Performance Verification of Road Traffic Survey Equipment Based on AI Image Recognition
Abstract : This study aims to address the limitations of conventional embedded-type traffic volume survey equipment and to examine the field applicability and effectiveness of a non-intrusive traffic volume survey system utilizing artificial intelligence (AI)-based video recognition technology. For this purpose, embedded AVCs, license plate recognition-based AVCs, and AI-based video recognition AVCs were installed under identical conditions at pilot sites on National Routes 34 and 42.July to November 2024, performance evaluations were conducted under various weather conditions (daytime, nighttime, sunrise, fog, and rain) and time periods. The AI AVC demonstrated the highest vehicle classification accuracy across all conditions (ranging89.5% to 95.9%) and maintained a stable traffic volume accuracy of 92.4% to 96.7%. Particularly under challenging environments such as nighttime, rainfall, and fog, the AI AVC maintained high recognition performance through -based deep learning algorithms, proving superior adaptability compared to traditional embeddedlicense plate-based devices. While the embedded AVC showed excellent traffic volume accuracy (up to 97.8%), it exhibited limited classification accuracy (83~87%). The license plate AVC, on the other hand, showed significant performance fluctuations due to sensitivity to environmental changes. In conclusion, the AI-based video recognition AVC demonstrated excellent detection performance and operational flexibility in real-world conditions, indicating strong potential as a replacementcomplementary technology to existing traffic survey methods. The results of this study may serve as foundational data for transitioning to intelligent transportation system (ITS)-based information collection frameworks and for the nationwide advancement of traffic volume survey systems.
Keyword : AI AVC, Embedded AVC, License-plate AVC, Traffic monitoring, Vehicle classification
http://dx.doi.org/10.29056/jncist.2025.10.01
A Study on Short-Term Call Option Price Prediction Using eXplainable Artificial Intelligence
Abstract : This study aims to validate the performance of machine learning-based option pricing prediction model and ensure transparency in the prediction by utilizing explainable artificial intelligence (XAI) technique. I conducted a comparative analysis of the predictive performance between the Black-Scholes and the XGBoost models using call option dataApple Inc. Both models employed the same input variables including underlying asset price, strike price, days to expiration, time-varying risk-free rate, and 30-day historical volatility, with actual call option prices as the target output variable. The experimental results demonstrate that the XGBoost model significantly outperformed the Black-Scholes model in predictive accuracy. The XGBoost model achieved approximately 87.9% performance improvement based on MSE, with error reductions of 62.9% and 65.3% in MAE and RMSE, respectively. Through actual versus predicted call option price comparisons, I confirmed that the XGBoost model exhibited consistent predictive accuracy across all price ranges. Feature importance analysis using SHAP identified underlying asset price and strike price as the most critical variables for option price prediction. This study distinguishes itselfexisting research by employing XAI technique to provide transparency in machine learning-based option price prediction.
Keyword : Business Analytics, Financial Engineering, Financial Mathematics, Financial AI, Financial Derivatives
http://dx.doi.org/10.29056/jncist.2025.10.02
Abstract : With the rapid advancement of Artificial Intelligence (AI) technologies, universities are increasingly integrating AI tools to enhance learning outcomes. This study explores the psychological mediators that link the educational use of AI with students’ learning outcomes in higher education and derives practical implications for its effective integration in university teaching. To this end, a systematic review was conducted on prior studies addressing AI utilization, academic self-efficacy, learning engagement, and self-regulated learning, followed by an analysis of how AI implementation contributes to learning performance. The findings indicate that the application of generative AI in higher education positively influences learners’ academic self-efficacy, learning engagement, and self-regulated learning. Moreover, the educational use of AI was found to improve learning outcomes through these psychological constructs. The study concludes by discussing key considerations for the pedagogically effective incorporation of AI into university education. Overall, this research advances the understanding of psychological mechanisms mediating the relationship between AI use and learning achievement, thereby elucidating their respective roles and boundaries in fostering academic success.
Keyword : Artificial Intelligence, Academic Performance, Academic Self-Efficacy, Learning Engagement, Self-Regulated Learning
http://dx.doi.org/10.29056/jncist.2025.10.03
A Study on Personal Branding Education Strategies Using AI-Based Digital Autobiographies
Abstract : The advancement of digital technology offers new opportunities for adult learners to explore and express their identities. In particular, digital autobiographies integrated with AI technology are emerging as a crucial tool for learners to systematically record their experiences and values, thereby reinforcing self-reflection and self-identity. Such digital autobiographies transcend mere historical recording; they can be effectively utilized to construct personal branding. Personal branding, the systematic management of one's strengths, values, and identity to build a unique image, is becoming a key strategy for professional success and social relationship building. This study aims to explore educational strategies that utilize AI-based digital autobiography education to support the self-reflection and personal branding of middle-aged and older adult learners. To achieve this, the concept and educational potential of digital autobiographies were theoretically examined, and a needs analysis was conducted with 30 middle-aged and older learners. Based on the analysis, strategic components were derived, and following an expert validity review, five strategic directions and 15 detailed guidelines were ultimately proposed. These strategic directions include: enhancing self-understanding-centered reflection, a cyclical learning structure focused on practice and feedback, establishing an AI-driven customized feedback and learning support system, fostering an emotionally stable and social sharing environment, and implementing an integrated personal branding model, with three specific guidelines detailed for each strategy. This research empirically demonstrates the educational potential of AI-based digital autobiographies as a learning tool for promoting self-exploration and social identity formation among the middle-aged and older population. It also provides foundational data for the future design and practical application of related educational programs.
Keyword : Digital Autobiography, Personal Branding, Lifelong Education, Adult Learners, AI-based Education, Digital Literacy
http://dx.doi.org/10.29056/jncist.2025.10.04
A Study on Lane Operation Strategies to Improve Traffic Speed at Tunnel Entrances
Abstract : On freeways with two lanes in each direction, the lanes are operated as a passing lane and a travel lane. In sections with many heavy vehicles, substantial delays occur when relatively slow heavy vehicles attempt to overtake on the uphill approaches to tunnel entrances. This study proposes a strategy to minimize lane changes in these uphill approach segments to reduce such delays. To this end, we selected a real-world tunnel-approach segment that experiences recurrent delay, developed scenarios based on traffic volume and the proportion of heavy vehicles, and analyzed each scenario using the microscopic traffic simulation software VISSIM. The results indicate that, for short segments of 100 m, prohibiting lane changes has little effect on average travel speed. By contrast, for kilometer-scale segments, as the controlled segment length increased in 1-km increments—from 570 m (average speed 84.076 km/h) up to 8,570 m (average speed 96.23 km/h)—the average speed rose by 12.154 km/h. Therefore, for repeated uphill tunnel-approach segments, actively adjusting the length of continuous solid white lane lines (lane-change prohibition) to minimize overtaking between passenger cars and heavy vehicles should be considered as an effective measure to reduce overall delay.
Keyword : Tunnel Entrance, recurring Congestion, Average Travel Speed, Lane Change, Uphill Section
http://dx.doi.org/10.29056/jncist.2025.10.05
Development and Validation of a Door-Lock Management System that Predicts Battery Replacement Timing
Abstract : This study proposes a combination-battery-based monitoring solution-comprising an alkalinelithium primary cell and a lithium-ion capacitor (LIC)-to predict battery replacement timing in hotel and lodging facility door locks, and presents the results of prototype implementation and performance verification through accelerated and field experiments. The proposed method leverages the proportional relationship between capacitor voltage and stored energy to indirectly estimate the state of the primary cell, while an ESP32 (Wi-Fi) module periodically measures and transmits the battery and capacitor voltages to a central management system, which determines replacement timing in real time. Accelerated testing identified an additional 163 days of stable operation after end-of-discharge by utilizing residual energy, followed by a 60-day preservation period until complete depletion. Furthermore, extending the lower operating threshold of the alkaline battery3.6 V to 2.8 V demonstrated the potential to more than double the usable lifetime. Field deployment at the Seoul Grand Hyatt, together with controlled laboratory trials, verified the reliability of the entire processdata acquisition and communication to monitoring and alerting. These findings demonstrate that the proposed system enables a predictive maintenance framework for door-lock operations, thereby enhancing the efficiency and reliability of smart hotel and smart building management.
Keyword : Door lock, Battery management, Replacement timing prediction, Lithium-ion capacitor, IoT monitoring
http://dx.doi.org/10.29056/jncist.2025.10.06
Abstract : This study quantitatively evaluates the performance of rockfall protection fences installed along roadside slopes to prevent human and property damage caused by rockfalls and examines the adequacy of the current design standards. The standard rockfall fence used in Korea is designed to absorb approximately 50 kJ of energy when a 400 kg rock fallsa height of 12.5 m; however, it does not sufficiently account for the wide range of rockfall conditions involving various masses and heights that can occur in the field. To address this limitation, the MI-66 CRSP v4.0 simulation program was employed to conduct 1,000 simulations of rockfalls with masses ranging400 kg to 800 kg and heights5 m to 20 m. The maximum and average rockfall energies as well as bounce heights were calculated for each condition. The results indicate that increasing fall height had the greatest effect on rockfall energy, with rock mass also contributing proportionally to energy growth. For conditions exceeding 15 m in height and 500 kg in mass, the maximum rockfall energy surpassed 50 kJ, exceeding the absorption capacity of current fences. Bounce heights increased with fall height but showed relatively little dependence on rock mass. This study quantitatively demonstrates the limitations of the current 50 kJ design standard and proposes reinforcement measures, including upgrading post specifications and installing double-layered wire mesh, to improve energy absorption capacity. These findings are expected to contribute to the establishment of more rational design standards, enhanced road safety, and improved maintenance efficiency.
Keyword : Rockfall Protection Fence, Roadside Slope, Rockfall Simulation, Design Standard, Road Safety
http://dx.doi.org/10.29056/jncist.2025.10.07
Abstract : As the senior generation emerges as a key user group in the digital financial market, there is a growing need for mobile banking service strategies that reflect their characteristics. This study aims to examine whether gamification elements-introduced to promote mobile banking app usage-positively influence not only the MZ generation but also the active senior generation. Through literature review, case analysis, and survey research, appropriate gamification elements were identified, and guidelines for enhancing user experience (UX) in mobile banking apps were proposed. The analysis revealed that, for active seniors, onboarding and progress bars improved comprehension and accessibility in the account inquiry/transfer stage, while progress bars enhanced visibility and recognizability through visual structuring in the spending history/management stage. In the asset management/preference stage, progress bars elicited positive responses in terms of satisfaction and manipulability, whereas onboarding and quests showed relatively low responses. Overall, progress bars consistently generated stable positive responses across all stages, whereas onboarding elicited varying responses depending on stage characteristics. These findings highlight the necessity of stage-specific UX strategies that reflect the characteristics of active seniors when designing mobile banking apps. This study holds academic significance by providing empirical evidence on active seniors, while also offering practical implications and concrete guidelines for financial institutions seeking to develop senior-friendly services.
Keyword : Active Senior, Gamification, Mobile Banking App, User experience
http://dx.doi.org/10.29056/jncist.2025.10.08
Abstract : This study aims to analyze the impact of the usefulness of design fairs on the selection of design sub-majors and to empirically verify the mediating effect of satisfaction in this relationship. Design fairs function as educational environments that provide learning and aesthetic experiences, playing a crucial role in students' major selection processes. While previous research confirmed that the usefulness and satisfaction of design fair experiences influence students' confidence in major selection, it failed to specifically verify the mediating role of satisfaction. Therefore, this study systematically analyzed the mediating effect using hierarchical regression analysis. The research subjects were 44 first-year studentsthe Department of Digital Contents Design at O University located in southern Gyeonggi Province, and a survey was conducted based on their participation in the 2024 Seoul Design Festival. The analysis results showed that both learning utility (β=0.436, p<.05) and aesthetic utility (β=0.480, p<.001) had significant positive effects on satisfaction, and aesthetic utility (β=0.560, p<.05) had a direct impact on major selection helpfulness. The mediation analysis revealed that satisfaction played a partial mediating role in the relationship between aesthetic utility and major selection helpfulness (β=.560→.484, p<.05). This study provides theoretical and practical foundations for improving the quality of design education and career guidance programs by identifying the structural relationships within design fair experiences.
Keyword : Design Exhibition, Major Selection, Mediating Effect, Learning Utility, Aesthetic Utility, Satisfaction
http://dx.doi.org/10.29056/jncist.2025.10.09
An RBV Analysis of SHIFT UP's Competitive Advantage
Abstract : Amid intensifying global competition, sustainable advantage is imperative. Using the resource based view (RBV) and VRIO, this study examines how Korean developer Shift Up converts internal assets into superiority. Drawing on analysis of financials, corporate disclosures, and secondary sources, we identify three reinforcing bundles: an inimitable art style and intellectual property shaped by founder Hyung Tae Kim; an efficient development capability that delivers AAA titles such as Stellar Blade at lower cost and faster time to market; and strategic partnerships with Sony Interactive Entertainment and Tencent that extend distribution, marketing, and capital access. We study 2016 to 2025, including the launches of Destiny Child, Goddess of Victory: Nikke, and Stellar Blade, and the firm’s 2024 IPO. Each resource satisfies VRIO criteria, and together they create a synergistic economic moat that rivals struggle to replicate. The findings enrich RBV scholarship by showing dynamic orchestration of heterogeneous assets in digital game production and offer implications for competitive resilience in creative industries.
Keyword : RBV, VRIO framework, Shift UP, Strategic diversification, Game industry
http://dx.doi.org/10.29056/jncist.2025.10.10
The Multi-layered Magic Circles in Digital Games and Theatre
Abstract : This study aims to analyze the multi-layered structure of the magic circle in digital games and theatre, and to examine the layers and interrelationships of the magic circle that are commonly revealed in both media. The research findings confirm that the magic circle in both media is a topological space where four layers-narrative-representation (L1), performance-interaction (L2), socio-cultural (L3), and meta-critical (L4)-dynamically interpenetrate and operate organically. Digital games demonstrate an overlapping of virtual worlds among avatars, players, and spectators, while participatory theatre forms fluid boundaries where role reversals between audience and actors occur beyond the fourth wall. These permeable boundaries form voluntary communitas and reveal the potential for transformation through the reconstruction of rules via meta-simulation. Furthermore, participants become active agents of meaning-making through a double consciousness that allows them to perceive fiction while simultaneously immersing themselves in the virtual world. The magic circle serves as a space for safe failure where alternative futures can be rehearsed, providing emergent possibilities for democratic participation and cultural production.
Keyword : Magic Circle, Digital Games, Participatory Theatre, Double Consciousness, Meta-Simulation
http://dx.doi.org/10.29056/jncist.2025.10.11
Abstract : After COVID-19, masks have become a medium of visual expression and brand communication beyond the hygiene products used every day. Based on these social changes, this study aims to investigate the effect of visual design factors such as color, shape, logo location, material, printing, and finishing on consumer satisfaction and purchase intention, and to compare cultural differences between Korea and the United States. For general consumers in both countries, we focused on the path through which visual factors lead to consumers' purchase intentions, and to find out this, a conceptual model was established after reviewing previous studies, and factor analysis, regression analysis, structural equation, and mediating effect verification were conducted on online questionnaire data. As a result of the analysis, it can be seen that all five factors had a significant influence on design satisfaction. It was found that satisfaction with the design increased the high purchase intention. Among them, it was found that Korean consumers preferred understated expressions, and in the United States, the preference for individual graphics and bold colors was relatively strong. In conclusion, this study proposes a combination guide and information design strategy of color, shape, logo placement, material, and finish reflecting consumer preferences by country, and presents follow-up studies using actual purchase and repurchase data and group experiments to compensate for cross-sectional design and self-report limitations.
Keyword : Mask design, visual elements, visual elements, user experience
http://dx.doi.org/10.29056/jncist.2025.10.12
Table of Contents