Landslide prediction is a critical task for ensuring public safety and preventing economic loss in regions prone to such natural disasters. Traditional models for landslide prediction often lack accuracy and precision because of the intricate interactions between various factors that lead to landslide events. To tackle this issue, we introduce an innovative hybrid approach for landslide prediction that combines Time Series Analysis (TSA), Genetic Algorithm (GA), and Support Vector Machine (SVM). TSA decomposes landslide displacement data into trend, seasonal, and residual components, improving the clarity of the data. GA optimizes the hyperparameters of SVM, ensuring the most effective application of the SVM. Finally, the SVM is trained on detrended data, producing a model capable of accurately predicting future landslides. Our experimental outcomes manifest that the TSA-GA-SVM model we advanced performs far better than the individual TSA and SVM models when it comes to forecasting landslide displacement. The hybrid model achieved a mean absolute error of 0.15 m compared to 0.42 m for TSA and 0.38 m for SVM alone. Sensitivity analysis revealed that increasing GA population size improved model stability, while higher mutation rates led to more variable predictions. The model showed good generalization ability, performing well across different regions and under various geological and hydrological conditions. This research not only advances the state of the art in landslide prediction but also provides a practical tool for authorities to implement in their disaster prevention and management strategies.

This paper presents a novel hybrid model that integrates predictive and optimization techniques to enhance the scheduling and management of electricity generation in large-scale power systems, with a focus on the variability of photovoltaic (PV) energy. By combining a long short-term memory (LSTM) neural network with an optimization framework, the model forecasts PV power generation over a one-month horizon using historical data, validated against actual production. The optimization component, built on a refined large-scale power system model, incorporates these predictions using a block representation approach to simulate diverse generation technologies, including natural gas, fossil fuel-based thermal units, hydroelectric, PV, nuclear, and wind power plants. This integrated approach addresses the stochastic nature of renewable sources, distinguishing it from prior studies that focus solely on prediction or optimization. The Argentine Interconnection System (SADI) serves as the case study, leveraging over a decade of time-series data to evaluate the model’s performance. Results demonstrate reliable prediction and scheduling capabilities, achieving a low prediction error of approximately 0.01% for key PV sources. Implemented in Python within the Spyder environment, with TensorFlow and Keras for LSTM predictions and PYOMO for optimization, the model offers a practical and effective solution for system operators to optimize resource allocation in renewable-heavy power systems.

This paper presents a literature review for inventory control of deteriorating items since 2018. A classification including 18 classes and 33 subclasses is offered to categorize inventory control models, constraints, and solution methods used in previous studies. Providing standard classes in this field, such as demand, deterioration, shortages, number of warehouses, and time value of money alongside new classes, for example, the type of model costs and supply chain, inventory constraints, number of supply chain levels, time horizon, lead time, considering multi-item models, preservation technology, financial conditions, non-instantaneous deteriorating items, environmental issues, and solution methods made this classification more comprehensive. A brief history and explanation are given to understand each class better, and related articles are grouped in these classes. The research gaps and a crucial aspect that paves the way for future research are presented in each category. A broad view of the future of this topic is provided, and exciting opportunities are highlighted for researchers to contribute to this field and inspire them to explore these potential areas of research. The potential for future research in this subject is vast and promising; this article offers numerous opportunities for researchers to make significant contributions. The results show that the best ways to extend this topic are using variable deterioration rates, costs, and demand functions, considering realistic assumptions, including allowable shortages with partial backlogging, two warehouses, inflation and discounts, preservation technology, uncertain lead time, and environmental issues. Developing cyclic (if possible), multi-item, and production models with financial conditions and various inventory constraints is an excellent way to develop existing models. Finally, solving the proposed models using exact methods to find the global answer is a great effort to contribute to this field.

Safety is a top concern for the civil aviation industry, and the risk of safety accidents will increase due to pilot fatigue. To ensure the safety of civil aviation, this paper proposes a method to solve the crew scheduling problem considering pilot fatigue. In order to reflect individual differences and fatigue levels of pilots, an improved three-stage alertness calculation model is first proposed based on subjective and objective perspectives to represent pilots’ alertness levels and fatigue working duration quantitatively. Then, for the crew scheduling problem considering pilot fatigue, a mixed integer programming model is constructed to simultaneously achieve the optimization objectives of reducing the overall scheduling cost and crew fatigue working duration. Next, since the actual crew scheduling problem is large-scale, a solution algorithm based on a column generation framework is developed to improve the quality and efficiency of solving the large-scale crew scheduling problem. Furthermore, in the case study, we collected actual data from an airline company to validate the effectiveness of our proposed method. Finally, through multiple experimental comparisons and analyses, to balance the two optimization objectives mentioned above, it is more reasonable to handle pilot fatigue working duration with soft constraints. Sensitivity analysis reveals the variation rules of the crew cost and fatigue, providing some valuable managerial insights for the crew scheduling problem considering pilot fatigue.

This paper presents a new integrated framework combining the Joint Replenishment Problem (JRP) with a generalized Vendor Managed Inventory (VMI) system. The model under consideration represents a three-level supply chain consisting of a supplier, manufacturer, and retailer. The model incorporates multiple product types, each produced on a dedicated machine at the manufacturer, subject to setup costs, and major and minor ordering costs. The primary objective of this research is to optimize a set of critical decision variables, including the common order interval, order frequencies for each item, backorder levels at the retailer, and production initiation times at the manufacturer for each product type, under both deterministic and stochastic demand scenarios. This analysis will provide valuable insights for improving joint replenishment operations in manufacturing. The research begins with a deterministic model fit for the particular issue area derived from the canonical JRP. Within a VMI context, the manufacturer, acting as the supply chain leader, utilizes shared information to derive initial feasible solutions. Subsequently, an optimization technique is employed, combining marginal cost-based and cumulative cost-based algorithms, while leveraging embedded discrete Markov chain decomposition method adapting Jacobi stepping method to determine steady-state probabilities. A cost function is then defined for each action state within this framework. The integration of the VMI policy into the JRP model can significantly reduce the whole cost of the supply chain, through balancing between production initiation and backorders under both the deterministic and stochastic models.

The new economic context has brought new challenges to the supply chain and has increased the complexity of its processes. The digitalization; as one of these challenges, is a rapidly evolving paradigm that transforms supply chains by integrating data and communication technologies to optimize operations, enhance sustainability, and improve overall performance. Digital twin technology emerged as one of the most promising digital tools that offer an innovative approach to supply chain management. However, the adoption of digital twins in the supply chain is still in its early stages. Previous research papers presented limited overviews of the applications of digital twin technology in supply chain systems that need to be extended, as it is inevitably a work in progress. In this matter, we conducted a systematic literature review built upon 31 articles to determine the applications of supply chain digital twins (SCDT). This study is divided into three core themes; the first is a comprehensive review of the paradigm of digital supply chain with a focus on digital twin technology and its primary features. The second theme presents an analysis of the 31 papers where we explore the different purposes of SCDTs and their integration. in the third theme by using VOSviewer to conduct a network analysis. We aim; through this paper, to contribute significantly to the supply chain management field by summarizing and analyzing existing research and developments in the applications of digital twins in the different areas of supply chains.

Hybrid laminated composites, integrating High Strength (HS) carbon and glass fibers (E, S) within an epoxy matrix, deliver an optimal compromise between lightweight design, mechanical strength, and cost-effectiveness for applications in industrial, aerospace, automotive, and civil engineering sectors. This study presents a three-dimensional optimization of mechanical performance through a multi-objective genetic algorithm (MOGA) under static loading conditions. The design variables encompass the number of plies (6 to 12), fiber orientation angles (-90°≤ θ ≤90°), and ply materials: HS-Carbon/Epoxy (CF-EP), E-Glass/Epoxy (EG-EP) and S-Glass/Epoxy (SG-EP). A constraint mandating 25% CF-EP placement at the core to maximize stiffness while minimizing stresses. The objectives are to enhance the longitudinal modulus (Ex) and reduce von-Mises stress, while ensuring compliance with the Tsai-Wu failure criterion. An analytical model, implemented in MATLAB, incorporates stiffness matrices, Tsai-Wu failure indices, and von-Mises stress calculations, demonstrating a 30% increase in stiffness and effective mitigation of stress concentrations through centralized CF-EP placement. These findings are corroborated by finite element method (FEM) simulations conducted in ANSYS, which exhibit strong agreement with analytical predictions. This hybrid methodology offers a strong framework for developing high-performance laminated composites, significantly impacting applications requiring structural reliability and efficiency.

Metaheuristic algorithms are easy, flexible and nature-inspired algorithms used to optimize functions. To make metaheuristic algorithms better, multiple algorithms are combined and hybridized. In this context, a hybrid algorithm (HSSAOA) was developed by adapting the exploration phase of the arithmetic optimization algorithm (AOA) to the position update part of the salp swarm algorithm (SSA) of the leader salps/salps. And also, there have also been a few new additions to the SSA. The proposed HSSAOA was tested in three different groups using 22 benchmark functions and compared with 7 well-known algorithms. HSSAOA optimized the best results in a total of 16 benchmark functions in each group. In addition, a statistically significant difference was obtained compared to other algorithms.

This decade has observed an upsurge in the eco-friendly materials because of the development of composites using natural fibers. These composites are made from renewable resources and are gaining popularity for their high performance in engineering applications. Industries are increasingly interested in using materials that are sustainable and resource-efficient. This research proposes a new innovative hybrid composite developed using coir and kenaf fibers, carbon nanotubes acting as a nanofiller, and a matrix made up of epoxy resin, detailing how they are fabricated, tested, and optimized based on different weight percentages. The weight percentages considered for CNT nanoparticles are 0, 1, 2, and 3 wt.%, coir, and kenaf fibers are considered in weight percentages of 12, 13, 14, and 15, whereas thickness is regarded as 2,3,4 and 5 mm. This research evaluates the mechanical features of this hybrid composite fabricated using a vacuum bag molding process. The different composite samples are tested using mechanical tests and subsequently optimized using the design of experiment (i.e., Taguchi method) and analysis of variance (ANOVA) method to arbitrate the best weight percent combination of the innovative hybrid composite. On the basis of the optimization results, the best composite sample obtained includes, 3 wt% of CNT, 15 wt% of kenaf, 15 wt% of Coir, and 4mm thickness of the sample, as it yields the highest tensile modulus and strength among all the hybrid composite samples. The outcomes from the research indicate that the hybridization of kenaf fibers into coir fibers, along with CNTs as fillers in the hybrid composite has enhanced the overall tensile strength, and flexural strength of the hybrid composite in comparison to the coir composite and kenaf composite alone, depicting the superiority of natural fiber hybrid composite over synthetic fiber hybrid composite.

Artificial neural networks (ANNs) are one of the artificial intelligence techniques used in real-world problems and applications encountered in almost all industries such as education, health, chemistry, food, informatics, logistics, transportation. ANN is widely used in many techniques such as optimization, modelling, classification and forecasting, and many empirical studies have been carried out in areas such as planning, inventory management, maintenance, quality control, econometrics, supply chain management and logistics related to ANN. The most important and just as hard stage of ANNs is the learning process. This process is about finding optimal values in the search space for different datasets. In this process, the values generated by training algorithms are used as network parameters and are directly effective in the success of the neural network (NN). In classical training techniques, problems such as local optimum and slow convergence are encountered. Meta-heuristic algorithms for the training of ANNs in the face of this negative situation have been used in many studies as an alternative. In this study, a new hybrid algorithm namely MVOSANN is suggested for the training of ANNs, using Simulated annealing (SA) and Multi-verse optimizer (MVO) algorithms. The suggested MVOSANN algorithm has been experimented on 12 prevalently classification datasets. The productivity of MVOSANN has been compared with 12 well-recognized and current meta-heuristic algorithms. Experimental results show that MVOSANN produces very successful and competitive results.