IFIP Open Access Digital Library
Not a member yet
22614 research outputs found
Sort by
Network Simulator-Centric Compositional Testing
No full textPart 2: Full Papers with ArtefactInternational audienceThis article introduces a novel methodology, Network Simulator-centric Compositional Testing (NSCT), to enhance the verification of network protocols with a particular focus on time-varying network properties. NSCT follows a Model-Based Testing (MBT) approach. These approaches usually struggle to test and represent time-varying network properties. NSCT also aims to achieve more accurate and reproducible protocol testing. It is implemented using the Ivy tool and the Shadow network simulator. This enables online debugging of real protocol implementations. A case study on an implementation of QUIC (picoquic) is presented, revealing an error in its compliance with a time-varying specification. This error has subsequently been rectified, highlighting NSCT’s effectiveness in uncovering and addressing real-world protocol implementation issues. The article underscores NSCT’s potential in advancing protocol testing methodologies, offering a notable contribution to the field of network protocol verification
DMFDT: Data Management Framework for Digital Twin
No full textPart 3: Fostering Collaborative and Interoperable Digital Models for Digital Twins: MethodsInternational audienceDigital Twin (DT) provides a digital representation of a real-world entity (process or product) that is continuously synchronized with a specified frequency. In this regard, DT utilizes a set of models that capture the various aspects of the real system to provide a deeper understanding and analysis of its real counterpart. The data within the DT holds paramount significance and serves as the foundation for model updating, refining, interoperability, validity, usability, etc. Accordingly, DT requires rigorous data management throughout its entire life cycle. This paper explores data knowledge areas related to DT (i.e., data governance, architecture, modeling, integration, interoperability, quality, uncertainty,visualization, and security) and also highlights their best practices, and proposes a Data Management Framework for Digital Twin (DMFDT) to facilitate a better understanding of the DT data related requirements and proven practices. Validation and application of the DMFDT is done through the high-level DT architecture and a case study of the proposed framework is also presented by a DT developed to study the mobility system at the University of Bordeaux in France
CM-DIR: A Method to Support the Specification of the User’s Dynamic Behavior in Recommender Systems
No full textInternational audienceThis paper introduces CM-DIR (Conceptual Model -Dynamicity and Interaction in Recommender Systems), a method that we have originally devised to help with the description of dynamic user behavior in Recommender Systems (RS). The identification of user behavior change is paramount to optimize RSs to face new contexts of use and adapt their user interface accordingly. Capturing dynamic user behavior is, however, not an easy task and most RS systems do not take this information into account during the requirements engineering process. The proposed method leverages user and Behaviour-Driven Development (BDD) stories and extends them to describe dynamicity. The method has been evaluated with a panel of experts through semi-structured interviews and the results suggest that the CM-DIR is necessary and comprehensible and that the proposed extension allows flexibility and adaptability while ensuring informative specifications
Mapping Forest Height with Multifrequency SAR, InSAR, and Multispectral Datasets
No full textPart 6: Intelligent Computational SystemsInternational audienceRemote Sensing has been widely used for monitoring forests, namely for the retrieval of structural parameters such as the Forest Height (FH). The reason behind the use of remote sensing is the fact that measuring the FH through field campaigns is expensive and non-scalable. The resort to Airborne Laser Scanning campaigns, despite its high accuracy, have the same limitations. Therefore, Synthetic Aperture Radar (SAR) and Multispectral sensors carried by spaceborne platforms are widely used to address this problem. This paper evaluates the effects of combining a dataset that includes multifrequency backscatter (L and C bands) and multispectral variables, with Interferometric SAR (InSAR) variables (Coherence and Phase) for FH mapping resorting to a locally calibrated regression methodology. To make it more suitable for operational scenarios, only free access data is used, and the calibration sets are small. The scope of this study is the Mediterranean forests, and it has achieved a R2/RMSE ranging from 50.33–72.01%/1.55–2.50m in the validation and 56.22–75.48%/0.77–2.34 m for the operational scenarios. The addition of the InSAR variables leads to an improvement of 0.63% in the R2 and 0.02m in the RMSE
Continual Learning Supporting Human-Robot Collaboration
No full textPart 2: Human-Robot CollaborationInternational audienceContinual learning, a Machine Learning (ML) approach, entails incremental learning and knowledge building over tasks. This study applies continual object detection capability of a human-robot collaborative assembly system facing frequent product rotation. The research is structured into three phases: Initially, hierarchical clustering is employed to establish a parts structure, facilitating the incorporation of new parts and a ML model is pre-trained based on the clustered part. Subsequently, the ML model is used to verify the accuracy of the assembly sequence and enable the robot to accurately select parts during the assembly process. Finally, the actual assembly task by operator and robot is carried out. Any unidentified part is associated in real-time with its category and used to continuously update the dataset for training the model. This facilitates ongoing and continual learning
Comparative Analysis of Time Series and Machine Learning Models for Air Quality Prediction Utilizing IoT Data
No full textPart 2: The 13th Workshop on “Mining Humanistic Data” (MHDW)International audienceAir pollution has been shown to have serious negative effects on people’s health, the environment, and the economy. It is becoming more and more crucial to model, predict, and monitor air quality, particularly in urban areas. Air quality prediction is challenging because of the dynamic nature, instability, and high spatial and temporal variability of particles and pollutants. Internet of things technologies and machine learning offer an efficient way to address these challenges and enables the implementation of effective air quality prediction models. This paper aims to provide a comparative analysis of time series and machine learning methods for air quality prediction based on data collected through IoT sensors. These methods have been evaluated for PM10, PM2.5, and Air Quality Index (AQI) particles. The results indicate that while deep learning models (LSTM) perform better for the air quality index, ARIMA and SVM algorithms best predict the concentrations of the researched air pollutants (PM2.5, PM10)
Net Zero Strategies: Empowering Climate Change Solutions Through Advanced Analytics and Time Series
No full textPart 2: The 13th Workshop on “Mining Humanistic Data” (MHDW)International audienceThis study conducts a comprehensive analysis of CO2 emissions trends from 1990 to 2020 across critical sectors including agriculture, buildings, electricity, industry, oil and gas, and waste. Leveraging a robust dataset of 35,000 observations, we explore emission patterns and their impact on climate change mitigation efforts. Employing advanced time series models—such as the Drift Method, Holt Linear Method, Damped Trend Method, and ARIMA—we forecast emissions and evaluate these models using RMSE, MAE, and MAPE to gauge their predictive accuracy. This research aims to assess the feasibility for various countries to meet the Paris Agreement’s goal of limiting global warming to 1.5 ∘C, aligned with the IPCC AR6 report’s benchmarks for emissions peaking by 2025 and a 43% reduction by 2030. Our analysis reveals critical insights into emission trajectories, underscoring the urgency and practicality of achieving global climate objectives. The findings serve as a crucial guide for policymakers in crafting informed, sustainable development strategies
Digital Technology Enabled Education for Sustainable Development in South Africa: A Case Study of a University of Technology
No full textPart 5: ICT Curriculum and EducationInternational audienceSouth Africa is confronted with various social development challenges, and the higher education sector is well-positioned to address these challenges by developing and implementing appropriate solutions. Furthermore, as technology becomes more prominent in the global development agenda, its adoption in higher education holds great promise for advancing education by overcoming some traditional barriers and transforming pedagogical practices and learning outcomes. This study seeks to understand how the integration digital technologies with teaching and learning activities can facilitate education for sustainable development (ESD) in the context of a university of technology to enhance students’ learning outcomes and competencies needed in today’s job market. The research design adopted is a qualitative case study approach. Semi-structured interviews were conducted with purposively selected participants at a university of technology. The data analysis was guided by the Unified theory of acceptance and use technology (UTAUT) as a theoretical lens to identify patterns and interpretations of digital technology enabled ESD. Data analysis suggests that facilitating conditions such as educator training, adequate policies and strategies, and reliable digital infrastructure are important determinants of the behavioral intentions to use digital technology to facilitate ESD. Furthermore, social factors such as digital divide can impede adequate use of technology in ESD. The study contributes towards understanding factors that enable effective integration of digital technologies to facilitate ESD in higher education
Advanced Time Block Analysis for Manual Assembly Tasks in Manufacturing Through Machine Learning Approaches
No full textPart 3: Computer Vision-based Digital Twin and Digital Services for Dynamic Production and Logistics EnvironmentInternational audienceThe management of assembly tasks within manufacturing, which traditionally relies on using stopwatches and video review, is both labour-intensive and prone to errors. This paper explores an approach utilizing machine learning (ML) and human pose estimation technologies to automate and enhance the classification and management of time blocks for manual assembly tasks in manufacturing environments. We developed and tested ML models capable of classifying manual assembly actions by converting video clips into a time series coordinate dataset via a human pose estimation library. The research highlights the potential of these technologies to significantly reduce the reliance on manual methods by providing a more adaptable, efficient, and scalable system for time data management. Our findings demonstrate accuracy variances across different actions, underscoring the challenges and potential of integrating ML in real-world manufacturing settings. This study provides a promising direction towards revolutionizing traditional practices and enhancing operational efficiencies in manufacturing
A Mobile Air-Purification Device and Digital Twin for Managing Hazardous Gases at Industrial Sites
No full textPart 3: Computer Vision-based Digital Twin and Digital Services for Dynamic Production and Logistics EnvironmentInternational audienceThe presence of hazardous gases in industrial sites where various substances are utilized poses safety concerns to workers. In particular, failure to promptly address hazardous situations can result in severe consequences such as significant casualties. Whereas many industrial sites have adopted air-purification systems to remove hazardous gases, most of those systems are fixed and focus on specific hazardous gases. Therefore, a safety system and response plan that can effectively manage the generation and dispersion of hazardous gases as well as for addressing unexpected situations must be established in industrial sites. This study proposes the development and application of a digital twin for the intelligent control mobile air-purification devices. The mobile air-purification device selects the appropriate removal module based on the type of hazardous gas and predicts the dispersion of gas through the digital twin. Based on the predicted results and derived purification priorities, the mobile air-purification device autonomously drives to the leakage point and removes the leaked hazardous gas. Hence, a digital twin is constructed and utilized to predict the concentration of hazardous gases in the field based on sensor data from the environment. Based on indoor experiments and simulations, we validate the effectiveness of the proposed methodology, which is expected to facilitate crisis response and improve worker safety in industrial settings