Leveraging Generative Artificial Intelligence for Prototyping: Ambidextrous Thinking in Engineering Design
Article Sidebar
Main Article Content
Abstract
Rapid economic and technological changes make the future increasingly unpredictable in the current society. Consequently, an urgent need exists to cultivate human resources capable of making significant contributions, particularly in education, where design education programs play a crucial role. In engineering design, “ambidextrous thinking” has also garnered significant attention. This study has two main objectives: (1) To develop a new evaluation method and provide guidelines for future studies by performing large language model (LLM)-based evaluations of prototyping, specifically along the axes of exploration and exploitation. (2) To quantitatively and qualitatively analyze the impact of design education on engineering. This study evaluates 31 product redesigns by third-year students enrolled in the Design Engineering course at Kogakuin University. In the LLM-based evaluation of “exploitation,” 70% of the top 10 proposals suggested by ChatGPT received the highest rating from the class instructor. In addition, in the “exploration” evaluation, incorporating the concept of “darkness” into the existing definition revealed the potential for a more effective evaluation of prototypes.
Article Details
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
References
Leifer L, Steinert M. Dancing with ambiguity: Causality behavior, design thinking, and triple-loop-learning. Inf Knowl Syst Manag. 2011;10(1–4):151–173.
Faste R. Ambidextrous thinking. Innov Mech Eng Curric 1990s. 1994.
Berger-Tal O, Nathan J, Meron E, Saltz D. The exploration-exploitation dilemma: a multidisciplinary framework. PLoS One. 2014;9(4):e95693.
Yen AZ, Hsu WL. Three questions concerning the use of large language models to facilitate mathematics learning. Conf Empir Methods Nat Lang Process. 2023.
Fede G, Rocchesso D, Dow S, Andolina S. The idea machine: LLM-based expansion, rewriting, combination, and suggestion of ideas. Proc 14th Conf Creativity Cognition. 2022 Jun;623–627.
Pierson KC, Ha MJ. Usage of ChatGPT for engineering design and analysis tool development. AIAA SciTech 2024 Forum. 2024;0914.
Cheung LH, Dall'Asta JC, Di Marco G. Exploring large language model as a design partner through verbal and non-verbal conversation in architectural design process. Proc SIGraDi 2023 Conf. 2023 Nov 20–24; São Paulo, Brazil. São Paulo: Blucher; 2024. p. 1049–1060.
Bossou K, Ackerman M. Should machines evaluate us? Opportunities and challenges. International Conference Innov Comput Cloud Comput. 2021.
Sanwuala G, Fatima SS. A study of automated evaluation of student’s examination paper using machine learning techniques. International Conference Comput Commun Intell Syst (ICCCIS). 2021;1049–1054.
Xu W, Kotecha MC, McAdams DA. How good is ChatGPT? An exploratory study on ChatGPT’s performance in engineering design tasks and subjective decision-making. Proc Des Soc. 2024;2307–2316.
Chiarello F, Barandoni S, Majda Škec M, Fantoni G. Generative large language models in engineering design: Opportunities and challenges. Proc Des Soc. 2024;4:1959–1968.
Freire SK, Foosherian M, Wang C, Niforatos E. Harnessing large language models for cognitive assistants in factories. 5th International Conference Convers User Interfaces; 2023 Jul 19–21; Eindhoven, Netherlands. New York: ACM; 2023. Article 44, p. 1–6.
Aikawa Y, Tamura R, Xu C, Ge X, Misaki D. Introducing augmented Post-it: an AR prototype for engaging body movements in online GPT-supported brainstorming. Adjunct Proc 36th ACM Symp User Interface Softw Technol (UIST '23 Adjunct). 2023. Article 6, p. 1–3.
Maher ML, Fisher DH. Using AI to evaluate creative designs. In: Duffy A, Nagai Y, Taura T, eds. DS 73-1 Proc 2nd International Conference Design Creativity Vol 1. 2012. p. 45–54.
Okamoto M, Murakami T. Proposal of defining exploration and exploitation in engineering design and evaluating the degree of exploration by natural language processing. Proc ASME 2022 Int Des Eng Tech Conf Comput Inf Eng Conf. Vol 6: 34th International Conference Des Theory Methodol (DTM); 2022 Aug 14–17; St. Louis, MO, USA. ASME; 2022.
Leifer L, Steinert M. Dancing with ambiguity: Causality behavior, design thinking, and triple-loop-learning. Inf Knowl Syst Manag. 2011;10(1–4):151–173.
Ge X, Leifer L. Design thinking at the core: Learn new ways of thinking and doing by reframing. Proc Am Soc Mech Eng. 2017.
Lande M. Catalysts for design thinking and engineering thinking: Fostering ambidextrous mindsets for innovation. Int J Eng Educ. 2016;32(3):1356–1363.
Okamoto M, Murakami T. Proposal of cluster analysis method for products considering exploration and exploitation in engineering design. Proc Des Soc. 2023;2995–3004.
Bushnell T, Steber S, Matta A, Cutkosky M, Leifer L. Using a ‘dark horse’ prototype to manage innovative teams. 3rd International Conference Integr Des Eng Manag Innov; 2013 Sep; Delft, Netherlands. p. 8.
Wang J, Liang Y, Meng F, Sun Z, Shi H, Li Z, Xu J, Qu J, Zhou J. Is ChatGPT a good NLG evaluator A preliminary study. Proc 4th New Frontiers Summarization Workshop. 2023;1–11.
Kocmi T, Federmann C. Large language models are state-of-the-art evaluators of translation quality. Proc 24th Annu Conf Eur Assoc Mach Transl. 2023 Jun 12–16; Tampere, Finland. Tampere: EAMT; 2023. p. 193–203.