SWAMP - SoftWare in Additive Manufacturing Processes

Additive Manufacturing (AM) is an umbrella term for processes that incrementally layer material in order to print in three dimensions. Combined with software (SW) that automatically calculates the instructions for the AM machine, arbitrary objects can be brought to life without the need to plan for large-scale manufacturing processes. Hence, prototypes can be quickly developed, as well as individual solutions, making the technology a pillar for the fourth industrial revolution. One of the most widely used techniques is Fused Deposition Modeling (FDM), which melts and layers diverse materials such as polymers, metals, or even cement. However, this technology still suffers from several errors that may occur during the print process, as well as a high combinatorial complexity regarding parameters that can be tuned for each print.

In this project, we investigate the extent to which best practices from software engineering can improve the internal and external quality of the additive manufactured "hardware", i.e., the printed object, as well as the software. For this, we study the three steps of the additive manufacturing process including the configuration of the print, the generation of print instructions, as well as the monitoring and adaptation of the actual printing process. With software repository mining techniques, we aim at supporting the print configuration by gathering and analyzing metric values of print instructions to reveal possible failures. Using mutation testing paradigms, we want to simplify the replication process of print failures by forcing artificial failures and check whether quality assurance techniques can detect the "mutant". Finally, by using the concept of autonomic computing, we aim at increasing the machine's self-awareness to adapt the print process with special regards to time restrictions.