AI-driven automation for structural design has been actively studied in structural engineering. In particular, reinforcement learning (RL) has attracted attention as a framework in which an agent interacts with an environment to autonomously search for optimal design solutions in complex design spaces. This study proposes an automated design model for rectangular reinforced-concrete (RC) columns based on a multi-agent Double Deep Q-Network (Double DQN). Extending prior RL-based automation developed for RC beam design to column members, the proposed environment explicitly incorporates key column-specific behaviors, including axial force–bending moment (P–M) interaction and moment magnification due to column buckling. Four agents independently determine the section width (b), section depth (h), number of longitudinal bars (n), and bar size. The reward function combines (i) penalty terms for violations of ACI 318-19 design constraints and (ii) an economic reward defined relative to an approximate optimal cost predicted by a quadratic regression model. After training for approximately 10,000 episodes, the proposed multi-agent Double DQN consistently generated ACI-compliant column designs across all test load cases and produced solutions with improved cost efficiency compared with the approximate optimal baseline. These results demonstrate the feasibility and practical potential of multi-agent RL for automated RC column section design.

본 연구의 목적은 초등 예비교사가 C hatGPT-4 o를 활용하여 과학 평가 문항을 개발하는 과정에서 나타난 사전 인식과 설계 경험의 특성을 탐색하는 데 있다. 이를 위해 비수도권 소재 A 교육대학교의 초등 예비교사 5명을 선정하 고, 2015 개정 초등과학 교육과정 ‘지구와 우주’ 영역을 중심으로 평가 문항을 생성·검토·수정하는 설계 활동을 수행하도록 하였다. 연구 자료는 사전 서면 응답, 문항 초안 및 최종본, ChatGPT 상호작용 로그와 프롬프트 입력 이유 기 록, 반성적 저널, 반구조화된 면담을 통해 수집되었다. 자료 분석은 두 단계로 이루어졌다. 첫째, 문항 생성 이전에 형 성된 인식은 사전 서면 응답을 중심으로 귀납적 내용 분석을 실시하였다. 둘째, 문항 검토·보완 단계에서의 설계 경험 은 사례 기반 과정 중심 질적 분석을 통해 AI Assessment Scale (AIAS) 틀을 참조하여 해석하였다. 분석 결과, 사전 인식은 (1) Gen AI를 설계 부담 완화와 사고 확장을 지원하는 도구로 기대하는 ‘역할 기대’와, (2) 생성 결과에 대한 신뢰와 불확실성이 공존하며 과학적·교육과정 타당성 검증과 최종 판단 책임은 교사에게 귀속된다는 ‘한계 인식 및 인 간 책임’의 두 축으로 구조화되었다. 또한 설계 경험 분석에서는 예비교사들이 Gen AI가 제시한 문항을 그대로 사용하 기보다 사고 구조의 재설계, 인지 수준 조정, 평가 기준 및 판단 책임, 문항 범위 및 표현 조정을 통해 인간 주도로 재 구성하는 양상이 나타났다. 이러한 활용은 전 사례에서 AIAS Level 4의 특성과 관련지어 해석될 수 있으며, Gen AI가 설계를 지원하더라도 핵심 판단과 책임은 예비교사에게 남아 있음을 보여준다. 본 연구는 사전 인식과 설계 경험을 구 분하여 분석함으로써, 초등 예비교사 교육에서 Gen AI 활용과 평가 전문성 논의를 확장하는 데 기초 자료를 제공한다.

This study aims to develop an underground expressway design for an exit area to mitigate traffic congestion. Thus, we explain the necessity of underground expressways and three reasons for persistent congestion on underground expressways despite an increase in supply. We focus on the first reason, which is a complicated traffic-flow conflict in the exit area, and analyze the traffic flow based on various conditions, such as the exit rate to a nearby interchange and the exit location for underground and ground roads. Consequently, we identify three factors that affect congestion in the exit area. The first factor is the exit rate, where a higher exit rate corresponds to a more severe congestion. The second is the exit location of two roads. When the exit of a road that exhibits a higher exit rate is placed on a curb side, the average delay is reduced. The final factor is the length of the lane-change section, where a longer lane-change section correspond to less congestion. However, after a certain length, the change in congestion is negligible. Based on these results, we suggest revised design guidelines for underground expressways in terms of exit location and the length of lane-change sections.

This study develops and evaluates a prompt-driven large language model (LLM) agent for section design of doubly reinforced concrete (RC) beams. Using Google Gemini (Gems), an engineering “expert” that operates without fine-tuning by uploading ACI-318 provisions, sample design documents, and a database of prior beam designs was developed. The agent interprets code clauses, formulas, and constraints from these materials and retrieves similar design cases to propose an initial solution. It then incorporates user-specified natural-language constraints—most notably a strength-ratio cap (design strength ≤ 105% of required strength)—to iteratively refine toward safe and economical designs. Beyond reporting member size and reinforcement details, the agent provides step-by-step computational justifications for moment and shear checks, increasing verifiability and instructional value. We benchmark the LLM-generated designs against results from the commercial program MIDAS/Design+ and observe close agreement. In several scenarios, the constraint-guided LLM solutions are more material-efficient while remaining code-compliant. The workflow also supports batch processing from spreadsheet inputs, enabling practical automation across multiple beams. The approach requires no additional model training or coding making it accessible to non-developer practitioners. Results indicate that a general-purpose LLM, properly grounded with code documents and examples, can achieve practice-level performance with transparent reasoning. This demonstrates a viable approach to AI-assisted structural design that is explainable, interactive, and readily integrated with engineering workflows.

This study investigates the traditional costumes featured in The Purple Hairpin, a representative work of Cantonese Yue opera, and explores their creative adaptation into modern fashion design. Cantonese Yue opera costumes, known for their symbolic patterns, colors, and craftsmanship, embody the cultural and aesthetic identity of the region. Through comprehensive literature review, field research, and analysis of museum artifacts and performance images, the structural and symbolic characteristics of key characters’ costumes were systematically examined. Based on these findings, two modern womenswear designs were developed: one inspired by the Xiaoguzhuang worn by noblewomen and another based on the Yuanling, a traditional official’s robe. Each design aimed to harmonize traditional aesthetics with contemporary sensibilities and functionality. Patterns for these designs were drafted and tested using Style3D software to conduct virtual fitting simulations, which allowed for evaluating wearability, aesthetic qualities, and structural stability. The results demonstrate that the distinctive symbolic elements and artistic values of traditional Yue opera costumes can be successfully reinterpreted into modern fashion, while digital tools enhance the efficiency and precision of pattern development. This research contributes to preserving and revitalizing traditional costume culture by providing a practical methodology for modern application, as well as offering insights for future global promotion and creative use of Chinese traditional dress in contemporary fashion contexts.

The number of significant issues on many welding processes are often connected to high productivity and manufacturability at low costs. The research on welding processes in the literature has reported several research activities, but there is still scope for improvement in most industrial settings. The primary goal of this research is to determine the best super-TIG welding settings to use for groove welding. First, in order to determine the quality characteristics and risks associated with them, concepts and frameworks of quality by design (QbD) which is a new standard in pharmaceutical area in order to improve drug qualities were integrated into this process optimization. Second, stepwise experimental design approaches including a factorial design as well as a response surface methodology (RSM) were customized and performed for this specific automated super-TIG welding process. Third, based on experimental design results, the optimal operating conditions with both design space (i.e., acceptable range of operating conditions) and safe operating space (i.e., safe range of operating conditions) were obtained. Finally, a case study including QbD steps, stepwise experimental design approaches, design and operating spaces, the optimal factor settings, and their association validation results was conducted for verification purposes.

Reinforcement learning (RL) is successfully applied to various engineering fields. RL is generally used for structural control cases to develop the control algorithms. On the other hand, a machine learning (ML) is adopted in various research to make automated structural design model for reinforced concrete (RC) beam members. In this case, ML models are developed to produce results that are as similar to those of training data as possible. The ML model developed in this way is difficult to produce better results than the training data. However, in reinforcement learning, an agent learns to make decisions by interacting with an environment. Therefore, the RL agent can find better design solution than the training data. In the structural design process (environment), the action of RL agent represent design variables of RC beam. Because the number of design variables of RC beam section is many, multi-agent DQN (Deep Q-Network) was used in this study to effectively find the optimal design solution. Among various versions of DQN, Double Q-Learning (DDQN) that not only improves accuracy in estimating the action-values but also improves the policy learned was used in this study. American Concrete Institute (318) was selected as the design codes for optimal structural design of RC beam and it was used to train the RL model without any hand-labeled dataset. Six agents of DDQN provides actions for beam with, beam depth, bottom rebar size, number of bottom rebar, top rebar size, and shear stirrup size, respectively. Six agents of DDQN were trained for 5,000 episodes and the performance of the multi-agent of DDQN was evaluated with 100 test design cases that is not used for training. Based on this study, it can be seen that the multi-agent RL algorithm can provide successfully structural design results of doubly reinforced beam.

Tuned Mass Dampers (TMDs) are widely used to mitigate structural vibrations in buildings and bridges. However, conventional optimization methods often struggle to achieve optimal performance due to the complexity of structural dynamics. This study proposes the NN-L-BFGS-B algorithm, which combines Artificial Neural Networks (ANNs) for global exploration and L-BFGS-B for local exploitation to efficiently optimize TMD parameters. A ten-story shear-building model with a TMD is used for validation. The proposed method achieves the lowest H₂ norm compared to previous studies, demonstrating improved optimization performance. Additionally, NN-L-BFGS-B effectively balances computational efficiency and accuracy, making it adaptable to various engineering optimization problems.

Automated structural design methods for reinforced concrete (RC) beam members have been widely studied with various techniques to date. Recently, artificial intelligence has been actively applied to various engineering fields. In this study, machine learning (ML) is adopted to make automated structural design model for RC beam members. Among various machine learning methods, a supervised learning was selected. When a supervised learning is applied to development of ML-based prediction model, datasets for training and test are required. Therefore, the datasets for rectangular and t-shaped RC beams was constructed by commercial structural design software of MIDAS. Five supervised learning algorithms, such as Decision Tree (DT), Random Forest (RF), K-Nearest Neighbor (KNN), Artificial Neural Networks (ANN), eXtreme Gradient Boosting (XGBoost) were used to develop the automated structural design model. Design moment (Mu), design shear force (Vu), beam length, uniform load (wu) were used for inputs of structural design model. Width and height of the designed section, diameter of top and bottom bars, number of top and bottom bars, diameter of stirrup bar were selected for outputs of structural design model. Performance evaluation of the developed structural design models was conducted using metrics sush as root mean square error (RMSE), mean square error (MSE), mean absolute error (MAE), and coefficient of determination (R2). This study presented that random forest provides the best structural design results for both rectangular and t-shaped RC beams.

중앙버스전용차로는 일반 도로 대비 높은 교통량과 반복적인 축하중이 작용하는 구간으로, 정차 및 출발 과정에서 발생 하는 국부적인 응력 집중으로 인해 포장 파손이 빈번하게 발생한다. 그러나 기존 도로 설계에서는 정적인 교통량을 기준 으로 축하중을 산정하여, 실제 교통 환경에서의 버스 유형별 차이, 재차 인원, 시간대별 하중 변화 등 동적인 요소를 충 분히 반영하지 못하는 한계가 존재한다. 이에 본 연구에서는 대중교통 빅데이터를 활용하여 중앙버스전용차로의 버스 유 형 및 시간대별 재차 인원을 반영한 새로운 축하중 산정 모델을 개발하였다. 이를 위해 서울시 열린 데이터 광장의 교통 정보를 활용하여 버스 유형 및 시간대별 재차 인원 데이터를 수집하고, 카카오맵 및 네이버 로드뷰 데이터를 이용해 결 측치를 보완하여 데이터셋을 구축하였다. 구축된 데이터셋을 활용하여 기존 ESAL(Equivalent Single Axle Load) 방식과 비교 분석한 결과, 새로운 축하중 모델에서는 기존 방식 대비 평균 111.8% 높은 축하중이 산정되었으며, 일부 구간에서 는 최대 128.9%까지 차이가 발생하는 것으로 나타났다. 이는 기존 포장 설계가 중앙버스전용차로의 실질적인 교통 하중 을 충분히 반영하지 못하고 있음을 시사하며, 추가적으로 버스 중하중의 가·감속의 영향을 고려한다면, 시간대별·노선별 실시간 축하중 변화를 보다 정밀하게 분석할 수 있으며, 이를 통해 과소 산정된 설계 하중을 보완하고 포장 공용성을 향 상시킬 수 있는 최적의 설계 및 유지보수 전략 수립이 가능할 것으로 기대된다.

Performance-Based Seismic Design (PBSD) is an approach that evaluates how structures will perform under different levels of seismic activity. It focuses on ensuring that buildings not only withstand earthquakes but also meet specific performance objectives, such as minimizing damage or maintaining functionality after the event. Unlike traditional methods, PBSD allows for more tailored, cost-effective designs by considering varying degrees of acceptable damage based on the structure's importance and use. PBSD was introduced in Korea in 2016 to replace elastic design, which is inevitable to over-design to cope with all variables such as earthquakes and winds. When PBSD is applied to the structural design new building, One of the challenges of PBSD is the complexity involved in creating accurate inelastic analysis models. The process requires significant time and effort to analyze the results, as it involves detailed simulations of how structures will behave under seismic stress. Additionally, organizing and interpreting the analysis data to meet performance objectives can be labor-intensive and technically demanding. In order to solve this problem, a post-processor program was developed in this study. A post-processor was developed based on Excel program using Visual Basic for Applications(VBA). Because analysis outputs of Perform-3D, that is a commercial software for structural analysis and design, are very complicated, generation of tables and graphs for report is significant time and effort consuming task. When the developed post-processor is used to make the seismic design report, the required task time is significantly reduced.

The purpose of this study was to optimize the design of asphalt concrete pavements for Jeju Island by considering the regional characteristics of the island. This study employed an MEPDG program to determine the allowable traffic loads for class 4 vehicles by considering the axle loads, climate, and material properties. Samples of basalt asphalt concrete from Jeju were used to measure the dynamic modulus for material property estimation. The climate input was based on 30-year climate data from Jeju. The thicknesses and moduli of the subgrade, subbase, and asphalt layers were incorporated into the design. The regression-analysis program SPSS was used to develop a regression equation for the overlay design, factoring in the modulus and thickness ratios between the existing and overlay asphalt layers. A pavement-thickness design formula tailored to Jeju's characteristics was derived. An equivalent single-axle load factor (ESALF) formula was developed to facilitate traffic-load estimation for different roads, enabling the easy incorporation of varying traffic volumes into the design. The ESALF formula demonstrated a high correlation with the pavement thickness, subgrade conditions, and axle loads, whereas the pavementthickness design formula exhibited strong correlations with the pavement thickness, subgrade state, thickness ratios, and modulus ratios. The use of basalt aggregates in asphalt concrete pavements provides an economically viable and technically sound solution for Jeju. The proposed design methodology not only reduces costs but also enhances pavement performance and road safety. The developed formulas offer flexibility in adjusting designs based on specific traffic conditions, providing optimal pavement solutions for different road categories.

This study aimed to develop a systematic process for identifying components that need to be changed to reduce the Head Injury Criterion (HIC) during pedestrian headform tests. Through simulation and analysis, it was confirmed that the hood, hinge, hinge plate, cowl, fender, and fender bracket significantly influence HIC15. The study identified the specific impact of each component on HIC15, allowing for targeted improvements. The proposed process demonstrated superior performance compared to single-component optimization, yielding more significant reductions in HIC15. Multiple vehicle models were tested, confirming the process's effectiveness in consistently lowering HIC15 values.

최근 인공지능 분야에서 가장 활발히 연구되고 있는 거대 언어 모델은 교육에 대한 응용 가능성을 보 이며, 교육학의 거의 모든 분야에서 그 활용 방안이 연구되고 있다. 이러한 연구는 공학 교육에서도 주목 받고 있다. 그러나 구체적인 활용 분야와 방법에 대해서는 아직 많은 연구가 필요한 상황이다. 특히, 거대 언어 모델을 이용한 교육과정 설계와 개선에 대한 연구는 인공지능 공학과 교육학 두 분야에서 모두 중요한 연구 과제로 부각되고 있다. 이러한 응용 필요성에 대한 예시이자 전략으로써, 본 연구는 OpenAI에서 발표한 최신 거대 언어 모델인 ChatGPT-4o를 이용하여 한국과학기술원(KAIST) 공과대학 학부 전공 과 목과 S전자 DS부문(반도체사업부) 직무 사이의 연관성을 분석하고, 그 결과를 기반으로 대학과 기업체 양측에 반도체 산업 인력 양성과 채용에 대한 실질적인 응용 전략을 제안한다. 이를 위해 KAIST 공과대 학 학부과정에 개설된 모든 전공 과목과 S전자 DS부문(반도체사업부)의 직무기술서를 ChatGPT-4o에 학습시켜 각 과목이 특정 제품군, 직무와 가지는 연관성을 특정 범위와 기준에 의거하여 정량화된 점수로 평가했다. 또한, 각각의 직무, 전공, 과목별로 확보한 데이터를 기초적인 통계 분석을 통해 평가했으며, 구직자와 구인자의 활용 가능성에 초점을 두고 특정 전공의 각 직무별 연관성과 특정 직무의 각 전공별 연관성, 그리고 특정 직무 및 전공의 반도체 제품군별 연관성 등 다양한 조건에서 분석을 진행하였다. 또 한 본 전략에 대한 반도체 산업 실무자 견해를 수집하여 실제 전략으로의 활용 가능성을 검증하였다. 분 석 결과, 간단한 질문과 분석만으로도 전공, 교과목별로 유의미한 직무 연관성의 차이를 확인했다. 이러한 결과를 바탕으로 본 연구는 대학 교육과정의 개선과 기업 채용 및 양성 과정에서의 응용 전략을 제시한 다. 이 연구는 대학과 산업 간의 협력을 통해 인적자원 개발과 채용 효율성 증대에 기여할 것으로 기대한 다. 또한, 후속 연구로 구직자와 구인자, 교수자 등 본 연구의 효과를 확인할 수 있는 집단을 대상으로 한 대규모 설문조사 및 전문가그룹 대상 질적연구 등을 제안하여 실제 활용도와의 비교 분석 연구를 제안 한다. 결론적으로, 본 연구는 거대 언어 모델을 활용하여 필요한 인재를 양성하기 위한 교육 과정 설계의 구체적인 응용 가능성을 제시함으로써, 인공지능을 이용한 교육 분야에 대한 기여 방안을 모색한다.

Wearable technology is expected to maintain continuous marketability and prospects, with its scope gradually expanding beyond the fashion sector to encompass fashion accessories. Meanwhile, the wedding industry is currently reflecting consumer preferences that emphasize individuality and emotional connection. As wedding trends evolve, there is a growing interest in unique and differentiated wedding styles. Therefore, the purpose of this study is to create high-value designs by integrating wearable LED technology into wedding accessories and dresses to meet the emotional needs of modern consumers. To achieve this, we analyzed the LED wedding accessories currently available in the market. Based on the findings, we designed and developed new such accessories and dresses through planning, development, and production processes. First, the study found out that LED wedding accessories are gaining attention as high-value products. Second, a survey of the domestic market for LED wedding accessories highlighted the needs for wedding dress designs that can be paired with LED hairpins. Third, we used Lilypad Arduino’s Lily Tiny to design and develop LED wedding hairpins and dresses through a production process. Finally, by styling LED wedding hairpins and dresses together, we demonstrated the potential in creating products that blend emotion and technology, in line with the current wearable technology trends. Overall, this study offers a fresh perspective on design development in wedding accessories.