AN ASSESSMENT METHOD OF INTERVIEWS FOR MODELING ENGINEERING DESIGN PROCESS
DS 136: Proceedings of the Asia Design and Innovation Conference (ADIC) 2024
Year: 2024
Editor: Yong Se Kim; Yutaka Nomaguchi; Chun-Hsien Chen; Xiangyang Xin; Linna Hu; Meng Wang
Author: Nishida, Rempei; Nomaguchi, Yutaka; Fujita, Kikuo; Hoang, Le Ang; Omori, Takahiro; Sasaki, Yuji; Mori, Yuji; Nakagawa, Yasutada; Inoue, Michinobu; Hosokawa, Daisuke
Series: Other endorsed
Institution: Osaka University, Toshiba Corporation
Page(s): 055-063
Abstract
The design and development process that produces innovative products and services has been considered an important organizational capability. Modeling of the design and development process is generally conducted through interviews with experienced engineers. However, the success of such modeling depends heavily on the ability of the interviewer to converse with experienced engineers, and it is not clear what kind of conversation is effective for modeling. This has become a bottleneck in knowledge sharing and reuse within organizations. To solve this problem, this study aims to establish a method for clarifying and assessing the conversation structure of interviews for modeling the engineering design processes. The proposed method consists of two parts: a method for analyzing the conversation structure using an integrated three-layered process model, which the group of Osaka University has developed to represent the system design process, and a method for evaluating the relationship between questions and answers in the conversation based on text similarity measurement based on a large language model. As a case study, we take up an example of modeling a solder joint design and evaluation process carried out in a manufacturing company. We apply the proposed method to the analysis and assessment of the modeling process conducted by an expert interviewer and a novice interviewer for an engineer. We discuss the validity of the proposed method under the results of the case study.
Keywords: Design process, Modeling, Engineering knowledge, Text similarity, Conversation assessment