Development of an additive manufacturing decision support system (AMDSS)

Abstract

Additive manufacturing (AM) technology describes a set of processes capable of producing 3D physical products from CAD data directly. The rapid development of AM technologies and their wide applications makes the selection of the suitable process chains and materials a difficult task. Some researchers have tackled this problem by developing selectors that should assist users in their selections. The existing selector systems have some drawbacks: (і) often being outdated even before they were completely developed because new processes and materials are evolving continuously, (іі) representing only the point of view of their developers because users were not involved in the development process and (iii) not being holistic and able to help in all AM aspects for example process chains, materials, finishing methods and machines. This work has developed an updatable decision support system that assists users in their selections regarding AM process chains, materials, finishing methods, and machines. First, the study started by analyzing the available additive manufacturing selector systems and identifying their shortcomings. Secondly, the researcher identified target specifications for the new system, investigated different possible architectures for the system, selected knowledge based system (KBS) and database (DB) architecture to work together as a versatile tool that achieves the required target specifications. Next, the first version of the system was developed. Furthermore, verification and validation processes were made to test the developed system. Three case studies were used for the validation purpose: a typical consumer razor blade and two automotive components. These case studies were manufactured using AM technologies and then a comparison between real life decisions and the developed decision support system decisions were made. In addition, a number of interviews were performed in order to obtain users’ feedback about the first developed version. As a result of the feedback and evaluation a second version of the system was developed and evaluated. The results obtained from the second evaluation suggest that the second version is more effective than the first version during the selection process. To conclude, this study has shown that using KBS and DB together is effective to develop an updatable additive manufacturing decision support system. In addition, the user involvement in the development stage of the system enhances the system performance.