data lake thus creates the basis for the success of data-driven manufacturing.

Advantages at a glance

  • Storage of structured and unstructured data

  • Current and historical Big Data analyses

  • Fast computing power

  • Unlimited data availability

Gain insights from historical data

Thanks to a data lake, the evaluation of historical data is made possible. Events in production can be digitally reconstructed to gain new insights into processes in the past. This helps answer a wide variety of questions, such as how the environmental variables of temperature or humidity influence the production result. With such knowledge, the production result can be significantly optimized.

Compare key figures over months and years

The data stored in the data lake makes it possible, for example, to map the development of energy data over time. Enriched with additional information such as the machine status, valuable insights into potential energy savings can be identified. For example, it is possible to find out whether machines are consuming energy despite being at a standstill.

 

Data basis for Machine Learning

The data lake in the FORCAM FORCE EDGE solution creates the prerequisite for successful data-driven manufacturing. The data is made available to the desired innovative analysis apps and enables the continuous optimization of production.

Example of machine learning for predictive maintenance: Traditionally, maintenance was only carried out after a fault, i.e., reactively. This meant that a production interruption could not be prevented. It was also not quick enough to ensure that the required spare part was also in stock.
Today, problems with a machine can be diagnosed at an early stage using predictive maintenance. The data lake supplies the data for this. The ability to plan maintenance is improved enormously. Maintenance can be scheduled outside of a shift, and spare parts can be ordered in time. This significantly reduces the risk of production interruptions, waiting times, and downtimes.

Machine learning models require training data to predict values and calculate probabilities. These are available in the data lake and thus create the foundation for machine learning in production. Machine learning models require training data to predict values and calculate probabilities. These are available in the data lake and thus create the foundation for machine learning in production.

Ihr Ansprechpartner

Sie wollen mehr zu dem Thema erfahren? Gerne beantworte ich Ihnen als Ihr Ansprechpartner offene Fragen zu unseren Lösungen. Sie erreichen mich unter: Oliver.Hoffmann@forcam.com

Oliver Hoffmann

Co-CEO FORCAM GmbH

oliver.hoffmann@forcam.com