Smart Emergency introduces an AI-driven structure to adapt to emergency operations through real-time data analysis, IoT and telemedicine. System AI-driven vehicle routing, predictive analysis and dynamic traffic control increase the detection, resource allocation and medical decision-making. Unlike traditional approaches, Smart Emergency dynamically prioritizes emergency interventions, improving response times and patient outcomes. A functional prototype has been developed and tested using controlled test scenarios and applied to the street network of the city of Rabat. This prototype includes a hybrid optimization engine for vehicle dispatch, a dynamic mapping interface, and algorithms for identifying high-risk zones. Although not yet deployed in real-world emergency networks, the system demonstrates the feasibility of the proposed architecture and serves as a foundation for future pilot deployments. This study presents the system architecture, core components, and current implementation status, highlighting the platform’s potential scalability for both routine and large-scale crises. Smart Emergency introduces an AI-driven structure to adapt to emergency operations through real-time data analysis, IoT and telemedicine. System AI-driven vehicle routing, predictive analysis and dynamic traffic control increase the detection, resource allocation and medical decision-making. Unlike traditional approaches, Smart Emergency gives dynamic emergency interventions, improves response time and patient results.

Multi-objective optimization plays a fundamental role in solving complex engineering design problems characterized by conflicting objectives and nonlinear constraints. In this study, a novel hybrid optimization algorithm, named Multi-objective Lemurs Optimizer-Backtracking Search Algorithm (MOLOBSA), is proposed to improve the exploration and exploitation capabilities of existing metaheuristic methods. The proposed approach integrates the global exploration ability of the Lemurs Optimizer (LO) with the efficient mutation and crossover mechanisms of the Backtracking Search Algorithm (BSA) within a multi-objective optimization framework. The effectiveness of the proposed algorithm is evaluated using the CEC2020 multimodal multi-objective benchmark suite, where its performance is assessed using the PSP and IGDX performance indicators. In addition, the proposed method was successfully applied to the multi-objective design optimization of an I-beam structure, where the objectives were to minimize the structural weight and the maximum displacement under mechanical constraints. The obtained Pareto solutions exhibit better diversity and improved trade-off characteristics compared with those produced by the baseline algorithm.

This study introduces an enhanced variant of the Genghis Khan Shark Optimizer (GKSO) algorithm, designed to improve its search performance when solving the Optimal Power Flow (OPF) problem, particularly in power systems incorporating Flexible AC Transmission Systems (FACTS) devices and wind energy sources. The proposed approach, referred to as the Enhanced Genghis Khan Shark Optimizer (EGKSO), addresses the limitations of the original GKSO, which, despite its simplicity and effectiveness, may suffer from premature convergence and entrapment in local optima.To overcome these shortcomings, EGKSO incorporates significant improvements in both the exploration and exploitation phases of the optimization process. Exploration is enhanced through a stochastic mutation mechanism that promotes the discovery of new regions in the search space, while exploitation is refined using an adaptive strategy that intensifies the search around promising solutions. This combined mechanism improves the balance between global exploration and local exploitation, leading to more efficient convergence toward the global optimum.The performance of EGKSO is first evaluated using the CEC’17 benchmark functions and compared with several well-established meta-heuristic algorithms, including Artificial Ecosystem-Based Optimization (AEO), Marine Predators Algorithm (MPA), Manta Ray Foraging Optimization (MRFO), Hunter-Prey Optimization (HPO), and the original GKSO. The proposed algorithm is then validated on a complex and highly constrained OPF problem involving the integration of FACTS devices namely Thyristor Controlled Phase Shifter (TCPS), Thyristor-Controlled Series Compensator (TCSC), and Static VAR Compensator (SVC) as well as wind power generation.The results demonstrate that EGKSO achieves superior performance in terms of convergence speed, solution accuracy, and robustness when addressing complex OPF scenarios. These findings confirm the effectiveness of the proposed algorithm and highlight its potential for solving real-world power system optimization problems.

This paper presents a real-time vision-based framework for emergency-aware adaptive traffic signal control using roof-mounted siren/light-bar detection as a common visual cue for emergency vehicles. Unlike acoustic sensing or full-vehicle classification, the proposed method detects the light-bar itself to improve robustness across heterogeneous emergency fleets. A YOLO26m object detector was trained on a curated dataset of 3,130 manually annotated images containing both activated and non-activated light-bar states, formulated as a single-class object-presence task independent of vehicle type. Across three training runs, the detector achieved mean precision, recall, mAP50, and mAP50–95 of 0.73 ± 0.03, 0.69 ± 0.002, 0.73 ± 0.02, and 0.34 ± 0.01, respectively, with the best run reaching 0.76 precision and 0.75 mAP50. The detector was integrated as a supervisory layer over a pre-optimized Genetic Algorithm–Stochastic Gradient Descent (GA-SGD) traffic controller to activate temporary emergency priority and restore optimized operation after vehicle passage. Simulation of Urban Mobility (SUMO)-based system-level validation was performed on a modeled Souissi roundabout in Rabat, Morocco, comparing Fixed-time, GA-SGD, and the proposed GA-SGD with vision-triggered priority override. In the selected stress scenario, the proposed framework reduced ambulance travel time to 56 s, compared with 70 s for GA-SGD and 64 s for Fixed-time control, while lowering ambulance waiting time from 21 s to 15 s and total network waiting time to 19,887 s. The results support the feasibility of the proposed framework for pre-deployment intelligent traffic management studies.

The Backtracking Search Algorithm (BSA) has emerged as a promising stochastic optimization method. This paper introduces a novel hybrid evolutionary algorithm, termed LOBSA, integrating the strengths of BSA and Lemurs Optimizer (LO). The hybrid approach significantly improves global exploration and convergence speed, validated through rigorous tests on 23 benchmark functions from the CEC 2013 suite, encompassing unimodal, multimodal, and fixed dimension multimodal functions. Compared with state-of-the-art algorithms, LOBSA presents a relative improvement, achieving superior results and outperforming traditional BSA by up to 35% of global performance gain in terms of solution accuracy. Moreover, the applicability and robustness of LOBSA were demonstrated in practical constrained optimization and a fluid–structure interaction problem involving the dynamic analysis and optimization of a submerged boat propeller, demonstrating both computational efficiency and real-world applicability.

Traffic congestion remains a pressing challenge for urban mobility, affecting travel times, safety, and environmental sustainability. This paper presents a practical hybrid optimization framework for traffic signal timing that integrates a YOLO-based vehicle detection module with a SUMO-based traffic simulation and optimization module. The system was developed in collaboration with the National Road Safety Agency (NARSA) in Morocco as part of a larger Intelligent Transportation System initiative. The detection module, built using the YOLO object detection architecture, has been tested on recorded video streams due to current regulatory restrictions on installing live cameras; it is designed to operate identically on live feeds once authorized. The simulation and optimization module allows users to select a modeled intersection within a web application, specify traffic demand and constraints, and run simulations in SUMO to compute optimal traffic signal cycle times using a hybrid Genetic Algorithm (GA) and Stochastic Gradient Descent (SGD) approach. The current study evaluates the optimization framework in simulated conditions for the Bab Zaer intersection in Rabat, Morocco, under varying traffic demands. Results show substantial reductions in total and per-vehicle waiting times compared to baseline configurations. While this work focuses on simulation-based validation, the proposed architecture supports future integration with live detection data and remote programming of traffic controllers, enabling scalable deployment in real-world traffic networks.

Employee transportation is an operational challenge for numerous industrial organizations seeking to enhance their overall productivity. This challenge is classified as an NP-hard problem, as it is an extended variant of the Vehicle Routing Problem (VRP), named Employee Bus Routing Problem (EBRP). In recent years, the EBRP has been extensively studied using various approaches, including heuristics, metaheuristics, and exact algorithms. However, efficiently solving large instances of the bi-objective EBRP remains a significant challenge for researchers. This paper examines a practical application of the EBRP for a large Moroccan industrial group that provides shuttle services for its employees across multiple regions. This company aims to operate a fleet of heterogeneous buses to ensure the comfort of thousands of employees while minimizing operational costs. Therefore, we propose a bi-colony Ant Colony Algorithm (BC-ACO) with two competing colonies: one focused on minimizing total cost and the other on reducing the maximum travel time for passengers. Each colony independently explores potential solutions while iteratively sharing information with one another throughout the optimization process. To evaluate the efficiency of the proposed approach, we conduct a comparative analysis against three other methods: a weighted sum bi-objective Genetic Algorithm (BGA), bi-objective ACO with one colony (BACO) and an exact approach. The results indicate that the BC-ACO yields solutions that provide a better trade-off and within a minimized execution time, making it a viable option for large-scale applications in the real world.

Smart cities increasingly rely on intelligent technologies to improve urban infrastructure, sustainability, and quality of life. Traffic flow prediction is essential for the optimization of the transportation system, reducing congestion and improving mobility. However, real-world traffic data are often noisy, limited in size, and lack sufficient features to capture the flow dynamics and temporal dependencies, making accurate prediction a significant challenge. Previous studies have shown that recurrent neural network (RNN) variants, such as LSTM and GRU, are well-suited for time series forecasting tasks, but their performance is highly sensitive to hyperparameter settings. This study proposes a hybrid approach that integrates GRU with a metaheuristic optimization algorithm to address this challenge. After effective preprocessing steps and a sliding time window are applied to structure the data, particle swarm optimization (PSO) is utilized to optimize the hyperparameters of the GRU. The model’s performance is evaluated using RMSE, MAE, and R2, and compared against several baseline approaches, including LSTM, CNN-LSTM, and a manually configured GRU. According to the experimental findings, the GRU model that was manually adjusted performed the best overall. However, the PSO-GRU model demonstrated competitive results, confirming that metaheuristics offer a promising alternative when manual tuning is not feasible despite the higher computational costs.

In recent times, there has been a growing demand for effective solutions to address the optimal power flow (OPF) problem. This increased attention is driven by the necessity to ensure reliable and optimal grid operations, considering factors such as generation uncertainty and rising demand. The OPF problem is designed with specific objectives to optimize power system variables while adhering to certain constraints. Therefore, this study presents a variant of an innovative optimization approach called the coati optimization algorithm (COA), which is based on swarm intelligence. Traditional algorithms may suffer from poor exploitation and require numerous iterations to achieve the global optimum when dealing with complex problems. To address this, the developed leader coati optimization approach (LCOA), based on the superiority of the feasible solution (SF) mechanism, enhances the exploitation capability of the standard COA and avoids getting trapped in locally optimal solutions. This is achieved by implementing a leader-based mutation-selection approach during the exploitation phase in each generation of COA. To evaluate the effectiveness of the suggested approach, a performance validation process was conducted using the CEC’17 benchmark test suites, demonstrating its superiority over COA and other recent algorithms. Additionally, three test systems adhering to IEEE standards, specifically the 30-bus, 57-bus, and 118-bus systems with fifteen case studies, were analyzed. The simulation results were assessed by comparing the performance and excellence of the proposed method with other well-designed optimization studies reported in the literature. The statistical analysis and simulation results conclusively demonstrate that the proposed LCOA exhibits superior convergence, robustness, efficiency, and high-quality feasible solutions for different OPF problems compared to the original COA and its competitors.

Assuming the stochastic uncertainty in real-life decision making problems, this paper, we study mean-risk model for stochastic Unmanned Aerial Vehicles Surveillance Routing (SUAVsSRP) problem. We use the bi-objective formulation of this problem involving stochastic travel times, the model tries to optimize expected travel times and travel time reliability simultaneously. The risk value is chosen as the reliability measure. In order to deal with these complexities, we propose a simheuristic algorithm with two dominance criteria. It incorporates a monte carlo simulation (MCS), a stochastic dominance criteria, and a pareto metaheuristic. The role of MCS is an effective method to assess and calculate probabilistic objective functions, and theoretically, we focus on a concise overview of the variant concepts of univariate stochastic orders, and combining one of them with mean-risk pareto dominance to filter non-dominated solutions, note that, the choice between these order relations depends on the risk measures studied. The validity of this new simheuristic algorithm is studied empirically and the results demonstrate the ability of the proposed technique to generate a well-distributed set of mean-risk pareto-optimal solutions.

Numerous approaches have been employed to optimize traffic signal systems in response to rapidly increasing traffic volumes. However, the sheer variety of methods can sometimes overwhelm and confuse early career researchers seeking guidance. Traffic signal control relies on various criteria that guide method selection, broadly categorized into fixed specifications identified through empirical data and adjustable parameters that vary by context. Given this complexity, integrating a multicriteria decision-making (MCDM) framework is essential. This study proposes a fuzzy-TOPSIS-based framework -where TOPSIS stands for Technique for Order of Preference by Similarity to Ideal Solution - that integrates both fixed and adjustable evaluation criteria to rank traffic signal control methods based on situational needs. The approach serves as a decision support tool that accounts for uncertainty in expert judgment while ensuring a systematic and transparent method selection process. This framework evaluates various traffic signal control methods, taking into account both fixed and adjustable criteria, to identify the most suitable solution for a given case study. The proposed method not only simplifies the decision-making process, but also provides a structured and reliable tool for researchers and practitioners in the field of intelligent traffic management.

In this paper, an enhanced version of the Non-dominated Sorting Genghis Khan Shark Optimizer (NSGKSO) is proposed to tackle complex multi-objective optimization problems, with a specific focus on the Optimal Power Flow (OPF) problem. The optimization incorporates renewable energy sources such as solar, wind, and small-hydro power, which introduce significant variability and complexity into the system. The proposed NSGKSO is based on the widely used NSGA-II framework but incorporates novel mechanisms to improve convergence and solution diversity. We evaluated the performance of NSGKSO on several OPF test cases and compared its results with other state-of-the-art algorithms. Our experimental results demonstrate that NSGKSO achieves a better balance between convergence speed and diversity preservation, making it well-suited for solving multi-objective OPF problems in the presence of renewable energy sources.

This overview is a recent literature on simulation-based multi-objective evolutionary algorithms (SMOEAs) capable of handling stochastic multiple objective functions. Special attention is given to stochastic multi-objective problems as well as to combinations of multi-objective evolutionary algorithms with simulation techniques. Then we illustrate the principale working of cooperation between Simulation and MOEAs, and discuss their application scope. Finally, it highlights recent important trends and closely related research fields.

The Backtracking Search Algorithm (BSA) stands out as a contemporary stochastic technique that has showcased its prowess in tackling intricate engineering challenges. Thus, a novel hybrid evolutionary algorithm designed to fully leverage the advantages of the BSA and the Lemurs Optimizer (LO). The core objective of this hybrid algorithm is to enhance and discover the global search ability of BSA and provide a robust optimization tool for finding global optima or high-quality solutions for a variety of complex benchmark functions commonly used in the field of optimization. LOBSA's effectiveness has been put to the test with 23 diverse benchmark functions, and its performance has been evaluated through a comparison with highly regarded state-of-the-art competitors. Besides, statistical results indicate that LOBSA presents promising and competitive outcomes, not only for its initial variant but also when compared to the other algorithms.

In response to the rising number of vehicles and corresponding increase in road accidents within urban environments, this study introduces a hybrid approach that combines clustering and classification of road accident severity to aid city planning and enhance public safety in smart cities. To address the imbalance in real-world accident data, which particularly affects the accurate prediction of fatal accidents often underrepresented in datasets, we employed clustering as a foundational structure for data segmentation. We developed a method based on two clusters for severity resampling. Within each cluster, we applied targeted resampling techniques, enhanced by SMOTE and ADASYN, to achieve a balanced representation of severity classes. Our comparative analysis reveals that our approach significantly outperforms models trained on both the original imbalanced dataset and those modified by standard resampling techniques. Specifically, we observed an increase in G-mean from 0.02 and 0.29 to 0.70 and an improvement in AUC-ROC by 35%, and 38% respectively. These enhancements highlight the efficacy of integrating clustering techniques into accident severity classification. This robust and adaptable methodology consistently improves predictions of accident severity in urban environments, demonstrating its effectiveness across various methodological setups.

In most of the existing multi-objective metaheuristics based on decomposition, the reference points and the subspaces are statically defined. In this paper, a new adaptive strategy based on Tchebycheff fractals is proposed. A fractal decomposition of the objective space based on Tchebycheff functions, and adaptive strategies for updating the reference points are performed. The proposed algorithm outperforms popular multi-objective evolutionary algorithms both in terms of the quality of the obtained Pareto fronts (convergence, cardinality, diversity) and the search time.

In this paper we propose an autonomous chaotic optimization algorithm, called Tornado, for high dimensional global optimization problems. The algorithm introduces advanced symmetrization, levelling and fine search strategies for an efficient and effective exploration of the search space and exploitation of the best found solutions. To our knowledge, this is the first accurate and fast autonomous chaotic algorithm solving large scale optimization problems. A panel of various benchmark problems with different properties was used to assess the performance of the proposed chaotic algorithm. The obtained results have shown the scalability of the algorithm in contrast to chaotic optimization algorithms encountered in the literature. Moreover, in comparison with some state-of-the-art metaheuristics (e.g. evolutionary algorithms, swarm intelligence), the computational results revealed that the proposed Tornado algorithm is an effective and efficient optimization algorithm. A panel of various benchmark problems with different properties was used to assess the performance of the proposed chaotic algorithm. The obtained results have shown the scalability of the algorithm in contrast to chaotic optimization algorithms encountered in the literature. Moreover, in comparison with some state-of-the-art metaheuristics (e.g. evolutionary algorithms, swarm intelligence), the computational results revealed that the proposed Tornado algorithm is an effective and efficient optimization algorithm.

Artificial neural networks (NNs) have shown remarkable success in a wide range of machine learning tasks. The activation function is a crucial component of NNs, as it introduces non-linearity and enables the network to learn complex representations. In this paper, we propose a novel activation function based on Hilbert basis, a mathematical concept from algebraic geometry. We formulate the Hilbert basis activation function and investigate its properties. We also compare its performance with popular activation functions such as ReLU and sigmoid through experiments on MNIST dataset under LeNet architecture. Our results show that the Hilbert basis activation function can improve the performance of NNs, achieving competitive accuracy and robustness via probability analysis.