Weidmüller says that its auto ML tool “democratises” the use of artificial intelligence (AI), and will provide the basis for more efficient production processes and new data-based business models. OEMs will be able to sell their machines based on their availability, or a guarantee of the number of parts that they can produce. Weidmüller believes that the ability to maximise machine productivity will result in return-on-investment times of a few months.
The company argues that existing ML tools are extremely demanding for traditional automation and machinery experts, who usually do not have the knowledge needed to develop the required models. Data analysis and model design are therefore carried out by data scientists, whose expertise is needed to apply AI or ML to the data and to develop models that can recognise anomalies or predict errors, for example.
The new tool guides users through model development, simplifying the process and allowing them to focus on their knowledge of machine and process behaviour. It will help companies to archive their application knowledge in reliable ML applications, as well as providing the components needed to implement AI.
An algorithm will learn typical data patterns for normal machine behaviour, based on historical data. Any deviations from these patterns may be the result of inefficiencies, minor malfunctions or more serious errors. The system can detect errors the first time they occur, even if they were previously unknown.
Weidmüller says that the auto ML tool will take on the role of a data science assistant, and guide the user through the process of creating models for anomaly detection, classification and prediction.