Assessing the capabilities of ChatGPT to improve additive manufacturing troubleshooting

This paper explores the potential of using Chat Generative Pre-trained Transformer (ChatGPT), a Large Language Model (LLM) developed by OpenAI, to address the main challenges and improve the efficiency of the Gcode generation process in Additive Manufacturing (AM), also known as 3D printing. The Gcode generation process, which controls the movements of the printer's extruder and the layer-by-layer build process, is a crucial step in the AM process and optimizing the Gcode is essential for ensuring the quality of the final product and reducing print time and waste. ChatGPT can be trained on existing Gcode data to generate optimized Gcode for specific polymeric materials, printers, and objects, as well as analyze and optimize the Gcode based on various printing parameters such as printing temperature, printing speed, bed temperature, fan speed, wipe distance, extrusion multiplier, layer thickness, and material flow. Here the capability of ChatGPT in performing complex tasks related to AM process optimization was demonstrated. In particular performance tests were conducted to evaluate ChatGPT's expertise in technical matters, focusing on the evaluation of printing parameters and bed detachment, warping, and stringing issues for Fused Filament Fabrication (FFF) methods using thermoplastic polyurethane polymer as feedstock material. This work provides effective feedback on the performance of ChatGPT and assesses its potential for use in the AM field. The use of ChatGPT for AM process optimization has the potential to revolutionize the industry by offering a user-friendly interface and utilizing machine learning algorithms to improve the efficiency and accuracy of the Gcode generation process and optimal printing parameters. Furthermore, the real-time optimization capabilities of ChatGPT can lead to significant time and material savings, making AM a more accessible and cost-effective solution for manufacturers and industry.(c) 2023 Kingfa Scientific and Technological Co. Ltd. Publishing services by Elsevier B.V. on behalf of KeAi Communications Co. Ltd. This is an open access article under the CC BY license (http://creativecommons. org/licenses/by/4.0/).