In assembly processes, the degree of automation is currently less than 15 percent. This is due, for example, to the high number of variants, small batch sizes, the diversity of assembly processes and the difficulty of providing parts in the correct way. However, the main reason why assembly processes cannot be automated is because parts are so hard to separate. If this problem is only recognized in the production phase, any subsequent changes made to a component are associated with major efforts and costs, or it no longer makes sense to automate a process.
In the past, assessing the ability to separate a component was always linked to an expert’s knowledge and experience. With NeuroCAD, Fraunhofer IPA has now developed an online tool which automates this assessment with the aid of machine learning methods. Users can upload their STEP files and, on a scale of one to ten, find out within the space of a few seconds how easy or difficult it is to separate a component. In addition, the tool evaluates the gripping surfaces of a part and how easily it can be aligned. Work is currently in progress to obtain further information, such as positioning capability. Besides evaluating components, the neural network also states a probability that its evaluation is correct.
- Free use of NeuroCAD based on a pre-trained neuronal network
- Advice on all queries regarding the automation-friendly design of components
- Development and implementation of machine learning methods for customer-specific issues related to assembly automation
- For product designers: Information about the ability to automate assembly processes already during the planning phase of a product
- For manufacturers of separating devices: internal tool for preparing offers, as well as an external tool for sales and marketing purposes