Abstract
This study proposes a comprehensive framework integrating eXplainable Artificial Intelligence (XAI) techniques with clustering-based context extraction to enhance energy consumption forecasting in modern office buildings. By leveraging explanation vectors derived from state-of-the-art XAI methods such as SHAP and LIME, our framework identifies latent operational contexts from sensor data aggregated at 15-min intervals. These contexts enable the tailoring of predictive models through feature augmentation, context-specific training, and transfer learning strategies, thereby improving forecasting accuracy compared to conventional approaches. To identify the best-performing models for each context, hyperparameter optimization via grid search is employed across multiple algorithmsincluding Gradient Boosting, Random Forest, and K-Nearest Neighbors. Extensive experiments demonstrate that context-aware models significantly outperform baseline methods, achieving up to a 7% improvement in the coefficient of determination (R-2) and a marked reduction in error metrics. Our findings underscore the importance of integrating XAI with data-driven modeling to enhance predictive performance and model interpretability, which are critical for practical energy management and decision-making in complex building environments.

Abstract
Renewable Energy Communities (RECs) hold potential for enhancing local energy flexibility and supporting a just energy transition. Yet most operate without intelligent coordination, limiting technical performance and raising concerns over fairness and the distribution of benefits. This study examines both the performance and equity dimensions of RECs by combining a critical review of technical, regulatory, and social barriers with simulation-based analysis of a real-world demonstrator developed in the EU-funded OPEVA project. Using real consumption and generation data, we model baseline, rule-based, and intelligent coordination scenarios, as well as expansion cases that integrate additional batteries or EV chargers into underserved households, setting to answer these two questions: Who benefits most from current and future deployments of flexibility technologies? And how can REC systems be expanded to not only aggregate performance gains but also equitable and fair outcomes for all participants? Results show that intelligent control reduces community-level peak demand, ramping, and energy costs while improving renewable self-consumption. However, these benefits are unevenly distributed, concentrated among participants already equipped with flexible assets or with higher demand. Expansion scenarios improve both technical performance and fairness, but inequities persist without deliberate policy intervention. We conclude with open challenges and propose policy and technical measures to ensure that RECs deliver not only efficiency gains but also just and inclusive outcomes.

Abstract
In today's rapidly evolving and competitive business environment, organizations must continuously innovate, leading to the development of new optimization techniques, methods, and tools to support decision-making. In project scheduling management, efficiency and effectiveness are crucial for organizational success, and the tools developed are designed to improve these two critical factors. This paper focuses on applying optimization techniques to project scheduling, with a particular emphasis on metaheuristics, specifically Simulated Annealing. A mathematical model was developed, incorporating the specific requirements of resource constrained project scheduling. A prototype was then implemented based on this model and tested using academic data to assess its effectiveness. The results demonstrated that the prototype could generate effective schedules and exhibited remarkable flexibility, adapting to different types of projects and multi-project environments. This article concludes that using metaheuristics, such as Simulated Annealing, provides a powerful and effective approach to solving complex project scheduling problems, offering significant advantages for organizations operating in dynamic and highly constrained environments.

Abstract
The limited in-person availability of administrative services at higher education institutions can delay the resolution of student queries and reduce satisfaction levels. To address this issue, we developed a conversational agent capable of understanding and responding to student questions in Portuguese using natural language processing and machine learning techniques. To enable non-technical management of the agent's knowledge base, a web-based service was implemented, allowing staff to update content and trigger model retraining. The system was evaluated by comparing multiple learning models, with the best performance achieved using Google's BERT language model combined with the DIET classifier, yielding an F1-score of 0.965. In a real-world deployment involving 256 questions, the chatbot achieved approximately 70% accuracy and received an average user satisfaction rating of 4.20 on a 0-5 scale. These results demonstrate the effectiveness of the proposed solution for improving accessibility and efficiency in academic student services.

Abstract
Identifying influential spreaders in temporal networks is crucial for understanding and controlling the dynamics of spreading. However, existing methods, such as temporal betweenness, closeness, pagerank, degree, and local path-based centrality, face several limitations, including high computational complexity, reliance on shortest paths, convergence issues, inability to capture influence dynamics with insufficient neighboring nodes, and a primary focus on local structural information. This paper presents PathSAGE, a novel method that addresses these problems. It integrates GraphSAGE, a deep learning model, to capture global node information while incorporating temporal local path counts as a key feature. Unlike other global feature-capturing methods, PathSAGE optimises computational complexity. Experimental results on thirteen real-world temporal networks demonstrate that PathSAGE outperforms the state-of-the-art methods in accurately identifying influential spreaders. PathSAGE exhibits a strong correlation with the Temporal Susceptible-Infected-Recovered (TSIR) model and achieves a relative improvement percentage (eta%) ranging from 0.12% to 70.70%. Additionally, PathSAGE attains the lowest average robustness value of 0.17, highlighting its effectiveness in identifying influential spreaders within temporal networks.

Abstract
Artificial Intelligence (AI) integration in supply chain systems is growing, and with it grows its potential impact on inter-organisational collaborative networks. We review existing literature on how different AI archetypes (Reflexive, Anticipatory, Supervisory, Prescriptive) could support Collaborative Supply Chain Management (CSCM) activities, and how they impact information sharing, collaborative decision-making, and trust among supply chain partners at different integration levels. Adopting a sociotechnical perspective, we synthesise existing literature and map the archetypes along four levels of AI integration, varying in scope and decision autonomy. The results are conceptual frameworks demonstrating how AI impacts collaboration dynamics as it evolves from a decision-support tool to an autonomous coordination agent. Findings show differentiated effects along archetypes and integration levels, with implications for CSCM governance, transparency, and resilience. We contribute to the discussion on human-AI collaboration in CSCM and offer a baseline for research on the human-centric values of Industry 5.0.