Asian Journal of Innovation and Policy, Asian Society for Innovation and Policy

PDF Citaion Measuring the Impact of Informatization on SMEs Business Processes : Based on a Machine Learning Analysis Approach

Yoonseo Won ; Jongtae Lee pp.251-277 https://doi.org/10.7545/ajip.2025.14.3.251

초록보기

Abstract

This study aims to suggest a machine learning-based analysis model to predict the business process effectiveness and to analyze the informatization level survey data of small and medium-sized enterprises (SMEs). The study also focuses on identifying the most effective methodology and providing practical insights for establishing informatization strategies. This study predicts the business process effectiveness of small and medium-sized enterprises (SMEs) by utilizing the survey data of their informatization level. Representative machine learning classification models - Random Forest, XGBoost, and LightGBM - were applied, and SHAP (SHapley Additive exPlanations) was then used to analyze the key variables influencing the prediction performance with the highest-performance analysis model. According to the findings, the Random Forest and LightGBM models demonstrated the best performance in terms of AUC and accuracy for predicting business process effectiveness. A SHAP analysis suggested that the informatization capabilities of enterprises and the use of appropriate information systems could be key variables for business process effectiveness.

PDF Citaion The Impact of Crisis-Responsive Dynamic Capabilities on Firm Performance in the Post-COVID-19 Era: Evidence Across Technology Levels

Minseo Kim ; Alden Kim ; Yeong-wha Sawng pp.278-298 https://doi.org/10.7545/AJIP.14.3.278

초록보기

Abstract

The COVID-19 pandemic heightened environmental uncertainty and challenged efficiency-oriented innovation strategies. This study examines how crisis-responsive dynamic capabilities affect firm performance in the post-COVID-19 era and whether these effects differ across technology levels. Using panel data on 174 firms from the Survey on Technology of Small and Medium Enterprises (2022– 2023), dynamic capabilities are measured by changes in R&D intensity and shifts in innovation types, while performance is captured by sales growth. The results show that dynamic capabilities do not uniformly improve short-term performance; changes in R&D intensity exhibit negative and nonlinear effects, reflecting adjustment costs and learning delays. Importantly, these effects vary by technology level, with pronounced nonlinear patterns observed among medium-high- and high-technology firms. The findings highlight the contingent and nonlinear nature of dynamic capabilities in the post-pandemic context.

PDF Citaion Analysis of Technological Trends in Aviation Industry Using Topic Modeling: The case of Korea

Ju-Yong Song ; Sang-Wan Jo ; Jeonghwan Jeon pp.299-329 https://doi.org/10.7545/AJIP.14.3.299

초록보기

Abstract

This study analyzes technological trends in Korea’s aviation industry using patent data. Latent Dirichlet Allocation (LDA) topic modeling is applied to domestic aviation technology patents to identify major technology areas and emerging themes, yielding ten topics. Yearly changes in each topic are examined through linear regression to quantify trend directions. The results show that “unmanned aerial vehicle (UAV) operation” is a promising and expanding field, while “rotor” and “manufacturing process” technologies exhibit declining trends, indicating that Korea’s aviation technology focus is shifting toward UAV-related domains. In addition, the technological competitiveness of patent applicants is assessed, revealing that established technology areas are largely dominated by global corporations, whereas competition among leading Korean firms is intensifying in emerging market fields. These findings provide evidence-based guidance for understanding Korea’s aviation technology landscape and for setting future R&D strategy priorities.

PDF Citaion Privacy Governance-Driven Design of AI-Powered Elderly Safety Monitoring for Cambodia

Monireach Tang ; Seingheng Hul ; May Thu pp.330-362 https://doi.org/10.7545/AJIP.14.3.330

초록보기

Abstract

Cloud-based elderly monitoring systems create surveillance risks with limited data protection enforcement. This study validates technical feasibility and cost-effectiveness of edge-based pose estimation for elderly safety monitoring in resource-constrained contexts, demonstrating how privacy principles can drive architectural decisions from inception. The proposed architecture comprises four 850nm NIR cameras and an NVIDIA Jetson Orin Nano edge platform. Privacy protection emerges from system constraints: immediate conversion of video frames to skeletal coordinates, pose-only storage, and permanent video deletion. The integrated YOLOv8n and MediaPipe pipeline achieves 100% on-device processing. Validation on 20 commercial CCTV videos demonstrates 91.3% keypoint detection at 20.53 FPS, confirming 24/7 monitoring capability without facial recognition technology. Edge architecture reduces 3-year total cost by 61% compared to cloud alternatives, expanding market accessibility to 168,000–252,000 elderly individuals in middle-income Cambodian urban households. This work demonstrates that governance principles can drive technical architecture in healthcare AI for Cambodia.

PDF Citaion Enhanced Thermal Management of Electric Vehicle Batteries Using Graphene-Based Nanofluid Coolant

Bernard Sagaiyaraj ; Sofea Zawawi ; Fareez Allany pp.363-379 https://doi.org/10.7545//ajip.2025.14.3.363

초록보기

Abstract

Efficient thermal management of prismatic lithium-ion batteries is critical for electric vehicle (EV) performance, safety, and longevity. The thermal conductivity of conventional water-based cooling systems is restricted, which raises battery temperature and increases energy consumption. This study examines AQUENE, a distinctive mixture of graphene-based nanofluid as an advanced coolant for prismatic battery packs. This graphene-based nanofluid offers improved thermal conductivity, stability, and minimal increase in viscosity compared to water. It consists of 1% functionalized graphene dissolved in deionized water. A 1.5-RT water-cooled chiller connected with a 4S1P prismatic battery pack under controlled charge-discharge cycles was implemented for experimental evaluation. Battery surface temperature, chiller energy consumption, pumping power, and energy efficiency ratio were the key performance indicators. AQUENE decreases the peak battery temperature by 1.01% compared to water. Energy efficiency improved by 23.43%, while chiller compressor energy consumption decreased by 22.46%. The results demonstrate that AQUENE enhances convective heat transfer, reduces thermal stress, and improves system energy efficiency without significant retrofitting. Implementation in EV battery thermal management systems can extend battery lifespan, enable more compact cooling designs, and reduce operational energy demand, contributing to sustainable mobility and decarbonization goals.

PDF Citaion Multiclass Classification of Coast in 3D LiDAR Point Clouds for Shoreline Erosion Monitoring

Sannareth Sou Sannareth Sou ; Surasak Boonkla Surasak Boonkla ; Hongly Va Hongly Va ; Jessada Karnjana Jessada Karnjana pp.380-401 https://doi.org/10.7545/AJIP.14.3.380

초록보기

Abstract

Coastal erosion poses a serious threat to ecosystems, infrastructure, and coastal livelihoods. Traditional monitoring using satellite or aerial imagery often struggles to distinguish between forest, beach, and sea regions, particularly in complex coastal environments. This study aims to enhance shoreline monitoring accuracy by introducing a LiDAR-based multiclass classification framework that effectively differentiates these terrain types. The proposed method integrates geometric and color-based features extracted from UAV-based LIDAR data collected in Khlung, Chanthaburi, Thailand. Six features were derived from segmented point cloud grids: elevation variability (v1), directional slope variability (v2), and the mean RGB values (Rmean, Gmean, Bmean, RGBmean). Four machine learning models, Random Forest, Decision Tree, Support Vector Machine, and Logistic Regression, were evaluated using cross-validation on 1500 samples. Results indicate that the Random Forest classifier achieved the highest accuracy of 99.78%, outperforming other models. The proposed approach effectively separates forest, beach, and sea regions, reducing misclassification errors and improving the reliability of coastal terrain classification. This study demonstrates the potential of combining LiDAR point cloud geometry and radiometric features for robust multiclass classification. The findings support more precise and automated coastal erosion analysis, offering a foundation for future integration with deep learning and hybrid image point cloud systems.

바로가기메뉴