Computation

28 pages, 1376 KB

Open AccessArticle

Multi-Objective Optimization of a Multi-Server Retrial Machine Repair System with Orbital Search and Synchronous Vacation

by Lee-Wen Chiu, Ming-Chin Chen, Tzu-Hsin Liu and Fu-Min Chang

Computation 2026, 14(7), 153; https://doi.org/10.3390/computation14070153 - 2 Jul 2026

Abstract

This paper investigates a multi-server retrial machine repair system that incorporates orbital search and a synchronous vacation mechanism. The system features standby units and examines two potential vacation scenarios for servers, reflecting real-world situations, such as technical staff in teaching hospitals taking periodic [...] Read more.

This paper investigates a multi-server retrial machine repair system that incorporates orbital search and a synchronous vacation mechanism. The system features standby units and examines two potential vacation scenarios for servers, reflecting real-world situations, such as technical staff in teaching hospitals taking periodic administrative or training vacations. We formulate a mathematical model using birth-and-death processes to establish the governing equation and propose a recursive matrix method to systematically derive the steady-state probabilities. Performance measures, including system availability, the expected number of failed units in the orbit, and expected waiting times, are derived. To address the conflicting objectives of minimizing total operating costs, minimizing expected waiting times, and maximizing system availability, we construct a tri-objective optimization problem. By implementing a multi-objective genetic algorithm, we identify a set of Pareto-optimal frontiers and reveal the explicit financial and operational trade-offs among these competing criteria. Numerical experiments and sensitivity analyses demonstrate that enhancing the automated retrial rate and managing the emergency repair rate are most critical to minimizing system downtime. Furthermore, the joint optimization of server capacities and vacation schedules effectively eliminates operational redundancy, showing that near-perfect equipment availability (up to 0.999) can be achieved with only marginal increases in cost. This research provides administrators with a robust decision-making framework to optimize technical resource management while ensuring near-perfect equipment availability in real-world environments. Full article

(This article belongs to the Section Computational Engineering)

31 pages, 3844 KB

Open AccessArticle

Competing Risks with Common Shocks: Joint Survival, Copulas, Censoring, Frailty, and Marshall–Olkin Models

by Cristian David Correa-Álvarez, Mario Cesar Jarramillo-Elorza and Osnamir Elias Bru-Cordero

Computation 2026, 14(7), 152; https://doi.org/10.3390/computation14070152 - 2 Jul 2026

Abstract

This study examines likelihood-based estimation of the joint survival function

S (t_{1}, t_{2}) = Pr {T^{(1)} > t_{1}, T^{(2)} > t_{2}}

for systems with two competing failure [...] Read more.

This study examines likelihood-based estimation of the joint survival function

S (t_{1}, t_{2}) = Pr {T^{(1)} > t_{1}, T^{(2)} > t_{2}}

for systems with two competing failure modes observed under right censoring. Rather than introducing a new distributional family, the study compares established dependence mechanisms within a common observed-data framework. Exponential and Weibull margins are combined with three types of dependence: Archimedean copulas, represented by the Gumbel and Clayton families; shared gamma frailty, used to model latent measurement-level heterogeneity; and Marshall–Olkin extensions, used to represent common shocks and simultaneous failures. The same observation scheme, likelihood construction, censoring design, and performance criteria are used across models. Model performance is evaluated through Monte Carlo simulation using bias, integrated mean squared error, and empirical coverage, and the workflow is illustrated with the Device G reliability data. The results show that ignoring dependence can distort joint survival estimates, especially under moderate or high censoring. They also show that copula, frailty, and Marshall–Olkin specifications can lead to different reliability assessments because they encode different stochastic mechanisms. The estimation workflow includes multi-start optimization and diagnostics for boundary solutions, Hessian stability, and irregular likelihood behavior. Full article

(This article belongs to the Section Computational Social Science)

18 pages, 755 KB

Open AccessArticle

Modeling of Biological Neural Networks Based on Neuronal Functions and Connectivity Patterns

by Hongfei Zhao, Yuhang Zhen, Yi Huang and Ying Liu

Computation 2026, 14(7), 151; https://doi.org/10.3390/computation14070151 - 1 Jul 2026

Abstract

Despite recent advances in the reconstruction of biological neural networks, the generative principles underlying the structural properties of these networks remain incompletely understood. Neurons in the brain belonging to different functional classes, such as sensory neurons, interneurons, or motor neurons, exhibit distinct connectivity [...] Read more.

Despite recent advances in the reconstruction of biological neural networks, the generative principles underlying the structural properties of these networks remain incompletely understood. Neurons in the brain belonging to different functional classes, such as sensory neurons, interneurons, or motor neurons, exhibit distinct connectivity asymmetry patterns. Here, we analyze the differences in connectivity patterns among these types, focusing on the asymmetry between in-degree and out-degree. Our analysis reveals that sensory neurons tend to exhibit a predominance of outgoing connections (negative asymmetry), motor neurons a predominance of incoming connections (positive asymmetry), and interneurons a more balanced connectivity profile. To capture these type-specific features, we propose an extended network growth model in which nodes are assigned to predefined functional types, each with distinct initial attractiveness for incoming and outgoing edges. Simulations demonstrate that our model can reproduce the observed asymmetry indices of different neuron types in biological neural networks and can also generate diverse degree distribution shapes. This work offers a phenomenological generative framework that links neuron type identity to connectivity asymmetry, and it provides a baseline for future studies that incorporate additional biological constraints. Full article

(This article belongs to the Section Computational Biology)

► Show Figures

Figure 1

18 pages, 7382 KB

Open AccessArticle

Computational Investigation of Friction Stir Processing of Ti-6Al-4V Alloy for Biomedical Applications Using FEM and Taguchi Design

by Nebojša Zdravković, Dragan S. Džunić, Živana Jovanovic Pešić and Dalibor Nikolić

Computation 2026, 14(7), 150; https://doi.org/10.3390/computation14070150 - 30 Jun 2026

Abstract

Friction stir processing (FSP) is an advanced solid-state surface modification technique for biomedical titanium alloys. This study presents a computational investigation of FSP applied to Ti-6Al-4V alloy through three-dimensional finite element modeling and Taguchi-based statistical optimization. A Taguchi L9 orthogonal array evaluated rotational [...] Read more.

Friction stir processing (FSP) is an advanced solid-state surface modification technique for biomedical titanium alloys. This study presents a computational investigation of FSP applied to Ti-6Al-4V alloy through three-dimensional finite element modeling and Taguchi-based statistical optimization. A Taguchi L9 orthogonal array evaluated rotational speed (400–1000 rpm), traverse speed (50–100 mm/min), shoulder diameter (6–18 mm), and pin diameter (2–6 mm), reducing the required simulations from 81 (full factorial) to nine (88.9% reduction). A calibrated friction model (μ = 0.35/0.25/0.20 for 400/800/1000 rpm, F = 6000 N) yielded maximum temperatures of 870–1384 °C; all predicted temperatures remained below the melting point of Ti-6Al-4V (1660 °C). These values are consistent with experimentally reported ranges for FSW/FSP of Ti-6Al-4V. Traverse speed is the dominant parameter (ANOVA contribution: 63.1%, F = 10.44), followed by rotational speed (26.7%) and shoulder diameter (4.1%). Simulation 3 (400 rpm, 100 mm/min, Ds = 18 mm, T_max = 870 °C) appears to be the most promising thermal condition for preserving the fine-grained α + β microstructure, as it remains below the β-transus temperature (980 °C) throughout the processed zone. Full article

(This article belongs to the Special Issue Advanced Computational Methods and Multiphysics Modeling in Bioengineering and Complex Systems)

► Show Figures

Figure 1

27 pages, 4696 KB

Open AccessArticle

Dynamic Defense Mechanism for Programmable Logic Controllers: A Heterogeneous Multi-Core Architecture with Rapid Nanosecond-Scale Threat Perception

by Delei Nie, Jingjing Hu, Xin Wang, Yu Li, Jiangxing Wu, Farrukh Hanif and Renhai Feng

Computation 2026, 14(7), 149; https://doi.org/10.3390/computation14070149 - 28 Jun 2026

Abstract

Existing PLC security solutions face a fundamental conflict between stringent real-time requirements and robust protection: traditional IT security mechanisms (e.g., encryption, authentication) introduce unacceptable latency, while software-based redundancy schemes operate at millisecond scale and remain vulnerable to common-cause failures. To bridge this gap, [...] Read more.

Existing PLC security solutions face a fundamental conflict between stringent real-time requirements and robust protection: traditional IT security mechanisms (e.g., encryption, authentication) introduce unacceptable latency, while software-based redundancy schemes operate at millisecond scale and remain vulnerable to common-cause failures. To bridge this gap, this study proposes MimicPLC v1.0, a dynamic defense mechanism based on a heterogeneous multi-core architecture that integrates threat perception, dynamic fault tolerance, and rapid recovery within a single chip, thereby reconciling real-time determinism with proactive security in industrial control systems. The architecture integrates three distinct CPU cores (MIPS, ARM, and RISC-V) within a single system-on-chip (ESC0830), coordinated by a dedicated hardware-based mimic scheduling subsystem. This subsystem performs real-time, loosely coupled, transaction-level consistency checks on the AHB-Lite bus operations of the heterogeneous processors, achieving nanosecond-scale arbitration latency for threat detection. We evaluate the proposed design using an industrial-strength testbed, incorporating a custom development board and the Synopsys Verdi simulation environment, under critical attack scenarios including Denial-of-Service (DoS), replay, code injection, and parameter overwrite attacks. The system maintains continuous operation through adaptive redundancy, demonstrating attack perception within 73 clock cycles and leveraging instruction-set asymmetry for effective threat containment. Rigorous validation, including 100 consecutive parameter override attacks, confirms a 100% interception rate within our tested attack scenarios, with zero false positives observed. The design complies with the IEC 61131-3 real-time standard, exhibiting a worst-case recovery duration of 9.3 ms and a 95% confidence interval for recovery latency of [4.0354, 4.0363] ms. This work pioneers a paradigm of rapid-detection endogenous security with nanosecond-scale arbitration for next-generation industrial control systems. Full article

(This article belongs to the Section Computational Engineering)

21 pages, 10357 KB

Open AccessArticle

A Multi-View Graph Learning Framework for Bearing Fault Diagnosis with Adaptive Fusion

by Xueyi Li, Chaolun Wang, Jiannan Dong, Zhilin Dong and Tianyang Wang

Computation 2026, 14(7), 148; https://doi.org/10.3390/computation14070148 - 27 Jun 2026

Abstract

Bearing fault diagnosis methods based on single sensors often suffer from reduced accuracy due to limited information. Although multi-sensor systems provide richer vibration information, the high dimensionality and complexity of these signals still pose challenges for effective feature extraction and fusion. In addition, [...] Read more.

Bearing fault diagnosis methods based on single sensors often suffer from reduced accuracy due to limited information. Although multi-sensor systems provide richer vibration information, the high dimensionality and complexity of these signals still pose challenges for effective feature extraction and fusion. In addition, many existing deep learning-based fusion methods rely on a single analysis domain or simple feature concatenation, making it difficult to fully exploit the complementarity among raw temporal signals, time-domain statistical features, and frequency-domain characteristics. To address these issues, this paper proposes a multi-view graph-based fault diagnosis framework with adaptive fusion, termed MDEGCN, for bearing condition identification. Specifically, non-overlapping vibration windows are treated as graph nodes, and three graph views are constructed to capture temporal proximity, time-domain similarity, and frequency-domain correlation, respectively. Each graph view is processed by an enhanced graph neural network branch to learn view-specific representations, and an adaptive, differentiable fusion mechanism is introduced to integrate complementary information from different views for final fault classification. Experiments on the Northeast Forestry University and Politecnico di Torino bearing datasets were conducted under a purged blocked split protocol to reduce potential information leakage between adjacent windows. Additional hard settings with a low training ratio further evaluate the robustness of the proposed framework under limited labelled data. The experimental results demonstrate that MDEGCN achieves competitive diagnostic performance and provides an effective multi-view representation learning strategy for bearing fault diagnosis. Full article

(This article belongs to the Special Issue Neural Network and Large Model-Driven Fault Diagnosis and Intelligent Operation and Maintenance for Rotating Machinery)

► Show Figures

Figure 1

13 pages, 10848 KB

Open AccessArticle

Numerical Prediction Study on Dynamic Characteristics of Key Components of a Variable-Speed Hydro-Generator Unit Under Load Rejection Conditions

by Tao Liu, Tengda Xu, Fei Ye, Huili Bi, Hongyu Chen, Xijie Song, Zan Zhou and Zhengwei Wang

Computation 2026, 14(7), 147; https://doi.org/10.3390/computation14070147 - 26 Jun 2026

Abstract

To evaluate the structural safety of variable-speed pumped-storage units under extreme transient conditions, this paper focuses on a variable-speed unit at a specific pumped-storage power plant. Based on boundary conditions measured during on-site load shedding tests, a three-dimensional, unidirectional fluid–structure interaction numerical model [...] Read more.

To evaluate the structural safety of variable-speed pumped-storage units under extreme transient conditions, this paper focuses on a variable-speed unit at a specific pumped-storage power plant. Based on boundary conditions measured during on-site load shedding tests, a three-dimensional, unidirectional fluid–structure interaction numerical model was established, incorporating stationary components such as the volute, base ring, top cover, and bottom ring. A numerical prediction and analysis of the dynamic stresses and deformations of key components were conducted for a hazardous scenario in which all units shed load simultaneously and the volute pressure reached its peak. The results show that during the load shedding process, the maximum static stress in the stationary components was 79.5 MPa, and the maximum displacement was 0.066 mm; both occurred 46.01 s after load shedding at the junction between the guide vane outlet edge and the top cover, and this value is far below the material’s yield strength of 490 MPa. Preliminary numerical evaluations indicate that the unit’s stationary components meet strength design requirements under this extreme transient condition. Furthermore, the study revealed the time lag mechanism between the peak hydraulic load and the peak structural stress in the top cover. The numerical prediction method established in this study can provide technical support for the structural safety assessment of transient processes in variable-speed units. Full article

(This article belongs to the Section Computational Engineering)

► Show Figures

Figure 1

32 pages, 2491 KB

Open AccessArticle

A Spectral-fPINN Framework for Fractional Optimal Control Problems

by Yonis Gulzar and Ishtiaq Ali

Computation 2026, 14(7), 146; https://doi.org/10.3390/computation14070146 - 25 Jun 2026

Abstract

Fractional optimal control problems provide an effective mathematical framework for modeling dynamical systems with memory, hereditary behavior, and anomalous diffusion effects. However, the nonlocal nature of Caputo fractional operators and the reduced regularity of fractional solutions pose significant challenges for the development of [...] Read more.

Fractional optimal control problems provide an effective mathematical framework for modeling dynamical systems with memory, hereditary behavior, and anomalous diffusion effects. However, the nonlocal nature of Caputo fractional operators and the reduced regularity of fractional solutions pose significant challenges for the development of accurate and efficient computational methods. In this paper, we develop a spectral-fractional Physics-Informed Neural Network (Spectral-fPINN) framework for solving fractional optimal control problems governed by Caputo fractional differential equations. The proposed methodology combines normalized shifted Legendre spectral approximations, fractional operational matrix formulations, and physics-informed optimization within a unified computational framework. Unlike conventional PINN and fPINN approaches, which directly approximate the unknown solution variables, the proposed framework predicts the spectral coefficient vectors associated with the shifted Legendre basis functions, yielding a low-dimensional global representation with improved approximation efficiency. Caputo fractional derivatives are evaluated through spectral operational matrices, while the resulting optimization problem is discretized using Gauss–Legendre quadrature and solved through gradient-based optimization. In addition, a theoretical analysis of the proposed Spectral-fPINN framework is presented, including approximation, consistency, stability, and convergence results, together with error estimates and residual control properties. Several benchmark linear and nonlinear fractional optimal control problems are investigated to validate the proposed methodology. The numerical results demonstrate excellent agreement with exact solutions, very small residual errors, and rapid spectral coefficient decay, confirming the high-order accuracy and robustness of the proposed approach. Overall, the proposed Spectral-fPINN framework provides an accurate, stable, and computationally efficient methodology for solving a broad class of fractional optimal control problems. Full article

(This article belongs to the Special Issue Nonlinear System Modelling and Control—2nd Edition)

► Show Figures

Figure 1

24 pages, 8781 KB

Open AccessArticle

Sub-Second Prediction of External Flow Fields Around a Ground Vehicle Using a Surrogate Model

by Roy Koomullil, Emmanuel Ramogi, Feroz Mohamed Iqbal, Peter Rynes, Vladimir Vantsevich, Vamshi Korivi and Nathan Tison

Computation 2026, 14(7), 145; https://doi.org/10.3390/computation14070145 - 25 Jun 2026

Abstract

Predicting the wind field around military vehicles during extended missions is crucial to avoid detectability by infrared (IR) devices. This is a challenging task because of the geometric complexity of the vehicles and the unpredictable nature of wind direction, which can shift abruptly [...] Read more.

Predicting the wind field around military vehicles during extended missions is crucial to avoid detectability by infrared (IR) devices. This is a challenging task because of the geometric complexity of the vehicles and the unpredictable nature of wind direction, which can shift abruptly and have a significant impact on the flow field and heat transfer. Computational fluid dynamics (CFD) is routinely used to calculate flow fields around ground vehicles. However, this requires extensive computational time and memory, making it unsuitable for real-time analysis. To address these challenges, this paper focuses on machine learning (ML) techniques for accurate wind field prediction in real time for unseen wind directions within the sampled range. Reduced order modeling (ROM) is used for dimensionality reduction of flow field data derived from high-fidelity CFD simulations. ML models are trained using low-dimensional data from the ROM, and the predicted low-dimensional data for unseen wind directions by the trained ML model is used to reconstruct the flow field. ROM, in conjunction with ML techniques, offers a substantial reduction in analysis time while maintaining the ability to predict the flow field accurately. In this study, a neural network architecture with three output formulations trained using ROM data was used for the predictions, and the accuracy of the formulations was evaluated by comparing them with the CFD results. An optimal ML model is identified by varying the number of hidden layers and neurons within those layers. The developed ROM- and ML-based approach was able to predict the unseen flow field in less than a second, while a single CFD simulation required approximately 2.6 h per wind direction. Full article

(This article belongs to the Special Issue Advances in Computational Methods for Fluid Flow—2nd Edition)

► Show Figures

Figure 1

41 pages, 11772 KB

Open AccessArticle

An Uncertainty-Aware Computational Framework for Dimensional Error Prediction in Ceramic Additive Manufacturing Under Variable Material and Process Conditions

by Mahmoud AlJamal, Nawal Louzi, Mohammad Q. Al-Jamal, Luay Tahat, Ala Mughaid and Qasim Aljamal

Computation 2026, 14(7), 144; https://doi.org/10.3390/computation14070144 - 24 Jun 2026

Abstract

Ceramic additive manufacturing offers strong potential for fabricating geometrically complex and application-specific components, yet achieving reliable dimensional fidelity remains challenging because dimensional deviation is governed by highly coupled material, process, thermal, and environmental factors. To address this problem, this study proposes an uncertainty-aware [...] Read more.

Ceramic additive manufacturing offers strong potential for fabricating geometrically complex and application-specific components, yet achieving reliable dimensional fidelity remains challenging because dimensional deviation is governed by highly coupled material, process, thermal, and environmental factors. To address this problem, this study proposes an uncertainty-aware computational framework for dimensional error prediction in ceramic 3D printing under variable material and process conditions. The contribution is positioned as a system-level integration of established learning, uncertainty estimation, calibration, and reliability-interpretation components within a ceramic additive manufacturing dimensional-error prediction workflow, rather than as a fundamental methodological breakthrough. The validation is conducted using the publicly available Ceramic 3D Printing Process Control Dataset, a 1000-sample tabular dataset, and the resulting findings are therefore interpreted as dataset-specific computational evidence rather than direct proof of industrial deployment readiness. The methodology begins with a structured data-driven preprocessing pipeline that transforms the Ceramic 3D Printing Process Control Dataset into a multi-condition feature space through data cleaning, one-hot material encoding, min–max normalization, and engineered descriptors capturing extrusion–speed balance, thermal gradients, cooling intensity, deposition density, and material-conditioned interactions. A multi-branch deep computational architecture is then developed to encode material, process, thermal-environmental, and engineered-feature streams separately, followed by adaptive cross-condition fusion to learn nonlinear dependencies across ceramic printing regimes. To improve reliability beyond deterministic regression, the framework jointly models aleatoric and epistemic uncertainty and incorporates calibration refinement to align predictive confidence with observed error behavior, thereby enabling preliminary reliability-oriented interpretation of stable and high-risk operating conditions. Experimental results demonstrate that the full model achieves the best overall within-dataset performance, with a test MAE of 0.0118, RMSE of 0.0172,

R^{2} = 0.999

, MAPE of 1.74%, calibration error of 0.003, PICP of 0.996, reliability score of 0.992, and a stable prediction rate of 98.7%. Although these values indicate strong predictive behavior under the current structured dataset, the exceptionally high

R^{2}

should be interpreted cautiously because external experimental validation, larger measured datasets, and cross-machine ceramic printing trials are still required. These findings show that the proposed framework provides an effective system-level computational strategy for dataset-specific reliability-aware dimensional quality prediction in ceramic additive manufacturing and offers a preliminary data-driven foundation for uncertainty-aware intelligent process optimization. Full article

(This article belongs to the Special Issue Computational Methods in Structural Optimization)

► Show Figures

Figure 1

28 pages, 2694 KB

Open AccessSystematic Review

Human Digital Twins in Personalized Medicine: A Systematic Review and Bibliometric–Thematic Synthesis of Methodological Advances and Clinical Applications

by Carlotta Fontana and Sina Zinatlou Ajabshir

Computation 2026, 14(7), 143; https://doi.org/10.3390/computation14070143 - 23 Jun 2026

Abstract

Human digital twins (HDTs) are patient-specific computational models that combine medical imaging, physiological measurements and predictive algorithms. They are moving from an exciting concept to a realistic clinical opportunity. The key question is no longer whether HDTs can be built. The key question [...] Read more.

Human digital twins (HDTs) are patient-specific computational models that combine medical imaging, physiological measurements and predictive algorithms. They are moving from an exciting concept to a realistic clinical opportunity. The key question is no longer whether HDTs can be built. The key question is which methods are mature enough to support clinical decisions and what is still missing for routine use. This systematic review maps the methodological landscape of HDTs and highlights practical bottlenecks that limit clinical translation. A PRISMA 2020 guided search of PubMed, Scopus, IEEE Xplore, and the Cochrane Library, covering publications from 2016 to 2026, identified 151 eligible studies. Bibliometric mapping and thematic synthesis were used to characterize research clusters, computational paradigms, and collaboration patterns. Three dominant application streams were identified: cardiovascular HDTs for hemodynamic simulation and procedural planning, musculoskeletal HDTs for biomechanics-driven orthopedic innovation, and neurological HDTs integrating neuroimaging with computational neuroscience. Across domains, the strongest technical trend is the rise in hybrid pipelines that combine physics-based simulation, including finite element and computational fluid dynamics models, with machine learning for segmentation, parameter identification, reduced-order modeling, and faster inference. However, reporting of verification, validation, uncertainty quantification, and explicit context of use remains uneven and prospective clinical evidence is still limited. Overall, the literature shows rapid progress toward clinically credible HDTs, while highlighting the need for scalable computation, standardized credibility pipelines, and workflow-integrated platforms to support safe and reproducible clinical adoption. Full article

(This article belongs to the Special Issue Computational Methods for Advanced Digital Twins in Biological and Engineered Systems)

► Show Figures

Graphical abstract

21 pages, 11433 KB

Open AccessArticle

Machine Learning-Assisted Synthesis of Self-Organizing SISO Control Systems with Guaranteed Lyapunov Stability

by Nurgul Shazhdekeyeva, Beket Kenzhegulov, Kamka Uteuliyeva, Gulash Kochshanova, Gulmira Nigmetova, Lyailya Kurmangaziyeva, Raigul Tuleuova, Saya Kenzhegulova and Raushan Moldasheva

Computation 2026, 14(6), 142; https://doi.org/10.3390/computation14060142 - 19 Jun 2026

Abstract

The proposed methodology combines analytical control laws with adaptive mechanisms and machine-learning-assisted modules based on regression trees, random forests, and extreme gradient boosting (XGBoost). Machine learning models are employed to approximate unknown nonlinear dynamics, compensate disturbances, and adjust controller parameters, while the overall [...] Read more.

The proposed methodology combines analytical control laws with adaptive mechanisms and machine-learning-assisted modules based on regression trees, random forests, and extreme gradient boosting (XGBoost). Machine learning models are employed to approximate unknown nonlinear dynamics, compensate disturbances, and adjust controller parameters, while the overall control structure is constrained by Lyapunov stability conditions. This ensures that the inclusion of data-driven components does not violate the fundamental requirement of system stability. The effectiveness of the proposed approach is evaluated through simulation experiments across three operating modes with varying degrees of nonlinearity and dynamic complexity. The results show that hybrid models incorporating ensemble machine learning methods improved performance compared with the analytical and adaptive baselines examined. XGBoost-based control achieves the lowest error values and the highest level of Lyapunov stability compliance (up to 99.3%). The main contribution of this study lies in the development of a unified synthesis framework in which machine learning is not used as a standalone control strategy but as a machine-learning-assisted support mechanism integrated into a theoretically grounded control architecture. The proposed approach provides a balance between adaptability, accuracy, and rigorous stability guarantees, suggesting potential applicability to simulation-based and offline-assisted control design tasks, while real-time embedded implementation requires additional computational optimization and validation. Full article

(This article belongs to the Section Computational Engineering)

► Show Figures

Figure 1

19 pages, 4025 KB

Open AccessArticle

Numerical and Experimental Studies on the Resistance of a Fast Catamaran in Accelerated Forward Speed Motion

by Apostolos Papanikolaou and Yan Xing-Kaeding

Computation 2026, 14(6), 141; https://doi.org/10.3390/computation14060141 - 18 Jun 2026

Abstract

This paper provides comprehensive numerical and experimental studies on the unsteady resistance of the world’s first battery-driven, zero-emissions high-speed catamaran, the MS Medstraum, in accelerated forward speed motion. These studies suggest that for a certain speed range of around Froude 0.50 (the [...] Read more.

This paper provides comprehensive numerical and experimental studies on the unsteady resistance of the world’s first battery-driven, zero-emissions high-speed catamaran, the MS Medstraum, in accelerated forward speed motion. These studies suggest that for a certain speed range of around Froude 0.50 (the so-called last hump of wave resistance), the corresponding unsteady resistance is significantly less than the originally anticipated value, namely, up to 40% less when adding to the steady resistance, the conventional added mass term. This surprising result could be explained by both experimental resistance tests and CFD calculations, as well as by inspection of the numerically generated wave patterns. Thus, care must be taken when applying the traditional approach to the unsteady resistance of a ship in accelerated or decelerated forward speed motion. As such, this positively affects the estimation of the required power capacity to accelerate the ship to full operational speed. This leads to reduced (fitted) battery weight and positively affects the ship’s displacement, allowing the vessel to achieve higher speeds. The present research finally yielded notable results of interest for seakeeping and ship maneuvering simulation studies; namely, comprehensive CFD simulations for the studied slender catamaran have shown that calculated added mass values for surge motion in real-flow conditions are up to six times higher than those initially estimated by ideal flow potential theory methods. Full article

(This article belongs to the Special Issue Advances in Computational Methods for Fluid Flow—2nd Edition)

► Show Figures

Figure 1

19 pages, 1907 KB

Open AccessArticle

An Enhanced Latency-Bounded GPU-Resident Pipeline for Real-Time Market Stream Visualization

by Donia Y. Badawood and Fahd M. Aldosari

Computation 2026, 14(6), 140; https://doi.org/10.3390/computation14060140 - 17 Jun 2026

Abstract

High-Frequency Trading (HFT) dashboards require rapid reception, aggregation, and visualization of order book and trade update streams that may arrive at multi-million message rates. Conventional CPU-based and CPU-GPU hybrid visualization pipelines can suffer from significant delays during periods of burst due to CPU-mediated [...] Read more.

High-Frequency Trading (HFT) dashboards require rapid reception, aggregation, and visualization of order book and trade update streams that may arrive at multi-million message rates. Conventional CPU-based and CPU-GPU hybrid visualization pipelines can suffer from significant delays during periods of burst due to CPU-mediated rendering, synchronization, kernel launch overhead, and copies on the host. This paper presents a visualization pipeline that is entirely resident on the graphics processor with zero-copy access to NIC accessible pinned buffers, persistent CUDA processing, fused stage execution of the parse-aggregate pipeline, and persistent CUDA OpenGL buffer interoperation. The goal is not to reach production status but rather to see whether host-to-host data movement can be decreased and whether the stages of GPU processing can be consolidated to improve latency, throughput and frame cadence in controlled HFT-style workloads. The evaluated workstation achieved a mean ingest-to-pixel latency of 6.3 ms using the proposed design compared to 29.4 ms for the current design, with sustained throughput of 10.2 million messages per second, which is 20 times greater than the current design, and a steady-state range of 185 to 192 frames per second with a burst floor of 178 frames per second for the proposed design. The improvement observed can be attributed to both the zero-copy ingestion and fused persistent kernel execution. Based on the obtained results, the proposed method of use of this technique in the implementation of real-time financial visualization under the proposed conditions is possible. More general testing is still required on other NICs, other generations of GPUs and PCIe configurations, workload traces, and actual exchange feeds. Full article

(This article belongs to the Section Computational Engineering)

► Show Figures

Figure 1

28 pages, 8867 KB

Open AccessArticle

From Particle Retention to Washout: Helical Bypass Geometry Reorganises Flow in Distal Anastomosis

by Sandor I. Bernad and Elena Silvia Bernad

Computation 2026, 14(6), 139; https://doi.org/10.3390/computation14060139 - 16 Jun 2026

Abstract

Current evaluation of bypass graft performance relies predominantly on wall shear stress metrics, even though thrombosis and atherogenesis are fundamentally governed by particle transport and residence within disturbed flow regions. This disconnect limits the ability of conventional hemodynamic indicators to capture mechanisms directly [...] Read more.

Current evaluation of bypass graft performance relies predominantly on wall shear stress metrics, even though thrombosis and atherogenesis are fundamentally governed by particle transport and residence within disturbed flow regions. This disconnect limits the ability of conventional hemodynamic indicators to capture mechanisms directly linked to graft failure. In this study, we investigate how helical bypass geometry reorganises the flow and, consequently, modifies transport behaviour within the distal anastomosis by combining experimentally validated flow visualisation with computational fluid dynamics under pulsatile conditions. Particle transport was quantified using a controlled injection of 151 tracers, enabling direct assessment of retention and washout across the graft–anastomosis system. The straight configuration exhibited persistent recirculation structures that promoted localised particle retention and delayed clearance. In contrast, the helical geometry disrupted these structures, enhancing flow mixing and accelerating downstream transport. At late stages of the cardiac cycle, the helical configuration reduced residual particle retention by approximately 43% compared to the straight bypass. These findings demonstrate a transition from recirculation-driven retention to washout-dominated transport, providing a mechanistic basis for interpreting bypass performance beyond shear-based metrics. This transport-centred perspective provides a mechanistic link between flow organisation and particle residence, supporting the functional relevance of helical graft design while remaining distinct from direct modelling of biological thrombosis or atherogenesis. Full article

(This article belongs to the Section Computational Engineering)

► Show Figures

Figure 1

31 pages, 8778 KB

Open AccessArticle

An Explainable Multimodal Deep Learning Framework for Thyroid Nodule Diagnosis in Ultrasound Imaging Using Hybrid Vision Transformers and Med-PaLM

by Sathya Jayaraman, Ramkumar Sivasakthivel, Jayapriya Jayapal and Balakrishnan Chinnaiyan

Computation 2026, 14(6), 138; https://doi.org/10.3390/computation14060138 - 16 Jun 2026

Abstract

Thyroid tumors rank among the most frequently occurring endocrine cancers because early detection helps doctors deliver effective treatments that lead to better patient results. Ultrasound imaging enables the detection of thyroid nodules, yet medical professionals struggle to differentiate between benign and malignant nodules [...] Read more.

Thyroid tumors rank among the most frequently occurring endocrine cancers because early detection helps doctors deliver effective treatments that lead to better patient results. Ultrasound imaging enables the detection of thyroid nodules, yet medical professionals struggle to differentiate between benign and malignant nodules through their diagnostic tests. This study introduces a new medical framework that enables thyroid nodule diagnosis through ultrasound imaging. The proposed model combines advanced segmentation with feature extraction, classification, and reasoning components to create a complete system. The specialized segmentation method shows accurate results when it detects nodule boundaries, which leads to better analysis of specific regions. The Hybrid Vision Transformer (HVT) operates to capture detailed textural information together with complete environmental patterns, which boosts its ability to classify different elements. The proposed framework incorporates a Large Language Model (LLM), specifically Med-PaLM, to provide context-aware clinical reasoning and interpretation. The structured evaluation process uses Thyroid Imaging Reporting and Data System (TI-RADS)-based feature scoring to compare model results with designated clinical standards. The diagnostic process is enhanced through the use of a language model, which delivers contextual understanding and produces valuable information from features that have been extracted. The proposed model achieves excellent performance with accuracy at 98.5%, precision at 98.7%, recall at 98.4%, and F1-score at 98.5%, which demonstrates its capacity for accurate and equivalent performance across different classifications. The experimental results demonstrate that the model achieves better results than existing methods. The combination of multimodal data with clinical reasoning improves both the accuracy and the user experience of the system. The proposed framework provides an efficient, interpretable, and scalable solution for thyroid nodule diagnosis. Full article

(This article belongs to the Section Computational Biology)

► Show Figures

Graphical abstract

22 pages, 867 KB

Open AccessArticle

RankBridge: Privacy-Preserving Rank-Based Explanation Clustering for Heterogeneous Federated Phishing Detection

by Panhapiseth Lim, Priyanka Kumar, Richard Zanni and Timothy Lambdin

Computation 2026, 14(6), 137; https://doi.org/10.3390/computation14060137 - 15 Jun 2026

Abstract

Federated learning lets organizations train a shared model without pooling private data. The standard method, Federated Averaging, requires all participants to use the same input features, a condition that fails in cross-sector phishing detection, where banks analyze URL structure and hospitals analyze email [...] Read more.

Federated learning lets organizations train a shared model without pooling private data. The standard method, Federated Averaging, requires all participants to use the same input features, a condition that fails in cross-sector phishing detection, where banks analyze URL structure and hospitals analyze email content. We present RankBridge, a system that groups participants by comparing ranked lists of SHapley Additive exPlanations (SHAP) feature importance rather than model weights or gradients. Each participant trains a local LightGBM model, extracts the top-K features by SHAP importance, and sends a 60-byte ranked list of feature indices to a central server. The server applies rank correlation and Ward’s hierarchical clustering to identify similarly threatened organizations. RankBridge operates in two modes: ModelShare, where models are also shared within each discovered group for prediction ensembling, and RankOnly, where the server returns only a group label and each participant keeps their model private. Across 32 participants in five organization types, RankBridge (ModelShare) achieves F1

= 0.853

(AUC

= 0.926

) on synthetic data and F1

= 0.772

(AUC

= 0.812

) on real phishing data, and it is the only method to outperform isolated local training on both. On real heterogeneous data the standard baselines adapted to LightGBM, including Federated Averaging, retain a moderate thresholded F1 (≈0.73) but their ranking quality collapses to near-random (AUC

\approx 0.59

, PR-AUC

\approx 0.66

), whereas RankBridge sustains AUC

= 0.812

and PR-AUC

= 0.819

. RankBridge recovers the correct organizational groupings with Normalized Mutual Information (NMI)

= 0.973

. The rank-based grouping channel itself transmits 60 bytes per participant per round, roughly 10,000× less than a full model upload. Full article

(This article belongs to the Special Issue Selected Papers from the 57th International Carnahan Conference on Security Technology (the 57th Annual ICCST))

► Show Figures

Figure 1

24 pages, 782 KB

Open AccessArticle

Optimal Control and Cost-Effectiveness Analysis of Porosity-Driven Bone Remodeling Dynamics

by Moustafa El-Shahed, Kadi Alowais and Yousef Alnafisah

Computation 2026, 14(6), 136; https://doi.org/10.3390/computation14060136 - 12 Jun 2026

Abstract

This paper develops an optimal control framework for a mechanical–structural model of bone remodeling that couples osteocytes, osteoblasts, and osteoclasts with bone density, incorporating porosity-dependent feedback mechanisms. To represent clinically relevant interventions, three bounded control functions are introduced: anabolic stimulation of osteoblast activity, [...] Read more.

This paper develops an optimal control framework for a mechanical–structural model of bone remodeling that couples osteocytes, osteoblasts, and osteoclasts with bone density, incorporating porosity-dependent feedback mechanisms. To represent clinically relevant interventions, three bounded control functions are introduced: anabolic stimulation of osteoblast activity, anti-resorptive suppression of osteoclast-mediated resorption, and structural modulation of porosity feedback. The controlled system is shown to be mathematically well-posed, and the necessary optimality conditions are derived via Pontryagin’s Maximum Principle, leading to explicit characterizations of the optimal controls. The resulting state–adjoint system is solved numerically using a forward–backward sweep method. Numerical results demonstrate that the optimal intervention effectively suppresses osteoclast activity and drives the system toward higher, more stable bone density levels than the uncontrolled dynamics. In particular, the anti-resorptive control consistently plays the dominant role in shaping the optimal strategy. A cost-effectiveness analysis based on ACER, ICER, and the efficient frontier shows that strategies involving anti-resorptive inhibition achieve the greatest therapeutic gains at moderate cost, while additional controls yield only marginal improvements. Sensitivity analysis further indicates that parameters associated with osteoclast dynamics and bone formation have the strongest influence on density-related outcomes. Full article

(This article belongs to the Section Computational Biology)

► Show Figures

Figure 1

20 pages, 1663 KB

Open AccessArticle

Jacobi Elliptic Function Solutions for the Conformable Resonant Nonlinear Schrödinger Equation with Parabolic Nonlinearity

by Du’a Al-zaleq, Lewa’ Alzaleq and Suboh Alkhushayni

Computation 2026, 14(6), 135; https://doi.org/10.3390/computation14060135 - 11 Jun 2026

Abstract

In this study, we utilize the

ϕ^{6}

-model expansion method to derive a diverse set of Jacobi elliptic function solutions for the conformable resonant Nonlinear Schrödinger Equation (NLSE) with parabolic law nonlinearity. As the modulus of the Jacobi elliptic functions approaches 1 [...] Read more.

In this study, we utilize the

ϕ^{6}

-model expansion method to derive a diverse set of Jacobi elliptic function solutions for the conformable resonant Nonlinear Schrödinger Equation (NLSE) with parabolic law nonlinearity. As the modulus of the Jacobi elliptic functions approaches 1 and 0, the solutions transform into hyperbolic and trigonometric functions, respectively. This methodology yields various exact traveling wave solutions, including kink solitons, singular solitons, periodic solutions, and singular periodic solutions. Notably, this work represents the first investigation into identifying Jacobi elliptic function solutions for the conformable resonant NLSE. These results enhance the understanding of the nonlinear dynamical properties intrinsic to the NLSE. We use graphical illustrations to highlight the dynamical features of the solutions. Moreover, our approach showcases versatility in addressing other nonlinear partial differential equations, offering insights applicable to nonlinear optics, fluid dynamics, and quantum physics. Full article

(This article belongs to the Section Computational Engineering)

► Show Figures

Figure 1

Journal Description

Computation

Latest Articles

Journal Menu

Journal Browser

Highly Accessed Articles

Latest Books

E-Mail Alert

News

Topics

Conferences

Special Issues

Further Information

Guidelines

MDPI Initiatives

Follow MDPI