Publications
-
Efficient AudioVisual Fusion Architectures for Emotion Recognition
Emotion recognition plays a critical role in the development of adaptive, human-aware intelligent systems. In this work, we propose an end-to-end audiovisual emotion recognition framework that integrates speech signals and facial expressions using lightweight deep learning architectures. To develop the end-to-end architecture, we first benchmark several pretrained convolutional neural networks, employing confidence interval estimation to […]
-
DRAGON: A Dynamic Risk-Aware Graph Optimization Network for Adaptive Building Evacuation Using Graph Convolutional Network and Q-Learning
Efficient evacuation route planning in dynamic environments remains a major challenge, particularly when environmental risks and accessibility conditions change rapidly during emergencies. Traditional shortest-path algorithms, while effective in static graphs, often fail to adapt to evolving conditions, leading to suboptimal or unsafe evacuation guidance. This study aims to develop an adaptive and intelligent evacuation routing […]
-
Coupling Pricing Decisions with Aircraft and Gate Scheduling in Airport Operations
Airlines increasingly emphasize operational flexibility to adapt schedules to fluctuating de mand. Pricing decisions directly influence expected passenger numbers, and when demand is insufficient for a large aircraft, companies may reassign a smaller one, provided capacity, aircraft–gate compatibility, and crew constraints are satisfied. Such changes require gate reassignment, making pricing, aircraft assignment, gate allocation, and […]
-
A NSGA-II Approach for Energy-Efficient Flexible Flow Shop Scheduling with Renewable Energy and Storage Integration
Manufacturing industries face increasing pressure to reduce energy costs while maintai ning production efficiency. Time-of-use electricity pricing structures offer opportunities for cost reduction through strategic production scheduling. Integrating renewable energy sources such as wind and Photovoltaic, along with Energy Storage System further enhances sustainability and economic benefits. This work presents a multi-objective optimization approach using […]
-
Meta-learning with similarity-weighted transfer for manufacturing optimization
The MBAE (Meta-Bayesian Active Explorer) framework addresses the critical challenge of sample efficiency in Industry 4.0 manufacturing optimization, where high-dimensional search spaces and expensive physical evaluations limit experimental budgets. By integrating hierarchical Bayesian meta-learning with similarity-weighted transfer, the model effectively overcomes the « cold-start » problem by leveraging historical task data while preventing negative transfer through a […]
-
Équilibrage des lignes de désassemblage avec des robots collaboratifs dans un environnement aléatoire
Le rôle de l’automatisation dans la fabrication moderne a connu une importance croissante au cours des dernières décennies. Cependant, les fabricants ne peuvent pas automatiser e!ca- cement de nombreuses tâches à ce jour et utilisent par conséquent des travailleurs humains, qui sont soumis à une capacité physique limitée. Ces dernières années, les robots collaboratifs peuvent […]
-
ADAP-GNN: Adaptive property-aware graph neural network for intrusion detection in IoT networks
New and sophisticated attacks are threatening Internet of Things (IoT) networks, compromising the security and trustworthiness of devices. Consequently, Network Intrusion Detection Systems (NIDS) have become critical for protecting these networks, and AI-based NIDS have emerged as a promising solution. A relatively new subfield of deep learning, Graph Neural Networks (GNNs), has further advanced this […]
-
Durability and microstructural evolution of metakaolin-based geopolymers under various coupled temperature and relative humidity conditions
This study investigates the long-term durability of a metakaolin-based geopolymer composite formulated with potassium silicate, fine sand, limestone filler, and short carbon fibers, after 9 months of exposure to three hygrothermal conditions: 20 ◦C – 100 % RH, 90 ◦C – 100 % RH, and 90 ◦C – 0 % RH. A comprehensive evaluation was […]
-
A Literature Review of Public Transport OD Matrix Estimation
Origin–Destination matrices (ODms) are a fundamental input for public transport planning and optimization, as they characterize travel demand across a network. Traditionally estimated from user surveys, ODms are now increasingly inferred from large-scale automatically collected data, such as Automated Fare Collection (AFC), Automated Passenger Counting (APC), and Automated Vehicle Location data (AVL). This review focuses […]
-
An innovative power converter based technique for on-site photovoltaic I-V characterization under natural irradiance
This paper presents a standalone PV curve tracer designed to extract current-voltage (I-V) and power-voltage (P-V) characteristics, as well as the five parameters of a Multiple-Diode Model (MDM) with identical diodes, effectively reducing it to a single-diode model for parameter extraction, under real sunlight conditions. The system consists of a custom-built synchronous boost converter operating […]
-
Hybrid Modeling of a Lithium-Ion Battery Using an Extended Shepherd Model Enhanced with an MLP Neural Network Model
Exact modeling of lithium‐ion batteries is essential for the optimal design and functioning of contemporary energy storage systems. This research introduces a hybrid modeling approach that integrates an extended Shepherd equivalent circuit model (ECM) with a multilayer perceptron (MLP) neural network to improve voltage prediction precision. The ECM parameters are determined utilizing the Red‐Tailed Hawk […]
-
A comparative environmental assessment of an automotive component processed by laser powder bed fusion (LPBF) versus CNC machining, with steel powder reuse impact analysis
The study reported in this paper presents a comparative life cycle assessment (LCA) of a maraging steel automotive hub carrier manufactured through using Laser Powder Bed Fusion (LPBF) metal additive manufacturing (metal AM) versus conventional Computer Numerical Controlled (CNC) machining. The manufacturing processes are modelled with primary data collected from dedicated metal AM unit and […]