Aware of the nefarious effects of excessive exploitation of natural resources and the greenhouse gases emissions linked to building sector, the concept of smart buildings emerged, referring to a building that uses clean energy efficiently. This requires intelligent control systems to manage the use of residential energy consuming devices, namely the HVAC (Heating, Ventilation, Air-conditioning) system. This system consumes up to 50% of the total energy used by a building. In this paper, we introduce a RL (Reinforcement Learning) and MPC-LSTM (Model Predictive Control-Long-Short Term Memory) hybrid control system that combines DNNs (Deep Neural Networks), through RL, with LSTM’s long-short memory technique and MPC’s control characteristics. The goal of our model is to maintain thermal comfort of residents while optimizing energy consumption. Consequently, to train and test our model, we generate our own dataset using a building model of a corporate building in Casablanca, Morocco, combined with weather data of the same city. Simulations confirm the robustness of our model as it outperforms basic control methods in terms of thermal comfort and energy consumption especially during summer. Compared to conventional methods, our approach resulted in a 45.4% and 70.9% reduction in energy consumption, in winter and summer, respectively. Our approach also resulted in 26 less comfort violations during winter. On the other hand, during summer, our approach found a compromise between energy consumption and comfort with no more than 2.5 °C above ideal temperature limit.

Access control systems rely increasingly on multimodal biometric and behavioral signals to enhance security and robustness against sophisticated attacks. However, when heterogeneous modalities provide conflicting evidence, such as valid biometric credentials accompanied by abnormal behavioral or acoustic patterns, traditional fusion strategies based on static thresholds or majority voting often fail, leading to false alarms or insecure authorization decisions. This paper addresses this critical limitation by proposing a contextual decision-making fusion framework designed to resolve conflicting multimodal evidence at the decision-making level. The proposed approach models access control as a decision-making problem in a context of uncertainty, where independent agents generate modality-specific evidence from authentication channels based on face, voice, and fingerprints. A centralized fusion mechanism integrates heterogeneous results using adaptive reliability weighting and contextual reasoning to resolve conflicts before operational decisions are made. Rather than treating each modality independently, the framework explicitly considers inconsistencies, uncertainties, and situational context when aggregating evidence. The framework is evaluated using public benchmarks, including VGGFace2, VoxCeleb2, and FVC2004, combined with controlled multimodal scenarios that induce conflicting evidence. Experimental results obtained under controlled contradiction scenarios show that the proposed fusion strategy reduces false alarms and improves decision consistency by approximately 18%. These results are interpreted within the scope of controlled multimodal simulations.

A Smart campus increasingly operates on the basis of data-driven operations, but an increasing demand for energy puts their control over costs and sustainability at risk. This study addresses the challenge of anticipating and managing energy consumption peaks in multi-campus environments by proposing a hybrid framework that combines advanced time-series forecasting models with a large language model (LLM)-driven multi-agent system. Based on the UNICON dataset, LSTM, CNN, GRU, and a combination architecture are trained and compared in terms of MAE and RMSE. The hybrid configuration achieves the greatest forecasting results by returning the minimum loss values. For the identification of critical periods, we employed a strategy based on median thresholding, which offers a categorization into low, normal, and extreme category, allowing the targeting of peak mitigation actions. We also introduce a multi-agent system based on the LLM, including the data aggregator, the forecaster, and the policy advisor, which create actionable policies informed by context. We also compare LLMs (Qwen-2.5, Gemma-2, Phi-4, Mistral, Llama-3.3) in terms of context accuracy, response relevance, semantic similarity, and retrieval/recall accuracy and fidelity, with Llama-3.3 achieving the best overall results. This framework has shown great potential, not only for energy consumption forecasting but also for developing precise policies on how to effectively manage energy consumption peaks.

This paper describes the benchmark dataset for the electric vehicle routing problem with time windows. It is designed to facilitate the large-scale and reproducible evaluation of routing approaches under diverse charging scenarios. It is an extension of the Homberger 1000-customer vehicle-routing benchmark dataset through the incorporation of computationally derived charging-station data. For the 60 base instances included in the dataset, charging-station locations are randomly generated within the customer-coordinate bounds, and two variants are provided, resulting in 120 benchmark problems used in the validation and baseline analyses. A normalized local customer-density score is derived for each station. It is used to determine charging rates and log-normal parameters for prices and waiting times. Two variants are included in the dataset. Variant A maintains the original customer time-window constraints, while Variant B relaxes customer due dates based on the distance from the depot, subject to the depot closing time. The dataset is complemented by instance files, station attributes, parameters, and scripts. It also includes the results of feasibility tests, baseline solver tests, difficulty analyses, and sensitivity tests. These results show that the benchmark includes both easier and harder instance classes under different charging settings. Overall, the dataset is intended to support its use as a reproducible benchmark.