Accepted Abstracts
SIP Conferences proudly recognize the high-quality research contributions submitted by authors worldwide. Each version of the conference showcases diverse and impactful studies advancing sustainability and innovation.
SIP2025 – Version 1
21–22 March 2025
The first edition of SIP2025 received an excellent range of research abstracts covering topics such as sustainable energy systems, environmental protection, and green technology innovation.
These abstracts reflected the global engagement and interdisciplinary collaboration encouraged by the SIP platform.
SIP2025 – Version 2
15–16 November 2025
The second edition further expanded international participation, featuring accepted abstracts focused on renewable energy integration, smart cities, and climate adaptation strategies.
It marked a milestone in the SIP series by strengthening collaboration between academia, industry, and policymakers.
SIP2026 Accepted Abstracts/Full Papers
Updated on March 25, 2026
Data Balance and Fairness in Mobility: The Impact of Gender Representation on AI-Based Models for Sustainable Mobility
Author: Katja Andrea Rösler
Affiliation: University of Applied Sciences Ruhr West, Mülheim an der Ruhr, Germany
Abstract: This study examines how gender representation within mobility datasets affects the fairness, reliability, and decision-support outcomes of artificial intelligence models used in sustainable transport planning. As data-driven approaches become more common in mobility systems, imbalances in underlying datasets can influence accessibility assessments, safety outputs, and optimization processes, creating unintended disparities between user groups. The research aims to evaluate how different gender distributions shape model results and influence key mobility indicators related to efficiency, safety, and service quality. The problem addressed centers on the underrepresentation of women in mobility and safety datasets and how this imbalance may affect model performance. The study uses open mobility data, sectoral statistics, and AI-supported scenario modeling to compare balanced and empirically observed gender distributions. Analytical methods include statistical evaluation, fairness assessment, and scenario-based transport simulations. Findings indicate that balanced datasets produce more equitable and safety-oriented recommendations without reducing system performance. The study contributes a structured approach for integrating gender representation into AI mobility modeling and demonstrates how demographic balance influences algorithmic outcomes. Tools used include statistical modeling frameworks, fairness evaluation metrics, and AI-based mobility analysis environments. Overall, the work highlights the importance of equitable data representation for responsible AI development and for supporting sustainable, inclusive mobility planning.
Keywords: AI Mobility Modeling, Gender Representation, Algorithmic Fairness, Sustainable Mobility, Data-Driven Analysis
An Integrated AI-Enhanced Post-Quantum Security Framework for 6G Edge IoT Ecosystems
Author: Hani Al-Balasmeh
Affiliation: University of Technology Bahrain, Bahrain
Abstract: This study proposes an integrated security architecture for sixth-generation (6G) edge-enabled Internet of Things (IoT) ecosystems to address emerging vulnerabilities associated with quantum computing and ultra-dense device environments. Conventional public-key cryptographic systems such as RSA and elliptic-curve cryptography are expected to become vulnerable in the presence of large-scale quantum computing capabilities. At the same time, future IoT systems require security mechanisms that maintain low latency, scalability, and energy efficiency. The proposed framework integrates lattice-based post-quantum cryptography, AI-based anomaly detection, and blockchain-supported federated learning to provide multi-layered security protection across distributed edge networks. Experimental evaluation using emulated IoT traffic with up to 100,000 connected devices demonstrates that lattice-based cryptographic schemes can reduce encryption latency by more than 50% compared with RSA and ECC while maintaining approximately 24,000 secure sessions per second. Additionally, a hybrid CNN–LSTM intrusion detection model achieved 96.7% classification accuracy with a 3.1% false-positive rate, outperforming traditional machine learning approaches. The federated learning layer supported by blockchain infrastructure enables decentralized and privacy-preserving model training while maintaining performance comparable to centralized learning. Overall, the proposed architecture provides a scalable and resilient framework for securing future 6G edge-IoT infrastructures against emerging post-quantum threats.
Keywords: Post-Quantum Cryptography, 6G Networks, Edge IoT Security, AI-Based Intrusion Detection, Federated Learning
Digital Transformation of Nepalese Cooperatives Through Akash DMS
Author: Tanka Prasad Adhikari
Affiliation: Akash Digital Pvt. Ltd., Nepal
Abstract: This study evaluates the digital transformation of Nepalese cooperatives through the implementation of the Akash Digital Management System (Akash DMS). The research addresses long-standing operational challenges including document mismanagement, manual loan processing, compliance gaps, and inefficiencies caused by paper-based workflows. The study investigates whether cloud-based automation, digital documentation, and workflow optimization can enhance transparency, governance, and operational performance across the sector. Using a mixed-methods approach involving workflow analysis, interviews with cooperative staff, and extraction of quantitative performance metrics from more than 500 cooperatives, the study assesses improvements across key operational indicators. The findings show substantial gains: document retrieval time decreased from several hours to seconds, loan-processing duration was reduced by 50–60%, manual error rates dropped by over 80%, and compliance readiness increased significantly. The Akash DMS framework integrates features such as Digital KYM, automated loan workflows, task management, expiry alerts, and cloud-based archives. These capabilities collectively reduce workload by up to 70% and improve accuracy by more than 80%, demonstrating strong alignment with sustainability and digital-governance objectives. The novelty of the study lies in its sector-scale empirical evaluation using real operational data from a diverse cooperative network. The results indicate that Akash DMS represents a scalable digital-governance model for emerging economies, supporting efficiency, accountability, and long-term institutional sustainability. Future work includes integrating AI-driven risk assessment, OCR-based document extraction, and blockchain-enabled secure archiving.
Keywords: Digital transformation, Cooperatives, Akash DMS, Automation, Document management, Nepal
Enhancing Healthcare Interoperability Using AI and Blockchain Integration
Author: Hani Al-Balasmeh
Affiliation: University of Technology Bahrain, Bahrain
Abstract: This study proposes a layered framework for secure and semantically reliable healthcare interoperability across heterogeneous clinical systems by integrating artificial intelligence (AI) and permissioned blockchain technologies. Existing interoperability solutions often face limitations related to semantic inconsistencies, fragmented governance structures, and excessive data exposure during cross-institutional information exchange. To address these challenges, the proposed architecture combines HL7 FHIR-based interoperability middleware for standardized data normalization, AI-driven semantic intelligence for clinical concept alignment across heterogeneous medical terminologies, and blockchain-based governance mechanisms for decentralized access control, consent management, and data provenance tracking. Experimental evaluation using simulated multi-institutional healthcare datasets demonstrates semantic alignment accuracy between 96% and 98%, outperforming traditional terminology mapping approaches. Privacy analysis indicates that the governance layer reduces unnecessary data exposure by more than 50%, while policy enforcement tests confirm the absence of unauthorized access events. Performance analysis shows that the system maintains end-to-end interoperability latency below 8 ms and transaction throughput exceeding 2000 transactions per second under increasing interoperability workloads. By integrating semantic confidence validation with decentralized governance, the proposed framework enables healthcare systems to exchange clinical data securely while preserving semantic integrity and patient privacy. The results suggest that AI–blockchain integration offers a scalable solution for next-generation healthcare data interoperability and trusted clinical data sharing.
Keywords: Healthcare interoperability; Artificial intelligence in healthcare; Blockchain governance; HL7 FHIR; Semantic data integration
Social Science and Humanities for Global Sustainability: Human Insight for a Sustainable Future
Author: Rajveer Kaur
Affiliation: Chandigarh University, India
Abstract: This study explores how social science and humanities perspectives contribute to global sustainability by examining the human factors that shape fairness, resilience, trust, and inclusive development. While technology and economic tools are essential, the research argues that long-term sustainability also depends on understanding how people live, interact, govern, and make collective decisions. Persistent global inequalities, such as the 692 million individuals living with extremely limited resources, highlight the continued need for social insight alongside technical solutions. Drawing on international assessments covering 222 countries, the study examines how variations in social cohesion, governance quality, equity, and leadership affect progress toward global development goals. Recent global reports indicate that only about 35% of assessed targets are on track, while nearly 18% have regressed since 2015, suggesting that social systems influence the success of sustainable development initiatives. Through conceptual analysis supported by global indices, the study investigates why some societies advance while others lag despite similar economic or technological conditions. The findings suggest that strengthening trust, unity, fairness, and collective resilience can accelerate progress and reduce long-standing disparities. This work emphasizes that sustainable transformation requires integrating human behavior, cultural norms, ethical frameworks, and governance practices into planning processes. Sustainable change must therefore blend technological innovation with social understanding to ensure that growth remains equitable, inclusive, and resilient.
Keywords: Social Sustainability, Governance and Development, Social Cohesion and Trust, Inclusive and Equitable Development, Global Sustainable Development Goals
Biomass-Derived Nanoarchitectured Silicon-Carbon Composites for Enhanced Supercapacitor Energy Storage
Authors: Adetomilola Victoria Fafure, Daniel Barasa Bem, Stanley Wambugu Kahuthu
Affiliation: Kenyatta University; PASET-RSIF, Kenya
Abstract: This study investigates biomass-derived silicon–carbon composites engineered for improved performance in supercapacitor energy storage systems. The work is motivated by the limitations of conventional carbon electrodes, which provide high power density but restricted energy density, and pure silicon electrodes, which face conductivity challenges and structural instability due to significant volume expansion. The research aims to develop hybrid Si–C materials that combine the strengths of both components while mitigating their individual drawbacks. The problem addressed involves achieving a material architecture capable of supporting fast ion transport, stable cycling, and improved electrochemical behavior. Using a green synthesis route consisting of silica extraction, magnesiothermic reduction, and low-temperature carbonization, the study produced composites with different Si:C ratios. A range of structural, morphological, and textural characterization techniques was employed, including XRD, FTIR, Raman spectroscopy, SEM/EDS, and N₂ sorption analysis. Electrochemical performance was evaluated through CV, GCD, and EIS measurements. The Si–C (8:2) composite demonstrated the most balanced performance, including a BET surface area of 120.50 m²/g, a specific capacitance of 467.57 F/g at 1 A/g, and an energy density of 405.88 Wh kg⁻¹. The findings indicate that hierarchical porosity and interfacial engineering enhance ion transport and reduce resistance, contributing to improved energy-storage characteristics. Overall, the study shows that biomass-derived Si–C composites can help close the performance gap between supercapacitors and batteries and offer a sustainable approach to next-generation energy storage systems.
Keywords: Silicon–Carbon Composites, Supercapacitors, Biomass-Derived Materials, Energy Storage, Porous Nanostructures
Agriculture, Food, and Environmental Sustainability: Building Climate-Smart Agriculture and Secure Food Chains
Author: Parvinder Singh
Affiliation: Amity University Noida, India
Abstract: This study examines the interconnected challenges and opportunities within agriculture, food systems, and environmental sustainability, emphasizing how resource-efficient land-use practices and emerging agricultural innovations can enhance productivity while reducing ecological degradation. The work is motivated by rising global food demand, increasing climate variability, soil degradation, and water scarcity, all of which threaten long-term food security. The study focuses on three strategic pathways: precision agriculture, climate-smart practices, and circular food-system approaches, each evaluated for their potential to improve yields, reduce waste, and support resilient production systems. The research aims to (1) measure the environmental performance of selected sustainable farming practices, (2) evaluate their economic feasibility for both smallholders and commercial producers, and (3) analyze how these interventions contribute to climate adaptation and resource efficiency. The problem addressed is the lack of integrated assessments that quantify both environmental and economic impacts across multiple sustainable agricultural innovations. A mixed-methods design is applied, combining field observations, stakeholder interviews, and environmental impact assessments with quantitative modeling supported by GIS, LCA tools, and simulation platforms such as R, ArcGIS, and Python-based analytical models. Preliminary findings indicate that precision agriculture reduces water use by up to 25% and improves nutrient-use efficiency, while agroforestry and other climate-smart techniques significantly enhance carbon sequestration. Circular strategies, such as food-waste valorization—further reduce emissions within the supply chain. The novelty lies in the study’s holistic evaluation framework, which simultaneously examines productivity, environmental outcomes, economic viability, and social acceptability. This integrated perspective provides a more comprehensive understanding than earlier studies that typically consider these dimensions in isolation. The results suggest that scaling climate-smart and resource-efficient innovations, supported by enabling policies and farmer training, can strengthen food security while lowering environmental impacts.
Keywords: Sustainable Agriculture, Land-Use Practices, Food Security, Agroforestry, Water Scarcity
Green-tech Complexity, Energy Consumption, and Environmental Quality: A Machine Learning Framework for Sustainable Solutions
Authors: Hashmat Ali1, Imad Ali1, Khan Baz2
Affiliation: Abbottabad University of Science and Technology1, Zhejiang Agriculture and Forestry University2; Pakistan, China
Abstract: This study examines the relationship between green technology complexity, energy consumption, and environmental quality using a combined econometric and machine-learning framework. Motivated by rising global energy demand and the urgency of decarbonization, the research evaluates how fossil fuel use, renewable energy adoption, and technological complexity jointly influence environmental outcomes. The analysis uses panel data from the top twenty-five carbon-emitting countries covering the period 1995-2023. First, a random-effects ordinary least squares model is applied to capture country-specific effects and identify baseline relationships among key variables. Second, a deep neural network model is employed to assess nonlinear interactions and predictive performance across the full variable set. The econometric results indicate that fossil fuel consumption and economic growth negatively affect environmental quality, while renewable energy use and green technology complexity contribute positively. The machine-learning model further confirms the importance of interaction effects, showing that the combined influence of green technology complexity and renewable energy consumption improves environmental quality more strongly than either factor alone. The study contributes by integrating traditional panel estimation with deep learning to capture both causal direction and complex nonlinear dynamics. The findings suggest that coordinated investment in renewable energy systems and green technological advancement can support emission reduction goals and enhance environmental quality in high-emission economies.
Keywords: Green Technological Innovation, Complexity Index, Machine Learning, Energy Consumption, Environmental Quality
Fossil Fuel Subsidies and Green Technology Impact on Carbon Emissions: A Study on Achieving Net-Zero Policy
Authors: Khan Baz1, Hashmat Ali2, Imad Ali2
Affiliation: Zhejiang Agriculture and Forestry University1; Abbottabad University of Science and Technology2; China, Pakistan
Abstract: This study analyzes the effects of fossil fuel subsidies, green technology adoption, renewable energy use, and forest carbon stocks on carbon dioxide emissions in high-emission economies. The research is motivated by the policy tension between subsidy-driven fossil fuel dependence and the transition toward low-carbon energy systems. Using panel data from the top twenty per capita carbon-emitting countries over the period 2010–2022, the study applies bootstrapped quantile regression to capture heterogeneous impacts across emission distributions. The results show that renewable energy consumption and forest carbon stocks reduce carbon emissions, while fossil fuel subsidies and green technology adoption are associated with higher emissions. Granger causality analysis identifies unidirectional causality from fossil fuel subsidies, green technology, and forest carbon stocks to carbon emissions, and bidirectional causality between renewable energy use and emissions. The findings suggest that subsidy structures can offset the benefits of technological progress when policy incentives remain misaligned. The study contributes by jointly examining fiscal policy, technological adoption, and natural carbon sinks within a distributional econometric framework. The results indicate that achieving net-zero targets requires phasing out fossil fuel subsidies, strengthening renewable energy deployment, and aligning green technology policies with emission-reduction objectives.
Keywords: Fossil Fuel Subsidies, Energy Policy, Green Technology, Eco-Friendly Technology
Industrial Wastewater Management and Cleaner Production Adoption in Stone-Cutting and Concrete Industries: A PLS-SEM Study from the Wadi Zomer Catchment, Palestine
Authors: Momen N. Alqub, Eldon R. Rene, Abdelhaleem I. Khader, Jaap Evers, Adel S. Yasin
Affiliation: Palestinian Water Authority; Al-Najah National University; IHE Delft – Institute for Water Education, Palestine; The Netherlands
Abstract: This study investigates the adoption of cleaner production strategies (CPS) as an industrial wastewater management approach in stone-cutting, concrete, and brick industries located in the Wadi Zomer catchment, a transboundary and environmentally sensitive region in Palestine. Industrial effluents from stone-cutting activities in the area contain high suspended solids that cause sewer blockages, wastewater treatment plant malfunction, and widespread environmental degradation. The proposed CPS separates sludge and water streams, enabling sludge reuse in brick and concrete manufacturing and recycling treated water for industrial cooling processes. Using Partial Least Squares Structural Equation Modeling (PLS-SEM), the study analyzes survey data collected from 45 stone-cutting and ready-mix concrete and brick factories. The results identify the availability of reuse options for effluents and sludge, along with factory owners’ awareness, as the strongest drivers of willingness to adopt CPS, exceeding the influence of enforced regulation or general sustainability perceptions. Willingness to adopt CPS shows a strong positive effect (β = 0.622) on reducing perceived environmental and economic impacts of industrial wastewater. The findings demonstrate that CPS adoption can significantly reduce pollutant loads to sewer networks and wastewater treatment facilities, lower transboundary wastewater-related costs, and improve resource efficiency. The study contributes empirical evidence from a conflict-affected region, highlighting cleaner production as a practical pathway toward sustainable industrial wastewater management under institutional and geopolitical constraints.
Keywords: Industrial Wastewater, Stone-Cutting Effluents, Cleaner Production Strategy, Resource Recovery, PLS-SEM, Transboundary Wastewater Management
Accelerating the 17 SDGs Actualization in Africa’s Emerging Economies: The Role of Innovation, Collaboration, and Policy through ESG Leadership
Authors: Lukman Abayomi Jimoh Rahim; Saheed Adesunkanmi Oyede
Affiliation: University of Jos, Nigeria
Abstract: This study examines how environmental, social, and governance (ESG) leadership influences progress toward the Sustainable Development Goals (SDGs) in Africa’s emerging economies, with specific attention to SDGs 9, 12, 13, 16, and 17. Despite increasing ESG adoption by governments and firms, sustainability performance in Sub-Saharan Africa remains limited, with recent assessments indicating that fewer than 15% of SDG targets are on track. The research addresses the gap between ESG policy adoption and measurable development outcomes by analyzing the mediating roles of innovation capability and cross-sector collaboration, and the moderating effect of policy coherence. A mixed-methods triangulation design is employed, combining Partial Least Squares Structural Equation Modeling (PLS-SEM) with semi-structured interviews. Quantitative analysis indicates that stronger ESG leadership is associated with improved SDG performance, while innovation and collaboration partially mediate this relationship. Policy coherence further strengthens outcomes, with integrated governance systems achieving 20–30% higher SDG indicator performance compared to fragmented policy approaches. The study contributes an Africa-specific ESG leadership framework that positions ESG practices not only as disclosure mechanisms but as operational drivers of sustainable development. The findings suggest that aligning ESG initiatives with innovation systems, collaborative partnerships, and coherent policy frameworks can accelerate SDG progress in emerging African economies.
Keywords: ESG Leadership, Innovation, Collaboration, Policy Coherence, Sustainable Development Goals
Supply Chain Resilience in the Era of Disruption: An Empirical Review of Risk Management, Digital Transformation, and Sustainability (2019–2025)
Authors: Shafiqul Islam; Zaharuzaman Bin Jamaluddin
Affiliation: Universiti Selangor (UNISEL), Malaysia
Abstract: This study presents an empirical integrative review of supply chain resilience research published between 2019 and 2025, with specific focus on risk management, digital transformation, sustainability, and network design under conditions of heightened disruption. The review synthesizes evidence from 60 peer-reviewed empirical studies employing structural equation modeling, panel econometrics, simulation modeling, network analysis, and qualitative case studies. The analysis demonstrates that supply chain resilience is a dynamic, multi-dimensional capability system rather than a static operational attribute. Formalized supply chain risk management, digital visibility and analytics, operational flexibility, relational governance, organizational learning, and sustainability-oriented practices emerge as mutually reinforcing enablers of resilience. The findings further show that digital transformation and sustainability initiatives contribute to resilience only when embedded into organizational routines and aligned across supply chain networks. Recent empirical developments extend resilience thinking toward the concept of supply chain viability, emphasizing long-term survivability through coordinated robustness, adaptability, and recovery. Despite substantial progress, the literature remains dominated by cross-sectional and firm-centric designs, with limited attention to capability orchestration across supply chain tiers. The study identifies clear directions for future research, including longitudinal and multi-tier empirical designs, refined resilience measurement, and integrated assessment of financial, environmental, and social performance outcomes.
Keywords: Supply Chain Resilience, Risk Management, Digital Transformation, Sustainability, Supply Chain Viability
Tourism as a Living Intelligence System: An AI-Native Framework for Sustainable Innovation and Governance
Author: Harilal Bhaskar
Affiliation: iSTEM, India
Abstract: This study proposes a conceptual reframing of tourism as a Living Intelligence System, defined as a continuously sensing, learning, and adaptive socio-technical system supported by artificial intelligence. Tourism governance is currently characterized by fragmented institutional arrangements, siloed data practices, and retrospective sustainability indicators, which limit the sector’s ability to respond to climate risk, overtourism, and resource constraints. Using a design-science and systems-theory methodology, the study introduces the Tourism Intelligence Grid (TIG), an AI-native and policy-aligned framework integrating five interdependent domains: policy and governance, knowledge and research, digital intelligence, tourism operations, and innovation partnerships. The framework positions artificial intelligence not as a standalone optimization tool but as a systemic mediator that enables real-time sustainability governance, anticipatory policy design, and continuous feedback between theory and practice. Environmental signals, operational data, and research outputs are embedded into governance processes to support adaptive decision-making and resilience. The contribution advances tourism research beyond smart tourism and indicator-based sustainability models toward intelligence-driven system design. The framework offers practical implications for policymakers, destination managers, and innovation ecosystems and provides a foundation for future empirical validation, digital twin simulations, and policy experimentation aimed at sustainable and adaptive tourism development.
Keywords: Tourism Governance, Artificial Intelligence, Digital Twins, Sustainability, Systems Theory, Innovation Ecosystems
Teacher AI Tool Adoption and Student 21st-Century Competencies in Rural Pakistan: A Descriptive-Correlational Study
Authors: Muhammad Kashif Majeed; Tunku Badariah Binti Ahmad
Affiliation: International Islamic University, Malaysia
Abstract: This study examines the relationship between teachers’ adoption of Artificial Intelligence (AI) tools and the development of students’ 21st-century competencies in rural secondary schools of Southern Punjab, Pakistan. Using a descriptive-correlational design, data were collected from 100 students (Grades 9–12) through a structured questionnaire measuring perceived teacher AI integration and student competencies (critical thinking, collaboration, communication, creativity, digital literacy, and adaptability). The instrument demonstrated strong internal consistency (Cronbach’s α > 0.80). Results indicate a moderately low level of teacher AI adoption (weighted mean = 2.46–2.51), constrained primarily by infrastructural limitations (85% reporting unreliable internet/electricity), limited device access, insufficient professional training (mean = 2.25), and financial barriers. AI usage was largely confined to mobile-based presentations and adaptive learning tools. Despite limited implementation, students reported positive perceptions of AI-enhanced pedagogy on competency development (weighted mean = 3.09). Pearson’s correlation analysis revealed a strong, statistically significant relationship between teacher AI adoption and student competencies (r = 0.72, p < 0.001). The findings suggest that even modest AI integration can positively influence skill development in resource-constrained contexts, provided that policy interventions support infrastructure, teacher capacity-building, and context-sensitive AI solutions designed for low-bandwidth rural environments.
Keywords: AI in Education, Teacher Technology Adoption, 21st-Century Skills, Rural Digital Divide, Educational Innovation
Digital Solution to Optimize a Unified Surface and Subsurface Simulation Framework for Geothermal Systems
Authors: Deepinder Jot Singh Aulakh1; Muhammad Faisal Iqbal2; Arash Behrang1
Affiliation: SLB (Schlumberger), Canada1; United Kingdom2
Abstract: This study presents a digital solution for optimizing geothermal energy systems through a unified simulation framework that integrates subsurface reservoir dynamics with surface facility operations. The increasing complexity of geothermal systems requires holistic modeling approaches capable of capturing the interdependencies between reservoir behavior and power generation processes. In response, the proposed framework combines reservoir simulation, surface gathering networks, and power plant operations within a single computational environment, enabling comprehensive system-level analysis and improved decision-making. The methodology is based on a bidirectional, time-synchronized data exchange mechanism implemented through an HTTP-based interface controlled by a Python orchestration layer. This integration allows continuous communication between subsurface and surface models at each simulation time step, thereby accurately representing dynamic interactions and evolving system conditions. Such a unified approach overcomes the limitations of traditional isolated modeling techniques and provides a more realistic representation of geothermal system performance. The framework is validated using a geothermal field consisting of eleven production wells and eight injection wells connected to a binary cycle power plant. Two case studies demonstrate its capabilities. The first evaluates the impact of reservoir depletion on long-term power generation, highlighting the necessity of integrated modeling for sustainable energy forecasting. The second investigates an operational optimization scenario involving the shutdown of a production well. Results show that the framework successfully redistributes flow and identifies optimized operational strategies that maintain a stable power output of approximately 1.59 MW while preserving system efficiency and resilience. Overall, the proposed digital integration approach enhances predictive accuracy, supports operational optimization, and improves strategic planning in geothermal energy systems. It contributes to advancing sustainable geothermal development by enabling efficient resource management, reducing operational risks, and supporting global efforts toward low-carbon and climate-resilient energy infrastructure.
Keywords: Digital Integration, Geothermal Energy, Integrated Reservoir–Surface Modeling, Optimization, System Level Optimization
Digital Solution to Optimize a Unified Surface and Subsurface Simulation Framework for Geothermal Systems
Authors: Deepinder Jot Singh Aulakh1; Muhammad Faisal Iqbal2; Arash Behrang1
Affiliation: SLB (Schlumberger), Canada1; United Kingdom2
Abstract: This study presents a digital solution for optimizing geothermal energy systems through a unified simulation framework that integrates subsurface reservoir dynamics with surface facility operations. The increasing complexity of geothermal systems requires holistic modeling approaches capable of capturing the interdependencies between reservoir behavior and power generation processes. In response, the proposed framework combines reservoir simulation, surface gathering networks, and power plant operations within a single computational environment, enabling comprehensive system-level analysis and improved decision-making. The methodology is based on a bidirectional, time-synchronized data exchange mechanism implemented through an HTTP-based interface controlled by a Python orchestration layer. This integration allows continuous communication between subsurface and surface models at each simulation time step, thereby accurately representing dynamic interactions and evolving system conditions. Such a unified approach overcomes the limitations of traditional isolated modeling techniques and provides a more realistic representation of geothermal system performance. The framework is validated using a geothermal field consisting of eleven production wells and eight injection wells connected to a binary cycle power plant. Two case studies demonstrate its capabilities. The first evaluates the impact of reservoir depletion on long-term power generation, highlighting the necessity of integrated modeling for sustainable energy forecasting. The second investigates an operational optimization scenario involving the shutdown of a production well. Results show that the framework successfully redistributes flow and identifies optimized operational strategies that maintain a stable power output of approximately 1.59 MW while preserving system efficiency and resilience. Overall, the proposed digital integration approach enhances predictive accuracy, supports operational optimization, and improves strategic planning in geothermal energy systems. It contributes to advancing sustainable geothermal development by enabling efficient resource management, reducing operational risks, and supporting global efforts toward low-carbon and climate-resilient energy infrastructure.
Keywords: Digital Integration, Geothermal Energy, Integrated Reservoir–Surface Modeling, Optimization, System Level Optimization
Simulation-Based Decision Support Framework for Cargo Bicycle and Microhub Integration in Urban Logistics
Authors: Vladyslav Shramenko1; Natalya Shramenko2
Affiliation: Karlsruhe University of Applied Sciences, Germany1; Baden-Württemberg Institute of Sustainable Mobility, Germany2
Abstract: This study proposes a simulation-based decision support framework for integrating cargo bicycles and microhubs into urban logistics systems, focusing on multi-criteria efficiency evaluation including cost, delivery time, and emissions. The framework is built on a discrete-event simulation model that captures the complete small-parcel delivery cycle under stochastic demand conditions and operational constraints. The model incorporates multiple transport modes, including cargo bicycles and conventional delivery vehicles, and allows systematic variation of key parameters such as demand intensity, fleet composition, logistics configurations, and policy interventions. The simulation generates integrated performance indicators and probabilistic distributions of key outputs, enabling risk-aware and data-driven decision-making. The proposed framework provides a flexible and scalable tool for evaluating sustainable urban logistics strategies and supports the transition toward low-emission and efficient last-mile delivery systems.
Keywords: Urban Logistics, Cargo Bicycle Systems, Microhub Integration, Discrete-Event Simulation, Sustainable Transport
Intelligent Thermal Runaway Detection and Prevention System for Electric Vehicle Batteries using Edge AI
Authors: Vishnu C1; D. Beaulah Princiba1
Affiliation: St. Joseph’s College of Engineering, India1
Abstract: This study presents an intelligent thermal runaway detection and prevention system for electric vehicle batteries using Edge Artificial Intelligence (AI). Thermal runaway poses a critical safety risk in electric vehicles, potentially leading to fire or explosion under uncontrolled temperature conditions. The proposed system integrates multiple sensors, including DHT11 for temperature, ACS770 for current measurement, voltage sensors, and MQ-2 gas sensors, to continuously monitor key battery parameters. Data processing is performed locally using an Arduino-integrated ESP8266 module, enabling real-time analysis and wireless communication with Internet of Things (IoT) platforms such as ThingSpeak. Lightweight machine learning algorithms deployed at the edge facilitate real-time anomaly detection and predictive analytics with low latency. Upon detection of abnormal operating conditions, the system activates preventive mechanisms such as cooling systems or emergency shutdown procedures. The proposed approach enhances battery safety, operational reliability, and lifespan by enabling early fault detection, autonomous response, and efficient energy management within an Edge AI-based framework.
Keywords: Electric Vehicles, Battery Safety, Thermal Runaway, Real-Time Monitoring, Predictive Maintenance, Edge Computing, Anomaly Detection
AI Power Sign Language Interpreter for Inclusive Communication
Authors: R. Jenitha1; M. Anusuya1; M. Muthupriya1; K. Nivitha1
Affiliation: PSR Engineering College, India1
Abstract: This study presents an Artificial Intelligence (AI)-based sign language interpretation system designed to enhance communication for hearing- and speech-impaired individuals. The system integrates computer vision and deep learning techniques, where MediaPipe is used to extract 21 hand landmarks and a Convolutional Neural Network (CNN) is employed for classification of American Sign Language (ASL) alphabets and control gestures. A real-time processing pipeline is implemented using a webcam interface, with outputs delivered through a Flask-based web application. Additional system functionalities include auto-suggestion, delay control, and text-to-speech conversion using gTTS and pyttsx3. The system achieves real-time performance in the range of 25–28 frames per second (FPS) and reports classification accuracy between 95% and 98%, enabling efficient gesture-to-text and speech translation. The proposed solution is applicable in education, healthcare, workplaces, and public service environments. Future work includes extension to dynamic gesture recognition and deployment on mobile platforms.
Keywords: Sign Language, Computer Vision, CNN, MediaPipe, Assistive Technology, Real-Time Recognition, Deep Learning