Using machine and process data to adapt 3D-CAD Models

Projekt Duration: 2017/06/01 until 2020/05/31

Nowadays, it is common for industrial workpieces to consist not of the once familiar straightforward geometries like cylindrical or conical areas but of three-dimensional surfaces, which require simultaneous five-axis processes if they are to be produced using conventional production methods such as milling. Parallel to this, there is a growing requirement on the market for customized products resulting in the need for very low-volume production of parts. Wherever highly complex part design is required, computer-based 3D CAD models are now generated. The production planner can then use model-assisted machine simulations to run preliminary tests on the planned production process.

However, in the majority of cases the 3D-CAD model and the actual workpiece are usually not identical, despite precision planning. The inaccuracies in the model impact the subsequent production processes and not all of the actual effects are taken into account in the simulation. Consequently, errors can occur during the production process and prevent the required part quality and dimensions from being achieved. In the majority of cases, several planning, simulation and production cycles are needed in order to achieve the desired production outcome. This gives rise not only to extended throughput times but also to high levels of waste.

Our Solution

The aim of the "AdaptCAD" project is to minimize both the amount of planning required and variation between the 3D-CAD-Modell and the actual part. This is achieved by recording machine and process data from the machine tool during the production process, transferring them automatically to a database and ensuring structured storage. The data collected in this way can then be used in machine simulation applications in order to generate the actual geometry of the workpiece in virtual form. This digital twin of the workpiece is then available to production planners who can subsequently design production programs on the basis of an exact set of data. This helps to minimize production errors, thus enabling more reliable process chains to be generated and waste to be reduced. This, in turn, can lower costs and resource consumption. The adaptive concept in conjunction with the precise data make it possible to produce the parts at the first attempt, which is particularly beneficial to the production of customized parts.


  • Automated generation of exact CAD models and machine simulations
  • Data consistency throughout the entire process chain (data acquisition, machine simulation, geometry adaptation and process planning)
  • Use of open communication standards such as OPC-UA
  • Transfer of simulation models in parameterized surfaces via adaption algorithms
  • Reduced planning effort, waste and machine times


  • CP autosport GmbH, Büren
  • Exapt Systemtechnik GmbH, Aachen
  • ModuleWorks GmbH, Aachen

Project Funding

European Fund for Regional Development (EFRE)

Project Manager

Project Management Jülich, Research Center Jülich GmbH