Interactive Extraction of 1D-CAE Modeling Requirements and Determination of Model Configuration Using Generative AI
生成系AIを用いた1D-CAEモデリング要件の対話的抽出とモデル構成の決定
- Delivery
- Available on this site
- Format
- Price
- Non-members (tax incl.)：￥1,100 Members (tax incl.)：￥880
- Publication code
- 20265363
- Paper/Info type
- Proceedings (Spring)
No.84-26
- Pages
- 1-5(Total 5 p)
- Date of publication
- May 2026
- Publisher
- JSAE
- Language
- Japanese
- Event
- 2026 JSAE Annual Congress (Spring)
Detailed Information
| Author(J) | 1) 内藤 恭兵, 2) 市村 純一, 3) 脇本 佑紀 |
|---|---|
| Author(E) | 1) Kyohei Naito, 2) Junichi Ichimura, 3) Yuki Wakimoto |
| Affiliation(J) | 1) ニュートンワークス, 2) ニュートンワークス／放送大学, 3) ニュートンワークス |
| Affiliation(E) | 1) NewtonWorks, 2) NewtonWorks/The Open University of Japan, 3) NewtonWorks |
| Abstract(J) | 工数制約により形骸化しがちな，実システムを解析目的に応じて理想化する「モデル化プロセス」に対し，生成AIとの対話を用いた支援手法を提案する．解析目的から適切な仮定や抽象化の指針を対話的に引き出し，工学的に妥当な仕様へと構造化することで，効率的かつ本質を捉えたモデル定義を行うプロセスについて論じる． Translation |
| Abstract(E) | We address the modeling process of idealizing real systems according to analysis objectives, which often becomes superficial due to time constraints. We propose a methodology to define the required information and extract it interactively using Generative AI. We discuss the process of eliciting appropriate assumptions and abstraction levels through dialogue, structuring them into engineering-consistent specifications to achieve model definitions that are both efficient and capture the essence. |
