DataPlan – Assistance system for a data-based inspection planning

During the manufacturing of components, various data streams from different sources are generated, each with different data formats and frequencies. These data streams need to be synchronized to an unified database in order to be used effectively for production process planning.

This planning is still often done by humans: Employees evaluate the relevance of inspection characteristics based on their experience. This results in high inspection efforts and the possibility that irrelevant characteristics may be included in process planning. This can be avoided by integrating data-driven approaches into existing production systems. Due to the increasing digitalization of manufacturing processes, more and more data is available that has not yet been sufficiently utilized.

In addition, artificial intelligence also offers added value for planning. It can evaluate large amounts of data. It also generates automated recommendations for action to support employees in their decision-making.

Objectives

The goal of the project is the development of an assistance system for a data-based inspection planning in additive-subtractive process chains. To this end, quality and manufacturing data are synchronized and fed back into production planning. This enables data-based predictions of workpiece quality, reducing the effort required for inspection planning.

Benefits

• Standardized procedure – reduced dependence on employee experience
• Prediction of workpiece quality – higher accuracy through the use of data
• Support for employees – visualization of workpiece quality and generation of recommendations for action

Approach

In the project “DataPlan”, we are working closely with the companies Brinkhaus GmbH and toolcraft AG to develop an assistance system for data-based inspection planning. For this goal, we are first developing a method for synchronizing different data streams, which are fed back into production planning. These data streams are then fed into a software solution for analysing the factors influencing the prediction results.

We are conducting research of the assistance system on three complementary use cases: subtractive, additive-subtractive, and hybrid-additive manufacturing. This allows us to examine the transferability of the project results and evaluate their application across processes. In addition, we are integrating machine learning models into CAM software that can predict workpiece quality and generate recommendations for action.

Are you also interested in a cooperation project?

Contact Sarah Andrews via email at andrews@ifw.uni-hannover.de or by phone at +49 511 762 18320.