Proposal for Design Evaluation Methodology Using Generative AI : Focusing on ChatGPT and Midjourney

This study proposes a design verification and evaluation methodology using generative AI, focusing on ChatGPT and Midjourney, to enhance the objectivity and efficiency of the design process. While generative AI is mainly used in the early stages of design, this research explores its application in verifying final design outcomes. Using image-to-text tool, extracted text prompts from design images are compared with original design descriptions to assess aesthetic similarities. The evaluation framework includes six categories: concept, style, shape, composition, color, and texture, with weighted scoring for quantitative analysis. The methodology is validated with award-winning design works, demonstrating reliability through repeated assessments. This approach enables automated verification at different design stages, providing insights for optimizing design options. By integrating AI into the design validation process, this study enhances evaluation objectivity and supports data-driven decision-making, ultimately improving market success potential.

Authors

Publication Details

Journal
Korea Institute of Design Research Society
Published
2025-03-31
DOI
https://doi.org/10.46248/kidrs.2025.1.338
Primary Topic
Technology and Data Analysis
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Proposal for Design Evaluation Methodology Using Generative AI : Focusing on ChatGPT and Midjourney

YongJoo Kim, Ken Nah, T. Lim
Korea Institute of Design Research Society
Technology and Data Analysis
article

Proposal for Design Evaluation Methodology Using Generative AI : Focusing on ChatGPT and Midjourney

YongJoo Kim, Ken Nah, T. Lim
article en

Abstract

This study proposes a design verification and evaluation methodology using generative AI, focusing on ChatGPT and Midjourney, to enhance the objectivity and efficiency of the design process. While generative AI is mainly used in the early stages of design, this research explores its application in verifying final design outcomes. Using image-to-text tool, extracted text prompts from design images are compared with original design descriptions to assess aesthetic similarities. The evaluation framework includes six categories: concept, style, shape, composition, color, and texture, with weighted scoring for quantitative analysis. The methodology is validated with award-winning design works, demonstrating reliability through repeated assessments. This approach enables automated verification at different design stages, providing insights for optimizing design options. By integrating AI into the design validation process, this study enhances evaluation objectivity and supports data-driven decision-making, ultimately improving market success potential.

Korea Institute of Design Research SocietyVol. 10(1)
Openalex Percentile: Top 93%
Technology and Data Analysis
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.