設計者の状態推定と発話内容に基づく設計論理抽出手法の検討

抄録

In the manufacturing industry, collaboration among designers is becoming more difficult due to the increasing scale and complexity of products. In this study, focus on design rationale, which is difficult to be converted into data, and aim to share it with a low burden. Design rationale can be extracted by analyzing the content of speech during design and estimating the designer's state. Automatic extraction method of design rationale information by combining the following states: automatically documenting recorded speech, analyzing the results of identifying important words using TF/IDF. Also estimating the skeletal structure from video recordings, estimating behavior from the data, and estimating physiological states from biometric signal information using smart glasses. We proposed a method for automatic extraction of design rationale information based on a combination of these two states. The results showed that the extraction of design rationale was more accurate than the extraction based on the analysis of speech content alone.