Archives of Computational Methods in Engineering最新文献

Alzheimer’s Disease (AD) is a neurodegenerative condition characterized by irreversible cognitive decline. Detecting AD early is challenging as symptoms typically manifest years after the disease onset, necessitating the identification of subtle biomarker changes, often detectable through various neuroimaging modalities. Computer-aided diagnostic models leveraging machine learning and deep learning offer promising avenues for analyzing diverse input modalities to aid in early AD detection. The present study aims to analyze recent trends in the methods utilized by researchers for early prediction of Alzheimer along with identifying key challenges in existing research. The study follows PRISMA methodology to provide a comprehensive analysis of studies published in the last five years, resulting in sixty-four studies. The studies are sourced from significant data repositories after careful inclusion and exclusion criteria. The analysis of studies reveals the utilization of various machine learning and deep learning architectures, emphasizing practitioner-oriented perspectives such as data sources, input modalities, feature extraction strategies, and validation techniques. Performance comparison of the methods elucidates the effectiveness of deep learning frameworks, particularly in handling multimodal data and facilitating multiclass classification. Notably, structural MRI emerges as the most utilized input modality, with potential improvements observed when combined with Diffusion Tensor Imaging (DTI). Furthermore, current challenges within the existing literature are addressed and provides recommendations for future research directions. This review serves as a valuable resource for both novice and experienced researchers, offering insights into the state of the art and guiding efforts towards improved Alzheimer’s disease prediction methodologies.

Over the past few decades, researchers have developed several metaheuristic algorithms following smart rules and strategies for solving high dimension optimization problems. However, such rapid advancements in the development of metaheuristic algorithms and their diverse applications present challenges in evaluating their relative effectiveness and adaptability to various problem types. To this end, this paper presents a comprehensive review and systematic evaluation on seagull optimization algorithm (SOA), one of the recent metaheuristic swarm optimization algorithms, that mimics migrating and hunting behaviour of seagull sea birds, for inspiring researchers to perform further research in this field. The review initiates with critical screening and evaluation of SOA published articles based on strict criteria to select 109 eligible articles for comprehensive review. This paper examines and explores past research on SOA including advancement, modifications, multi-objective versions, hybridization and real-world application areas. Additionally, the SOA articles percentage distribution was visualized in terms of improvements, journals, various publishers, and different domains of optimization problems. Key findings indicate that SOA performance have been improved mostly utilizing chaotic map strategy in 22% and levy flight mechanism in 18% of SOA publications. It was also revealed that Springer, IEEE and Elsevier have higher number of SOA publications having 19, 16 and 15%, respectively. Finally, concluding remarks and future research directions are provided for further investigation on SOA, particularly in the field of biomedical applications and sensitive tuning of its internal parameters.

This paper presents a detailed review of existing and emerging deep learning algorithms for time series forecasting in geotechnics and geoscience applications. Deep learning has shown promising results in addressing complex prediction problems involving large datasets and multiple interacting variables without requiring extensive feature extraction. This study provides an in-depth description of prominent deep learning methods, including recurrent neural networks (RNNs), convolutional neural networks (CNNs), generative adversarial network, deep belief network, reinforcement learning, attention and transformer algorithms as well as hybrid networks using a combination of these architectures. In addition, this paper summarizes the applications of these models in various fields, including mining and tunnelling, railway and road construction, seismology, slope stability, earth retaining and stabilizing structures, remote sensing, as well as scour and erosion. This review reveals that RNN-based models, particularly Long Short-Term Memory networks, are the most commonly used models for time series forecasting. The advantages of deep learning models over traditional machine learning, including their superior ability to handle complex patterns and process large-scale data more effectively, are discussed. Furthermore, in time series forecasting within the fields of geotechnics and geosciences, studies frequently reveal that deep learning methods tend to surpass traditional machine learning techniques in effectiveness.

Tunnels and subways have become essential features of the civil infrastructure system in recent years. Explosions inside tunnels and subway stations could endanger persons trapped inside and cause structural damage, resulting in more deaths and financial losses. Thus, a blast-resistant tunnel design is required to reduce the damage and fortify the tunnel against blast-related disasters. Several relevant literature publications on the subject were explored to acquire a better knowledge of how tunnels perform when subjected to blast loads. This paper examines various aspects of the behaviour and performance of tunnels subjected to internal explosions. It begins with a discussion of the importance of tunnels, then it explores the categories of internal explosions and their potential causes, the characteristics of blast waves in tunnel structures, and their effects on tunnel structures. It then delves into the response of blast resistance tunnels and the methodologies employed for assessing the blast resistance of tunnels, such as analytical approaches, numerical simulations, experimental testing, and the main failure modes of tunnels subjected to internal blast loading. Additionally, the factors that influence tunnel response are reviewed including the use of high-performance materials such as fibre-reinforced polymer (FRP) composites as well as ultra-high-performance concrete (UHPC). The review concludes by identifying potential areas for further research and development in the field of blast-resistant tunnel engineering.

Application placement in fog computing is a crucial aspect of designing and managing fog computing systems. As a consequence, optimized application management becomes an essential task. Numerous algorithms have been proposed in the literature for efficient placement of applications in fog to address the Fog Application Placement Problem (FAPP), however not much work was done in evaluation of optimization algorithms. The evaluations are needed for making practical and impactful decision that can contribute to the scalability, reliability, and cost-effectiveness of the fog. This evaluation work is targeted towards the search of most efficient algorithms for FAPP especially optimization algorithms. The evaluations presented in this work provide answers to the most essential research questions such as what is the need of optimization in fog environment, which optimization algorithm performs best and what is its future in fog computing. The evaluations in this paper firstly, focuses on metric-based evaluations, which evaluates Fog Utilization, Service Level Agreement (SLA) violation and Response time of various state of the art works collected from the literature The work also evaluates different type of Optimization algorithms and Optimization objectives side-by-side to see which type are best suited for FAPP.

Soils, particularly clayey soils, show desiccation cracking when drying. Soil desiccation cracking is a prevalent natural phenomenon involving complex physical processes and mechanisms, presenting significant challenges in developing numerical models. This review summarizes numerical methodologies for addressing soil cracking issues from microscopic to macroscopic scales. At microscales, the fundamental theory of the Young–Laplace equation and hemisphere approximation for water meniscus is introduced to investigate the attracting force between soil particles. Various numerical methods used to model the evolution of the water meniscus and the development of microcracks in soil are reviewed and compared here. At macroscales, coupled thermo-hydro-mechanical models are the mainstream approach for simulating desiccation cracking owing to varying temperature and moisture. Different numerical methods, such as mesh-based methods, mesh-free methods and particle-based methods, for addressing soil desiccation cracking are reviewed, including their advantages, disadvantages, and recommended application scenarios. Furthermore, the future perspectives for soil desiccation cracking are discussed combined with the peridynamics method, including the three-phase solid–liquid-gas medium model for water meniscus, parameter homogenization for multiscale models, thermo-hydro-mechanical coupling and elastoplastic peridynamic model, cracking and healing criteria, complex climatic and environmental conditions and the development of hybrid numerical models. This review provides not only an in-depth understanding of the mechanisms underlying soil desiccation cracking modelling but also numerical techniques for the digital implementation of theoretical models for soil desiccation cracking modeling.

Nowadays, the architectural design process became more complex, the exponential growth of the architectural field, Made the design more critical for the designers, and student to understand. While creating an outstanding building, it must be free of clashes and well-coordinated. The rapid development of computer technology has accelerated the progress of architectural technology, and the application of virtual reality technology has become more and more common. Virtual reality (VR) provides a completely digital world of interaction that enables the users to modify, edit, and transform digital elements responsively. It has strong integration and comprehensive characteristics and covers many technologies such as computer graphics, simulation, artificial intelligence, sensing, display, and network processing. This paper focuses on adapting the already-existing methods and tools in architecture to the VR environment under the sustainable and architectural design domain. For this purpose, this literature benefits from the semantically enriched data platforms of Scientific Articles on Virtual Reality, BIM, Architectural design, and green building, which in the end state the potentiality of Virtual Reality in the Architectural field through different aspects. It plays an important role in optimizing and simulating architectural buildings and improving them to an optimum design. It facilitates the design process to get the optimum design through different programs, which leads us to BIM and collaboration between team members in order to Imagine and coordinate between different trades. The VR illustrates the important role of Green Building, through simulation and tests of the building before initiating it. Not only the simulation but the life cycle. Whereas its importance in education for both Students and teachers.

The flow direction algorithm (FDA) has garnereds significant attention from researchers and is increasingly being utilized across various fields to address diverse optimization challenges. This algorithm draws on principles of flow direction towards the outlet with the lowest elevation in a drainage basin, emphasizing the differentiation in objective functions and the spatial distance to neighboring points. This study provides an extensive examination of the original FDA, as well as its modifications and hybridizations. Additionally, the applications of the FDA in various engineering domains are explained. In general, FDA’s average ranking is relatively high. Lastly, the paper outlines potential avenues for future research in the advancement of the FDA.

Ordinary, partial, and integral differential equations are indispensable tools across diverse scientific domains, enabling precise modeling of natural and engineered phenomena. The polynomial collocation method, a powerful numerical technique, has emerged as a robust approach for solving these equations efficiently. This review explores the evolution and applications of the collocation method, emphasizing its matrix-based formulation and utilization of polynomial sequences such as Chebyshev, Legendre, and Taylor series. Beginning with its inception in the late 20th century, the method has evolved to encompass a wide array of differential equation types, including integro-differential and fractional equations. Applications span mechanical vibrations, heat transfer, diffusion processes, wave propagation, environmental pollution modeling, medical uses, biomedical dynamics, and population ecology. The method’s efficacy lies in its ability to transform differential equations into algebraic systems using orthogonal polynomials at chosen collocation points, facilitating accurate numerical solutions across complex systems and diverse engineering and scientific disciplines. This approach circumvents the need for mesh generation and simplifies the computational complexity associated with traditional numerical methods. This comprehensive review consolidates theoretical foundations, methodological advancements, and practical applications, highlighting the method’s pivotal role in modern computational mathematics and its continued relevance in addressing complex scientific challenges.

This review examines the application of machine learning techniques for optimizing ceramic waste-based concrete, a sustainable alternative in construction. In the course of this work, numerous computational paradigms such as the Decision Trees, Random Forests, XGBoost, Artificial Neural Networks (ANNs), Bagging, AdaBoost, Gradient Boosting, Regression models as well as Support Vector Machines (SVMs) are discussed. Comparing to other models in this study, XGBoost and ANNs were shown to yield better results in terms of concrete properties hence revealing non-linear relationships in ceramic waste-concrete systems. However, there are also some shortcomings: small sample sizes were used, critical chemical features were not included, and critical hyperparameters were not tuned. The review emphasizes the need for larger, standardized datasets, incorporation of chemical composition data, and advanced techniques like deep learning and multi-objective optimization for future research. Such developments may further enhance the prediction precision and realism of the created model and subsequently ensure the long-lasting concrete through utilization of ceramic waste.