Sejong University · Department of Energy Resources & Geosystems Engineering

📢 News & Updates

📰 Awards

🏆 SMART-X Lab Sweeps Top Honors at the 2026 Mine-Tech Festa — 1st, 3rd, and 4th Places  (2026.09.03)

We are proud to announce another outstanding achievement by SMART-X Lab at the 2026 Mine-Tech Festa, held on September 3, 2026, at the Korea Mine Rehabilitation and Mineral Resources Corporation (KOMIR) in Wonju, Korea.

This year’s competition brought together 39 teams from 12 universities across Korea under the theme of “Resource Security and Supply Chain Strategy, AI Transformation (AX), and Digital Innovation in the Resources Industry.” Following the preliminary evaluation, eight teams advanced to the final round.

Remarkably, all three teams from SMART-X Lab advanced to the finals and finished among the top four teams overall, securing 1st, 3rd, and 4th places.

🥇 K-MIR — Gold Prize (1st Place)
🥈 SMART-X — Silver Prize (3rd Place)
🥈 OptiMine AI — Silver Prize (4th Place)

With K-MIR winning the Gold Prize, students supervised by Prof. Yosoon Choi have now won the top prize at the Mine-Tech Festa for four consecutive years (2023–2026).

🥇 1st Place: K-MIR — Agentic AI for Critical Mineral Supply Chain Intelligence

The K-MIR team, led by Taeheon Kim and three other members, received the Gold Prize (President’s Award of KOMIR) for developing:

“Critical Mineral Supply Chain Early Warning (K-RISK) and Agentic AI Decision Support Platform”

The platform integrates critical mineral data distributed across multiple institutions and introduces K-RISK (Korea Critical Mineral Supply Chain Integrated Risk Index) to quantitatively assess supply chain risks.

Beyond conventional risk monitoring, the system incorporates Agentic AI to analyze emerging supply chain threats, compare alternative response strategies, and support strategic decision-making.

🧠 Key Innovations

• Integrates fragmented critical mineral supply chain data into a unified platform

• Quantitatively evaluates supply chain vulnerability using the K-RISK index

• Detects emerging supply chain risks through an early-warning framework

• Employs AI agents to analyze and compare alternative response strategies

• Supports data-driven decision-making for critical mineral security

As global competition for critical minerals intensifies, the K-MIR platform demonstrates how data integration, quantitative risk assessment, and Agentic AI can be combined to support proactive and intelligent resource security strategies.

🥈 3rd Place: SMART-X — Physical AI for Autonomous Mineral Processing

The SMART-X team, led by Jimin Hong and four other members, received the Silver Prize (Korea Mine Reclamation Association President’s Award) for developing:

“Physical AI-Based Autonomous Operation System for Flotation Processes”

The project explores a new paradigm in mineral processing in which artificial intelligence moves beyond monitoring and prediction to directly interact with and control physical processes.

By integrating AI-based process perception, decision-making, and physical control, the system aims to establish a closed-loop autonomous flotation framework capable of responding dynamically to changing process conditions.

🤖 Technological Impact

• Introduces Physical AI into mineral processing operations

• Connects AI-based perception and decision-making with physical process control

• Enables closed-loop autonomous operation of flotation processes

• Reduces dependence on manual monitoring and operator intervention

• Provides a pathway toward intelligent and autonomous mineral processing plants

This work represents SMART-X Lab’s vision of moving from AI-assisted mining toward truly autonomous resource operations.

🥈 4th Place: OptiMine AI — AI Agents for Autonomous Open-Pit Mine Operations

The OptiMine AI team, led by Minkyoung Jang and another member, also received the Silver Prize (Korean Society of Mineral and Energy Resources Engineers President’s Award) for developing:

“AI Agent-Based Optimization of Ore Haulage Operations in Open-Pit Mines”

The system applies AI agents to open-pit mine operations to analyze operational conditions, evaluate alternative haulage strategies, and support optimized decision-making.

By combining artificial intelligence with mine operation models and simulation, OptiMine AI demonstrates how AI agents can move beyond data analysis and actively participate in operational planning and optimization.

🚚 Why It Matters

• Applies AI agents to real-world mine operation problems

• Supports intelligent planning of ore haulage operations

• Evaluates alternative operational strategies through simulation

• Improves data-driven decision-making in open-pit mines

• Advances the transition toward autonomous and self-optimizing mining systems

The project represents an important step toward mines in which AI does not simply predict operational outcomes, but actively determines how mining operations should be conducted.

🌐 Three Teams, One Vision: AI-Driven Transformation of the Resources Industry

Although the three teams addressed very different problems, their projects share a common vision.

K-MIR applies AI at the strategic level, supporting critical mineral supply chain security and national resource decision-making.

OptiMine AI applies AI at the operational level, optimizing production and haulage decisions in open-pit mines.

SMART-X applies AI at the process level, enabling autonomous interaction with physical mineral processing systems.

Together, the three projects demonstrate a multi-level framework for the digital and autonomous transformation of the resources industry:

Strategic Intelligence → Operational Intelligence → Process Intelligence

This reflects the broader research direction of SMART-X Lab: moving beyond conventional AI prediction toward AI-driven decision-making and autonomous operation of resource systems.

🏆 Four Consecutive Years of Mine-Tech Festa Championships

The 2026 Gold Prize marks the fourth consecutive year that a team supervised by Prof. Yosoon Choi has won the top prize at the Mine-Tech Festa.

2023 → Champion
2024 → Champion
2025 → Champion
2026 → Champion

This four-year winning streak reflects not only the creativity and dedication of our students, but also SMART-X Lab’s continued commitment to transforming innovative research ideas into practical technologies for the mining and resources industry.

Prof. Choi is also the only two-time recipient of the Minister of Trade, Industry and Energy Award for Outstanding Research Supervision at the Mine-Tech Festa, having received the award in 2020 and 2025.

These achievements highlight SMART-X Lab’s sustained excellence in research, student education, technological innovation, and the development of the next generation of professionals for the resources industry.

🎓 From Research to Innovation

The 2026 Mine-Tech Festa results are particularly meaningful because all three SMART-X Lab teams demonstrated different dimensions of the laboratory’s evolving research portfolio:

Critical Minerals + Agentic AI + Autonomous Mining + Physical AI

Rather than applying AI simply as an analytical tool, SMART-X Lab is developing intelligent systems capable of perceiving complex resource environments, making decisions, optimizing operations, and ultimately interacting autonomously with physical systems.

Our goal is to advance from:

AI Prediction → AI Decision → Autonomous Operation

and to develop next-generation intelligent resource systems that can contribute to safer, more efficient, more resilient, and more sustainable mining and resource industries.

📌 About the Mine-Tech Festa

The Mine-Tech Festa is one of Korea’s leading innovation competitions in the mining and resources sector, organized by the Korea Mine Rehabilitation and Mineral Resources Corporation (KOMIR) and supported by the Ministry of Trade, Industry and Energy and major professional societies in the resources industry.

The 2026 competition focused on resource security, critical mineral supply chains, AI transformation (AX), and digital innovation, bringing together talented university students to propose and demonstrate technologies addressing emerging challenges across the resources sector.

We sincerely congratulate all SMART-X Lab members who participated in this year’s competition.

Three finalist teams. Three awards. 1st, 3rd, and 4th places. And four consecutive years of championship wins.

Congratulations to K-MIR, SMART-X, and OptiMine AI! 🎉

Their achievements represent another important milestone in SMART-X Lab’s journey toward building AI-powered, autonomous, and sustainable resource systems — from Earth to Space.

🏆  SMART-X Lab Wins 1st Place at the 2025 Mine-Tech Festa  (2025.09.03)

We are proud to announce that SMART-X Lab has achieved first place (Gold Prize) at the prestigious 2025 Mine-Tech Festa, held on September 3, 2025 at the Korea Mine Reclamation and Mineral Resources Corporation (KOMIR). This national competition recognizes outstanding innovation in the mining and resource industries.

Our winning team, GeoInpainter, presented a groundbreaking technology titled:

"High-Precision Digital Terrain Modeling of Open-Pit Mines Powered by AI Inpainting"

This system utilizes advanced deep learning algorithms (YOLOv8, SAM, LaMa, etc.) to automatically detect and remove large machinery from drone imagery. It then reconstructs the terrain using inpainting and generates a high-precision, equipment-free DEM (Digital Elevation Model). The entire process is fully automated via a user-friendly GUI platform, enabling non-experts to produce clean and accurate DEMs for mining applications.

🧠 Technological Impact

• Removes equipment-induced terrain distortions from drone imagery

• Enhances the accuracy and reliability of open-pit DEMs

• Facilitates digital twin implementation in smart mining

• Applicable to slope stability analysis, blasting design, and volume estimation
  
  

🥉 Additional Recognition: 3rd Place for Team LoRa

Another SMART-X Lab team, LoRa, was also selected as a finalist and received the Bronze Prize (3rd Place) for their innovative system titled:

"LoRa-based Real-Time Truck Monitoring and Production Management System for Open-Pit Mines"

This low-power, cost-effective system integrates LoRa, GPS, BLE, and Starlink technologies to enable real-time truck tracking and production data monitoring in mining environments with limited communication infrastructure. The system has been successfully field-tested at a tin mine in Mongolia, demonstrating its practical value and scalability for resource-constrained mining sites.

🎓 Recognition for Research Leadership

In recognition of this achievement, Professor Yosoon Choi, director of SMART-X Lab, received the Ministerial Award for Outstanding Research Supervision from the Ministry of Trade, Industry and Energy. This marks a remarkable three-year winning streak for Professor Choi, whose teams have consecutively led innovations in mine digitalization since 2023.

📌 About Mine-Tech Festa

The Mine-Tech Festa is Korea’s premier innovation contest in the mining sector, featuring top university-industry teams across fields such as exploration, resource processing, mine safety, sustainability, and digital mining.

📰 New Publications

📰 New Publication Alert (2025.10.18)
 

Comparative performance of monofacial and bifacial PV–wind–flywheel systems for hydrogen refueling stations

Prof. Yosoon Choi and Dr. Shubhashish Bhakta have published a new study in Energy Conversion and Management (Elsevier, 2026), titled:
“Comparative performance of monofacial and bifacial PV–wind–flywheel systems for hydrogen refueling stations.”

💡 What’s the innovation?
This study introduces a novel hybrid renewable energy system integrating photovoltaic (PV), wind, and flywheel (FW) technologies to power hydrogen refueling stations (HRSs).
The research uniquely compares monofacial and bifacial PV configurations under real Korean climate and grid conditions, assessing their impact on hydrogen cost, energy efficiency, and CO₂ mitigation.

The study:

• Develops an integrated SAM–HOMER Pro simulation framework for dynamic modeling of bifacial PV–wind–flywheel hybrid systems.
  
  

• Quantifies how bifacial PV modules enhance energy yield and reduce system-level hydrogen production cost.
  
  

• Analyzes grid-tied and standalone HRS configurations, evaluating techno-economic performance and environmental benefits.
  
  

🌍 Why it matters
Hydrogen refueling stations are vital to achieving carbon-free mobility, yet they face challenges from renewable intermittency and energy cost fluctuations.
By integrating bifacial PV and flywheel storage, this research:

• Reduces the Levelized Hydrogen Cost (LHC) to 3.62 $/kg—1.9% lower than the monofacial system.
  
  

• Lowers electricity cost (LEC) by 2.7% and improves overall system efficiency by nearly 5%.
  
  

• Demonstrates an annual CO₂ reduction of 342,495 kg, highlighting strong environmental potential for renewable-based HRS deployment.
  
  

🧪 How was it validated?
The team simulated both monofacial and bifacial systems using real meteorological and tariff data from Korea, considering hourly variations in irradiance and wind speed.
Results showed that:

• The bifacial PV–wind–flywheel configuration generated 4.87% more PV energy and maintained >94% renewable fraction.
  
  

• Hydrogen output reached 22,278 kg/year, fully meeting station demand.
  
  

• Flywheel storage improved short-term stability and grid exchange management, enabling continuous hydrogen production.
  
  

🔗 Publication Details
Choi, Y.*, Bhakta, S. (2026). Comparative performance of monofacial and bifacial PV–wind–flywheel systems for hydrogen refueling stations.
Energy Conversion and Management, 348, 120634.
https://doi.org/10.1016/j.enconman.2025.120634 (SCIE)

📰 New Publication Alert (2025.09.25)

Performance prediction of wind energy-driven green hydrogen production: A case study based on optimized machine learning models for abandoned mine sites.

Prof. Yosoon Choi and his team have published a new study in International Journal of Hydrogen Energy (Elsevier, 2025), titled:
“Performance prediction of wind energy-driven green hydrogen production: A case study based on optimized machine learning models for abandoned mine sites.”

💡 What’s the innovation?

This research pioneers a machine learning–based framework for predicting the performance of green hydrogen production systems powered by wind energy. The study:

• Employs multiple machine learning models—including Support Vector Regression (SVR), Random Forest (RF), and Gradient Boosting Machines (GBM)—to forecast hydrogen production efficiency.

• Uses real wind speed data combined with electrolysis system performance metrics to build robust prediction models.

• Identifies key input variables (wind characteristics, system design parameters, and operational conditions) that most strongly influence hydrogen yield.

🌍 Why it matters

Wind-to-hydrogen systems are central to a sustainable energy transition, but their performance is highly variable due to wind intermittency and system nonlinearities. By applying advanced machine learning techniques, this study:

• Enhances the accuracy of hydrogen production forecasts.

• Enables optimized planning and operation of renewable-driven hydrogen plants.

• Contributes to improving the economic feasibility and scalability of green hydrogen as a cornerstone of carbon-neutral energy systems.

🧪 How was it validated?

The framework was tested using case study data from real-world wind energy systems coupled with electrolysis units. Results showed that:

• Gradient Boosting achieved the highest prediction accuracy among the tested models.

• Feature importance analysis revealed wind speed variability and system load as dominant factors for hydrogen output.

• The machine learning–driven predictions significantly outperformed conventional linear regression approaches.

🔗 Publication Details

Choi, Y.*, Bhakta, S. (2025) Performance prediction of wind energy-driven green hydrogen production: A case study based on optimized machine learning models for abandoned mine sites. International Journal of Hydrogen Energy 179, 151646. https://doi.org/10.1016/j.ijhydene.2025.151646 (SCIE)

📰 New Publication Alert (2025.08.13)

Profit-driven Scheduling of CO₂ Storage and Reuse in Salt Caverns for Solar-to-X Applications

Prof. Yosoon Choi and his team have published a novel study in Solar Energy (Elsevier, 2025), titled:
“Profit-driven scheduling of CO₂ storage and reuse in salt caverns for Solar-to-X applications: A two-stage optimization approach.”

💡 What’s the innovation?

This research proposes a profit-maximizing scheduling framework for the cyclic storage and reuse of CO₂ in underground salt caverns, specifically tailored for solar-to-X systems. The framework:

• Integrates CO₂ storage dynamics with Power-to-Gas (PtG) and methanation processes.

• Optimizes operational decisions considering electricity market prices, CO₂ availability, and cavern capacity.

• Balances economic returns with energy system stability in renewable-rich grids.

🌍 Why it matters

The intermittency of solar power creates challenges for continuous synthetic methane production. By strategically storing captured CO₂ during low-demand or low-price periods, and reusing it when solar generation is high, this approach:

• Enhances the economic feasibility of Solar-to-X pathways.

• Supports long-term CO₂ utilization and sequestration strategies.

• Contributes to carbon neutrality goals through integrated energy–carbon management.

🧪 How was it validated?

The proposed scheduling model was tested using real-world data from electricity markets and CO₂ supply scenarios. Key findings include:

• Profit improvements compared to conventional operation strategies.

• Increased utilization of renewable power for CO₂-based synthetic fuel production.

• Operational flexibility enabling both grid services and low-carbon fuel output.

🔗 Publication Details

Khavari, F., Choi, Y.* (2025) Profit-driven scheduling of CO2 storage and reuse in salt caverns for Solar-to-X applications: A two-stage optimization approach. Solar Energy 300, 113855. https://doi.org/10.1016/j.solener.2025.113855

📰 New Publication Alert (2025.07.14)

Database-Driven Solar Parking Optimization for SEVs
Prof. Yosoon Choi and his team have published a cutting-edge paper in Solar Energy Materials & Solar Cells (Elsevier, 2025), titled: 
“Design and implementation of a solar access database for optimizing parking of solar electric vehicles.”

🚗 What’s the innovation?
This study introduces a Solar Access database (SA-DB) that:

• Quantifies real-time and seasonal solar access across 1,388 parking bays,

• Eliminates the need for on-site 360° imaging or specialized devices,

• Provides optimal SEV parking recommendations through a mobile app,

• Enables time- and season-specific parking decisions to maximize solar energy gain.

🌞 Why it matters
SEVs spend most of their time parked, making solar access at the parking spot a crucial factor for energy efficiency. This work:

• Bridges data-driven GIS modeling with solar PV integration,

• Enhances smart mobility by informing SEV users where and when to park for maximum charge,

• Assists general drivers in selecting shaded spots to reduce vehicle cabin temperatures in summer.

🧪 How was it validated?
The SA-DB was constructed using fisheye sky imagery and QGIS, covering all parking lots at the Daeyeon Campus of Pukyong National University. It was further integrated into a mobile app that slashes solar analysis time from 125s to just 5s, enabling real-time decision-making without additional hardware.

🔗 Publication Details
Hong, J., & Choi, Y. (2025). Design and implementation of a solar access database for optimizing parking of solar electric vehicles. Solar Energy Materials and Solar Cells, 293, 113841. https://doi.org/10.1016/j.solmat.2025.113841

📰 New Publication Alert (2025.06.28)

Advanced Solar Parking Guidance Using VR and Solar Forecasting

Prof. Yosoon Choi and his team have published an innovative paper in Applied Energy (Elsevier, 2025), titled:
“Advanced parking assistance system for solar electric vehicles using 360° virtual reality imaging and real-time solar radiation forecasting.”

🚗 What’s the innovation?
The team developed an AI-powered smart parking assistant for solar electric vehicles (SEVs) that:

• Visualizes real-time solar conditions via 360° virtual reality imaging.

• Predicts solar radiation levels using advanced forecasting models.

• Guides drivers to optimal parking spots to maximize solar charging.

🌞 Why it matters
This system bridges mobility and renewable energy, solving a key challenge for SEVs: where to park for the best solar gain. The proposed framework integrates:

• Real-time weather data,

• Solar simulation based on shadow detection, and

• A user-friendly 3D interface for intuitive decision-making.

🧪 How was it validated?
The system was tested in an urban outdoor parking lot using UAV aerial imagery and virtual simulation tools. The study demonstrated how real-time solar optimization could significantly improve SEV energy self-sufficiency.

🔗 Publication Details
Hong, J., & Choi, Y. (2025). Advanced parking assistance system for solar electric vehicles using 360° virtual reality imaging and real-time solar radiation forecasting. Applied Energy, 355, 123064. https://doi.org/10.1016/j.apenergy.2024.123064

📰 New Publication Alert (2025.06.27)

A Novel KPI Framework for Haulage System Evaluation in Open-Pit Mining

We are pleased to announce the latest publication by Prof. Yosoon Choi and collaborators in the Ain Shams Engineering Journal (Elsevier, 2025). The study, titled "A novel integrated key performance indicator for evaluating open-pit mine haulage systems: application of GMG standards", proposes a groundbreaking approach to performance evaluation in open-pit mining.

🔍 What’s it about?
The research introduces an integrated KPI framework that combines nine individual metrics—based on standards by the Global Mining Guidelines Group (GMG)—into a single, comprehensive performance indicator. This methodology enhances decision-making by offering a more intuitive and actionable assessment of system performance.

📍 Where was it applied?
The framework was validated using real haulage operation data from Sungshin Cement’s limestone mine in South Korea. The study analyzed truck and loader performance across availability, utilization, and effectiveness, revealing key inefficiencies and offering practical optimization strategies.

💡 Why it matters
This integrated KPI simplifies complex operational data into a unified metric, enabling faster issue diagnosis, better equipment management, and smarter planning. It is a valuable tool for advancing sustainability and efficiency in the mining sector.

📘 Citation
Park, S., Jung, D., & Choi, Y. (2025). A novel integrated key performance indicator for evaluating open-pit mine haulage systems: application of GMG standards. Ain Shams Engineering Journal, 16, 103589. https://doi.org/10.1016/j.asej.2025.103589

© 2026 SMART-X Lab.