This study proposes a dynamic evaluation framework for diagnosing signal control adequacy using high-resolution Automated Traffic Signal Performance Measures (ATSPM) data. Traditional signal performance assessments have primarily relied on aggregated metrics, such as average delay and volume-to-capacity ratio, which are effective for evaluating overall operational efficiency but insufficient for capturing cycle-level control limitations and temporal variability. Although split failure-based measures, including the Split Failure Ratio (SFR), provide more direct insights into green time adequacy, most existing applications focus on the failure frequency within a fixed analysis period. To address this limitation, this study introduces a Dynamic Operational Strain (DOS) index that extends the split failure into a time-evolving state variable incorporating accumulation and recovery mechanisms. By modeling the recursive evolution of the operational strain, the proposed framework captures how often failures occur and how they persist or dissipate over time. Phase-level DOS measures are subsequently aggregated at the intersection level to derive a priority score reflecting structural control inadequacy. The framework is further applied to classify intersections using DOS–SFR quadrant analysis, enabling the identification of distinct operational patterns, such as persistent oversaturation, localized phase imbalance, intermittent strain accumulation, and stable control conditions. The results demonstrate that intersections with similar SFR values may exhibit substantially different temporal strain structures, highlighting the importance of a dynamic state-based evaluation. The proposed approach provides a diagnostic foundation for data-driven signal re-timing and future adaptive control strategies by shifting the signal performance assessment from static frequency-based measures to dynamic structural adequacy analysis.

External steel frame retrofitting effectively enhances the lateral resistance performance of existing reinforced concrete school buildings. However, when column shear failure occurs, ductility capacity is reduced, posing a risk of sudden collapse in high-seismic hazard areas. The purpose of this study is to propose a linear-elastic analysis-based retrofit process to reduce the likelihood of column shear failure and maximize retrofit effectiveness when applying external steel frame retrofitting. To achieve these objectives, a multiple-step process was presented and a case application was performed using quasi-static cyclic loading test results of a school building specimen retrofitted with an external steel frame system. Experimental results showed that strength, stiffness, and energy dissipation were improved by the retrofit system. However, the damage mechanism shifted from beam-column joints to column bases, resulting in reduced ductility capacity due to column shear failure. By applying the proposed process, the lateral stiffness ratio of the external steel frame columns was increased from 0.23 to 0.74, eliminating shear failure in existing reinforced concrete columns and confirming that it can contribute to improving ductility capacity.

This study proposes a data-driven framework for analyzing freeway driving behavior using multiple real-world trajectory datasets, and applies it consistently to mainline and ramp sections. The four large-scale datasets—namely highD, exiD, NGSIM I-80, and NGSIM US- 101—were processed through a unified preprocessing pipeline that converted all variables to International System Units(SI), resampled trajectories to 10 Hz, applied Savitzky-Golay smoothing to speed, and removed physically implausible and statistical outliers based on joint physical-statistical criteria. For each vehicle, 24 summary features were constructed from six longitudinal indicators–speed, acceleration, deceleration, time headway (THW), distance headway (DHW), and time-to-collision (TTC)–using their minimum, maximum, mean, and standard deviation. Indicator distributions by road type were compared using relative frequency histograms with common binning; then, principal component analysis (PCA) and K-means clustering were applied independently to each dataset. The leading principal components revealed interpretable axes related to longitudinal driving intensity (speed and acceleration level), safety margin (THW/DHW/TTC), and onramp sections; responsiveness was characterized by acceleration-deceleration variability, as observed within the analyzed datasets. Cluster interpretation yielded four relative driving behavior categories–aggressive, responsive, stable, and defensive–defined within each dataset based on indicator levels and variability rather than absolute thresholds.

최근 메타버스 기술의 확산으로, 2D 도면을 넘어 3D 공간을 직접 시뮬레이션하려는 실내 디자인 서비스 수요가 급증하고 있다. 그러나 기존 3D 재구성 기술은 높은 연산량에 따른 긴 처리 시간과 실제 공간과의 시각적 괴리, 그리고 재구성 후 수정이 어려워 실시간 서비 스 적용에 한계가 있다. 본 논문에서는 이러한 문제를 해결하기 위해 기존 PE3R 모델을 획 기적으로 개선한 IN3R(Interior New 3D Reconstruction) 프레임워크를 제안한다. IN3R은 YOLOE 기반의 단일 통합 파이프라인을 도입하여 객체 탐지, 분할, 의미 부여 과정을 경량 화함으로써 불필요한 연산을 제거하고 재구성 속도를 최적화하였다. 또한, 생성형 LLM을 결합하여 공간의 맥락을 분석하고, 사용자가 텍스트 프롬프트를 통해 가구 배치와 분위기 를 능동적으로 편집할 수 있는 인터랙티브 기능을 구현하였다. 실험 결과, 제안 모델은 기 존 PE3R 대비 재구성 품질의 저하 없이 평균 3.65배 빠른 처리 속도를 달성하였다. 본 프 레임워크는 실내 재구성 기술의 실용성을 확보하여, 향후 사용자 맞춤형 가상 인테리어 및 이커머스 연계형 3D 쇼룸 시스템의 핵심 기술로 활용될 것으로 기대된다.

선박 충돌 위험 평가는 항해 안전 확보를 위한 핵심 절차로, 실시간 의사결정 지원과 회피 기동 판단의 기초를 제공하며, 특히 자율운항선박 시대의 안전성 확보를 위해 그 중요성이 커지고 있다. 기존 연구들은 다양한 시나리오에 맞춘 지표와 모델을 제안해 왔으 나, 충돌 위험 변수와 변수 모델링 방법론 간의 구조적 연계에 대한 통합적 분석은 부족한 실정이다. 본 연구는 PRISMA 2020 지침에 따라 문헌 검토 절차를 체계화하고, 자율운항선박 기술개발이 본격적으로 진행된 2020년부터 2025년까지 발표된 관련 논문 중 75편을 선정하 여 선박 충돌 위험도 평가에 사용된 변수와 방법론을 각각 다섯 개의 범주로 분류하였다. 이를 통해 각 요소의 출현 빈도와 조합 경향을 통계적으로 분석하였으며, 변수-방법론 연계 행렬을 통해 연구 경향을 시각화하였다. 분석 결과, 대부분의 충돌 회피 모델이 여전히 운동 학 기반의 거리 중심 지표에 의존하고 있으며, 인간 요인이나 맥락 조건에 기반한 위험 평가는 상대적으로 적게 다루어졌다. 본 연구는 이러한 통합적 분석을 통해 향후 충돌 위험 평가 연구에서 확장되어야 할 변수와 기법의 방향을 제안하며, 제시한 5축 기반 분류체계는 향후 관련 연구자들이 연구 목적에 따라 적절한 변수와 방법론을 선택하고 설계하는데 유용한 개념적 틀로 활용될 수 있을 것으로 기대 한다.

This study investigates the framing patterns and content of news coverage surrounding the construction of the Islamic mosque in Daehyeon-dong, Buk-gu, Daegu. A total of 433 articles containing the keywords “Daegu” and “Islam” were collected from Naver News between 2021 and 2024 and coded according to the frames proposed by Bowe and Makki (2016). Subsequently, a corpus linguistic analysis was conducted for each frame to examine the keywords, collocations, and framing devices quantitatively and qualitatively. The findings revealed that, in the early stages of the event, the major frames appeared relatively balanced; however, over time, the reporting became fragmented, emphasizing either conflict or condemnation of hate. Furthermore, the distribution of keywords and collocations within each frame demonstrated that the same lexical items served different semantic roles depending on the frame. The use of framing devices also enabled a more detailed observation of how frames were constructed at the textual level. This study is significant in that it adopts an interdisciplinary approach combining framing analysis and corpus linguistics to provide a multidimensional understanding of how real social conflicts in multicultural contexts are constructed within news discourse.

Defect detection in manufacturing processes is a critical requirement for ensuring product reliability and maintaining production stability. As smart manufacturing environments continue to advance, the need for precise and robust vision-based inspection methods has become increasingly significant. This study proposes a hybrid defect analysis framework that integrates YOLOv5-based defect candidate detection with an Attention U-Net–based segmentation module. Experiments conducted on chromate-coated industrial images demonstrate that the proposed framework achieves an accuracy of 0.97, precision of 0.91, recall of 0.89, F1-score of 0.93, and IoU of 0.88, exhibiting stable performance even for small defects and irregular boundaries. The combination of region- of-interest extraction and attention-enhanced pixel-level segmentation improves both computational efficiency and boundary reconstruction quality. The findings extend the applicability of attention-based segmentation to industrial defect inspection and provide practical insights for deploying deep learning–based quality monitoring systems in automated manufacturing environments.

Freight-rate forecasting in the VLCC TD3C market remains challenged by abrupt regime shifts, pronounced volatility, and heterogeneity in real-time signals from oil prices, seaborne trade, vessel operations, and macroeconomic factors; these directly impact freight planning and chartering. This study presents a daily multivariate dataset with 4,267 samples covering 2014-02-01 to 2025-10-08, integrating crude benchmarks, fuel spreads, refinery margins, port congestion, inventory levels by region, plus detailed AIS-derived VLCC activity, speed, and operation states, scaled and split 80/10/10 for training, validation, and testing. The proposed framework combines a PyTorch Transformer—optimized using Optuna for d_model=128, 9 layers, 8 heads, a 14-day input window, and 5-day output—with Monte Carlo Dropout for uncertainty quantification. Diagnosis uses differential entropy and coefficient-of-variation to verify convergence with 90 separate runs, while a Kalman filter (Q=0.001, R=0.01) smooths the forecast trajectory and enhances temporal reliability. Experimental results show baseline Transformer achieves average MAE 5,259.4, MAPE 13.10%, and R²=0.74 across 1-5 day horizons, with volatility quality metrics declining at longer leads. Applying the Kalman filter reduces errors to MAE 4,326.1, MAPE 10.6%, and raises R² to 0.83; timing and extremity components of volatility quality scores are strengthened, providing a more robust basis for operational decisions. Monte Carlo backtesting for 82 Korean VLCCs over 598 trades finds the Kalman-smoothed strategy earns $108.5M (88.9% win rate, Sharpe ratio 0.83), substantially outperforming raw Transformer ($32.9M, 60.5%, 0.24) and random selection (near zero, 49.3%, 0.005). These results highlight the clear economic value added by calibrating uncertainty and post-processing forecasts, transforming predictive reliability into real-world freight portfolio improvement in the tanker market.

본 연구는 미국과 영국의 주요 언론사인 뉴욕타임스, 워싱턴포스트, 더 타임스, 가디언을 대상으로, 2014년부터 2024년까지 인공지능(AI) 관련 보도량을 검토하고, 2023년에서 2024년까지 ‘챗GPT’ 또는 ‘오픈AI’를 다룬 346편의 기사를 중심으로 핵심 의제와 뉴스 프레임을 분석하였다. 키워드 분석과 정성적 문맥 검토를 적용한 결과, 양국 언론은 공통적으 로 ‘챗봇’과 ‘생성’ 키워드를 중심으로 ‘생성형 AI’를 핵심 의제로 부각하 였다. 미국 언론은 ‘학생’ ‘데이터’ ‘레딧’ 등을 키워드로 교육 현장의 윤 리와 규제 쟁점, 데이터 소유권 등 정보 거버넌스 의제를 비교적 현저하 게 다루었고, 영국 언론은 ‘에너지’ ‘법’ ‘생산성’ 키워드를 통해 AI 운영 에 따른 물리적 인프라 부담, 영국 내 기업의 생산성 문제를 중심으로 환경 및 거시경제적 효과를 주로 강조하였다. 본 연구는 이러한 분석을 바탕으로 글로벌 미디어 환경 속에서 인공지능(AI) 관련 국내 정책 개발 과 사회제도적 대응을 모색하기 위한 기초적 논의의 틀을 제시한다.

본 연구는 관계외상을 경험한 중년 여성이 코칭적 맥락(context)에서 소명의 식을 어떻게 재정립하는지를 Van Manen의 해석학적 현상학으로 탐색하였다. 이를 위해 40–50대 여성 4명을 목적표집하여 반구조화 심층면담을 진행하고, 텍스트를 주제화한 뒤 해석적 글쓰기로 분석하였다. 분석 결과는 실존체의 네 차원(신체성·공간성·시간성·관계성)으로 정리 되었다. 신체성에서는 외상의 체 화와 감정 억압이 드러났고, 공간성에서는 안전 부재의 가정에서 공동체로의 이 동이 회복의 토대로 작동하였다. 시간성에서는 과거 상처의 반복–재해석–재 서사화의 흐름이 확인되었으며, 관계성에서는 단절에서 지지·공명으로의 전환 이 자기 이해의 확장을 이끌었다. 이 과정에서 코칭은 정서 탐색과 자기 성찰을 촉진하는 매개 맥락으로 기능했고, 소명의식은 특히 시간성과 관계성이 교차하 는 지점에서 재구성·확장되는 것으로 해석되었다. 이러한 결과는 외상 경험이 존재 이해의 전환으로 이어질 수 있음을 시사하며, 실존체-정렬형 코칭 프레임 의 기초 구성을 제시함과 동시에 표본 규모가 제한적이라는 점에서 일반화에는 신중함이 요구되지만 임상·교육·코칭 현장에서의 적용 가능성을 시사한다.

The purpose of this study is to design a participatory framework for voxel-LED public art that integrates artificial intelligence for sustainable urban engagement. Using a Research through Design (RtD) approach, the framework combines a voxel-LED installation with an interactive kiosk, enabling citizens to co-create content through generative AI and to experience adaptive responses based on environmental data. The proposed system indicates both technical feasibility and design potential, offering two core scenarios: generative co-creation and environment-aware adaptation. While still conceptual in scope, the results specify a replicable model for participatory urban art and provide a roadmap for future studies to evaluate user interaction and advance placemaking-oriented practices.

In this study, we propose a data-driven analytical framework for systematically analyzing the driving patterns of autonomous buses and quantitatively identifying risky driving behaviors at the road-segment level using operational data from real roads. The analysis was based on Basic Safety Message (BSM) data collected over 125 days from two Panta-G autonomous buses operating in the Pangyo Autonomous Driving Testbed. Key driving indicators included speed, acceleration, yaw rate, and elevation, which were mapped onto high-definition (HD) road maps. A hybrid clustering method combining self-organizing map (SOM) and k-means++ was applied, which resulted in eight distinct driving pattern clusters. Among these, four clusters exhibited characteristics associated with risky driving such as sudden acceleration, deceleration, and abrupt steering, and were spatially visualized using digital maps. These visualizations offer practical insights for real-time monitoring and localized risk assessment in autonomous vehicle operations. The proposed framework provides empirical evidence for evaluating the operational safety and reliability of autonomous buses based on repeated behavioral patterns. Its adaptability to diverse urban environments highlights its utility for intelligent traffic control systems and future mobility policy planning.

공유 유기 골격체(covalent organic frameworks, COF)는 기능성을 정밀하게 설계하고 제어할 수 있는 결정성 다 공성 소재로서, 차세대 연료전지 멤브레인으로 주목받고 있다. 표준 양성자 교환막인 나피온(Nafion)은 높은 비용과 좁은 가 용 범위 등의 한계에 직면해 있다. 본 논문은 COF를 다양한 고분자 매트릭스에 도입하여 이러한 단점을 극복하기 위한 최신 연구 전략을 심도 있게 다룬다. 특히 양성자 교환막 연료전지(proton exchange membrane fuel cells, PEMFC), 음이온 교환막 연료전지(anion exchange membrane fuel cells, AEMFC), 그리고 고온(high-temperature) PEMFC (HT-PEMFC)용 COF 기반 복합막의 설계와 성능 특성에 집중한다. 다양한 COF 기능화 및 복합화 전략을 통해 이온 전도도, 기계적 강도 및 운전 안정 성을 향상시킨 주요 연구들을 비평적으로 논하며, 연료전지의 전반적인 효율 향상에 대한 COF의 잠재력을 조명한다.

This study investigates how the term burnout was framed, transformed, and made to function in Korean public discourse from 2010 to 2024. Drawing on 9,206 news articles retrieved from the BigKinds database, the research applies an integrated analytical framework combining Entman’s (1993) four framing functions with Leventhal’s Common-Sense Model (CSM) of illness representation. Five dominant frames —medicalisation, individualisation, normalisation, risk framing, and metaphorisation— were identified through quantitative content analysis and qualitative coding with MAXQDA 2024 (qualitative and mixed-methods analysis software). Results show a marked shift from a medicalised frame in Period 1 (2010-2015) to an increasingly normalised frame in Period 3 (2021-2024), indicating a recontextualisation of burnout from an individual pathological condition to a shared everyday experience. Statistical analyses, including chi-square tests, interrupted time series analysis, and keyword network mapping, reveal significant temporal changes in frame distribution, functional complexity, and co-occurrence patterns. The framing functions were found to systematically structure CSM dimensions, shaping public perceptions of identity, causation, temporality, consequences, and controllability. These findings demonstrate that the reframing of burnout reflects not only shifts in linguistic representation but also broader socio-cultural and political reconfigurations, with implications for health communication, organisational policy, and public health strategies.

The semiconductor equipment industry is characterized by high technological intensity and a strong demand for customized functionality from client firms. This necessitates the collection and implementation of diverse client requirements, which in turn demands close cross-departmental collaboration and leads to highly complex project structures. To manage this complexity, many semiconductor-related companies operate global matrix organizations based on multiple reporting lines by region, function, and product. These structures entail intricate communication process and pose significant collaboration challenges. This study empirically investigates key factors influencing collaboration and project performance within global matrix organizations in the high-tech semiconductor equipment industry, applying the Technology-Organization-Environment (TOE) framework. The analysis identified significant effects of all examined factors on inter-organizational collaboration performance: three technological sub-factors (usability of communication tools, process standardization, and change management systems), two organizational sub-factors (clarity of roles and responsibilities, and robustness of information-sharing systems), and three environmental sub-factors (alignment of requirements, cultural understanding, and responsiveness to industry standards and regulations). Furthermore, the level of agile implementation was tested as a moderating variable in the relationships between TOE factors and collaboration performance. The results revealed significant moderating effects of agility in specific areas: the usability of communication tools and systematic change management (technological factors); information-sharing structure (organizational factor); and requirement alignment (environmental factor). These findings suggest that agile approaches do not operate as a single-factor solution but interact dynamically with various organizational conditions. By focusing on the underexplored dynamics of global matrix organizations in collaboration performance, this study provides a structured empirical analysis of their operational characteristics. Its theoretical contribution lies in offering an integrative perspective on technological, organizational, and environmental drivers of collaboration and project success, extending current research in high-tech project and organizational performance.

This study addresses the limitations of traditional Failure Modes and Effects Analysis (FMEA), which heavily relies on expert judgment and lacks the ability to effectively incorporate unstructured failure history data such as warranty claims and maintenance records into the design stage. To overcome these challenges, we propose an automated FMEA framework based on a Retrieval-Augmented Generation (RAG) architecture integrated with a Local Large Language Model (LLM) in a secure, locally managed environment. The proposed system stores claim and test data in a vector database and leverages the LLM to retrieve and analyze relevant information, enabling automatic extraction of new failure modes and dynamic updates to FMEA documents. Additionally, the system recalculates Risk Priority Number (RPN) by adjusting severity, occurrence, and detection ratings when recurring failures are detected. To improve response quality, we applied prompt engineering and optimized chunking parameters during data retrieval. This research demonstrates the feasibility of achieving a life cycle-integrated quality enhancement framework throughout the product lifecycle while ensuring data security.