Computation

38 pages, 5320 KB

Open AccessArticle

Joint Optimization of Preservation Technology, Hybrid Payment Policies, and Prepayment Discounts for Non-Instantaneously Deteriorating Items with Shortages

by El-Awady Attia and Md Sharif Uddin

Computation 2026, 14(8), 193; https://doi.org/10.3390/computation14080193 - 20 Aug 2026

Abstract

Retailers of non-instantaneously deteriorating items must jointly set inventory, preservation technology, and payment decisions. Preservation technology reduces deterioration, but excessive investment increases operational costs, making the determination of an optimal preservation level essential for maximizing profit. Although preservation technology, hybrid payment schemes, and [...] Read more.

Retailers of non-instantaneously deteriorating items must jointly set inventory, preservation technology, and payment decisions. Preservation technology reduces deterioration, but excessive investment increases operational costs, making the determination of an optimal preservation level essential for maximizing profit. Although preservation technology, hybrid payment schemes, and prepayment discounts have been studied individually, their joint treatment alongside partially backlogged shortages remains largely unexplored. To address this gap, this study develops an inventory model that simultaneously incorporates preservation technology investment, a hybrid payment structure, advance payment combined with trade credit, optionally supplemented by a prepayment discount, and partially backlogged shortages for non-instantaneously deteriorating items. A classical optimization approach is employed, yielding quasi-closed-form solutions for the shortage and replenishment timing across four trade credit scenarios, while the profit-maximizing preservation investment level is identified through sensitivity analysis. Numerical examples and sensitivity analysis show that increasing the number of prepayment installments lowers the discount rate offered by the supplier; because this forgone discount outweighs the benefit of retaining capital longer, the retailer’s profit falls. Profit responds most strongly to purchasing cost, the advance payment period, and lead time. These results give retailers a practical basis for balancing preservation investment, payment structure, and shortage policy to maximize profitability. In the sensitivity analysis, profit varies by more than 45% over the tested range of the purchasing cost and by up to 21% depending on the number of prepayment installments negotiated with the supplier. That gives retailers a concrete ranked basis for prioritizing which contract terms to negotiate first. Full article

(This article belongs to the Section Computational Social Science)

► Show Figures

Figure 1

23 pages, 2139 KB

Open AccessArticle

Physiological Signal-Guided Uncertainty Management for Autonomous UAVs in Human–UAV Supervisory Control

by Jun Che, Feng Zhu and Hanbin Xiao

Computation 2026, 14(8), 192; https://doi.org/10.3390/computation14080192 - 20 Aug 2026

Abstract

This paper presents a physiological signal-guided uncertainty-management framework that integrates real-time indicators of the operator’s supervisory state into UAV autonomy to improve obstacle avoidance and risk-aware decision-making under uncertain conditions. During UAV supervisory control, multimodal physiological signals, including heart rate variability, blood pressure, [...] Read more.

This paper presents a physiological signal-guided uncertainty-management framework that integrates real-time indicators of the operator’s supervisory state into UAV autonomy to improve obstacle avoidance and risk-aware decision-making under uncertain conditions. During UAV supervisory control, multimodal physiological signals, including heart rate variability, blood pressure, electrodermal activity, and other cardiovascular or stress-related measures, are time-synchronized with UAV telemetry, perceived obstacle fields, planner confidence, environmental uncertainty estimates, and operator intervention logs, including waypoint edits, overrides, and replanning commands. These heterogeneous data streams are fused using a Bayesian hierarchical state-space framework to estimate latent supervisory states representing trust miscalibration, risk sensitivity, and situational-awareness degradation. The estimated states are then incorporated into the UAV decision-making stack as bounded uncertainty-management parameters that regulate safety margins, replanning priority, and risk preference without relaxing hard safety constraints. The framework was evaluated in a simulation-based human-in-the-loop study involving 24 operators and 180 UAV obstacle avoidance missions. Performance was assessed using held-out-operator mission success AUROC, proxy-state RMSE, mission success rate, mean minimum obstacle clearance, mission risk index, operator override rate, and replanning latency. The results support the feasibility of using physiological, behavioral, vehicle, and environmental information to adapt UAV supervisory control to uncertainty in both the operating environment and human supervisory readiness. Full article

(This article belongs to the Topic Intelligent Systems and Immersive Technologies for Human–Machine Interaction and Collaboration)

► Show Figures

Figure 1

24 pages, 502 KB

Open AccessArticle

Performance and Computational Cost of Full and Parameter-Efficient Fine Tuning for Arabic Sentiment Classification Across Training Set Sizes

by Teif Aldaajani, Morooj Alqurashi, Sarah Aljuaid and Maha Jarallah Althobaiti

Computation 2026, 14(8), 191; https://doi.org/10.3390/computation14080191 - 19 Aug 2026

Abstract

Pre-trained language models are typically adapted to downstream tasks via full fine tuning. However, this entails substantial computational and memory overhead. Parameter-efficient alternatives such as Low-Rank Adaptation (LoRA), Quantized Low-Rank Adaptation (QLoRA), and frozen backbone mitigate these costs, but empirical evidence on how [...] Read more.

Pre-trained language models are typically adapted to downstream tasks via full fine tuning. However, this entails substantial computational and memory overhead. Parameter-efficient alternatives such as Low-Rank Adaptation (LoRA), Quantized Low-Rank Adaptation (QLoRA), and frozen backbone mitigate these costs, but empirical evidence on how their performance–cost trade-offs change under low labeled data in Arabic remains limited. This paper compares four adaptation strategies: full fine tuning, frozen backbone, LoRA, and QLoRA for Arabic binary sentiment classification on the Hotel Arabic Reviews Dataset, using CAMeLBERT-Mix as the pre-trained encoder. The methods are evaluated under a unified experimental setting at three labeled-data levels: the full training set, 100 samples per class, and 25 samples per class. The evaluation metrics are reported as means and standard deviations across five random seeds. At the full-data level, full fine tuning, LoRA, and QLoRA achieve macro-F1 scores between 0.9569 and 0.9579 and are comparable within seed variability, while the frozen backbone exhibits performance that is approximately ten points lower. LoRA and QLoRA use approximately 35.0% less peak GPU memory than full fine tuning but require longer training times. Under reduced-data conditions, full fine tuning outperforms all other adaptation strategies with the differences being statically significant. Full article

(This article belongs to the Special Issue Recent Advances on Computational Linguistics and Natural Language Processing—2nd Edition)

► Show Figures

Figure 1

24 pages, 590 KB

Open AccessArticle

A Two-Stage Matheuristic for the Capacitated Arc Routing Problem with Vehicle Dependence

by Hugo Alexer Pérez-Vicente, Jonás Velasco and Luis E. Urbán-Rivero

Computation 2026, 14(8), 190; https://doi.org/10.3390/computation14080190 - 18 Aug 2026

Abstract

In the capacitated arc routing problem (CARP), a fleet of capacitated vehicles based at a depot must cover the streets of a network where the demand is located at the lowest possible total cost. Waste collection, street sweeping, winter gritting, and mail delivery [...] Read more.

In the capacitated arc routing problem (CARP), a fleet of capacitated vehicles based at a depot must cover the streets of a network where the demand is located at the lowest possible total cost. Waste collection, street sweeping, winter gritting, and mail delivery are among its best-known applications. This work introduces the CARP with vehicle dependence (CARP-VD), an extension in which the cost of servicing an edge, and that of traversing it without service, are specific to each vehicle type and formulates it as a mixed-integer linear program. A two-stage matheuristic is proposed: the first stage distributes the required edges among the vehicles without exceeding their capacities, and the second builds the route of each vehicle. A bound is derived that limits the optimality loss of this decomposition by its own deadheading cost. Both approaches are evaluated on 47 benchmark instances adapted from the literature under a common one-hour budget, and their robustness is assessed over six scenarios that vary the parameters of the adaptation. The matheuristic returns good-quality solutions in a fraction of the time on the smaller instances, and on those in which almost every edge requires service it improves the best solutions found by a commercial solver applied to the complete model by up to 44%. Full article

(This article belongs to the Special Issue Advances in Computational Methods for Logistics and Supply Chain Optimization)

► Show Figures

Figure 1

14 pages, 966 KB

Open AccessArticle

Reduced-Order Computational Modeling of Small UAV Acoustic Signatures and SNR-Based Passive Detection Range Using Harmonic Aeroacoustic Scaling

by David Sanchez-Hernandez, Guillermo Urriolagoitia-Sosa, Gerardo Reyes-Ruiz, Beatriz Romero-Angeles, Jacobo Martinez-Reyes, Julian Patiño-Ortiz, Miguel Patiño-Ortíz, Flavio Arturo Dominguez-Pacheco, Claudia Hernández-Aguilar, Alfonso Trejo-Enriquez, Candy Esmeralda Hernandez-Bravo, Jonathan Rodolfo Guereca-Ibarra, Luis Itzcoatl Lugo-Chacón and Jorge Alberto Gomez-Niebla

Computation 2026, 14(8), 189; https://doi.org/10.3390/computation14080189 - 15 Aug 2026

Abstract

Small unmanned aerial vehicle (UAV) acoustic signatures are relevant to environmental noise assessment, passive monitoring, and preliminary detectability analysis. This study presents a physics-informed reduced-order framework that combines blade-passing frequency (BPF) harmonic synthesis, rotational-speed acoustic scaling, propagation, and signal-to-noise ratio (SNR) threshold crossing. [...] Read more.

Small unmanned aerial vehicle (UAV) acoustic signatures are relevant to environmental noise assessment, passive monitoring, and preliminary detectability analysis. This study presents a physics-informed reduced-order framework that combines blade-passing frequency (BPF) harmonic synthesis, rotational-speed acoustic scaling, propagation, and signal-to-noise ratio (SNR) threshold crossing. The RPM–OASPL law was calibrated using ten digitized measurements for a four-rotor DJI Phantom II with Original 9450 propellers and validated, without refitting, against eleven Aftermarket 9443 measurements. The fitted exponent was m = 5.285, and the fitted reference level was Lref = 77.01 dB(A) at 5000 RPM and 1 m. A 100,000-realization Monte Carlo analysis that propagated ±100 RPM and ±0.5 dB digitization bounds, together with residual scatter, produced total 95% intervals of 5.01–5.55 for m and 76.58–77.43 dB(A) for Lref. Calibration yielded RMSE = 0.39 dB and R² = 0.9979; independent-configuration validation yielded MAE = 2.27 dB, RMSE = 2.38 dB, and R² = 0.9168. At 5000 RPM, nominal free-field threshold-crossing distances were 70.9, 22.4, and 7.1 m for quiet rural, semi-urban, and urban scenarios. Combined statistical 95% screening intervals were 41.3–121.7, 13.0–38.5, and 4.1–12.2 m, respectively. Sensitivity analyses show that β controls harmonic-specific range but not normalized OASPL, while ground interference, band-limited masking, and detector processing can materially change operational range. The framework is therefore a rapid, interpretable screening tool rather than a universal detector performance model. Full article

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

► Show Figures

Figure 1

26 pages, 335 KB

Open AccessArticle

Proximal

Z

-Condensing Operators via Simulation Functions and Applications

by Moosa Gabeleh and Maggie Aphane

Computation 2026, 14(8), 188; https://doi.org/10.3390/computation14080188 - 14 Aug 2026

Abstract

In this paper, we introduce and study proximal

Z

-condensing operators in strictly convex Banach spaces by combining simulation functions with measures of noncompactness. A Darbo-type best proximity point theorem is established, and several consequences corresponding to nonlinear condensing conditions are obtained. As [...] Read more.

In this paper, we introduce and study proximal

Z

-condensing operators in strictly convex Banach spaces by combining simulation functions with measures of noncompactness. A Darbo-type best proximity point theorem is established, and several consequences corresponding to nonlinear condensing conditions are obtained. As an application, a system of nonlinear ordinary differential equations is embedded into a non-self operator problem on an enlarged product space; in this formulation, best proximity points are shown to be equivalent to classical solutions of the system. We also prove a Krasnoselskii-type best proximity point theorem for the sum of a simulation-function contraction and a compact operator and apply it to a nonlinear matrix-valued integral equation. Finally, a multiplicative best proximity point theorem is obtained in strictly convex Banach algebras and is used to study a nonlinear integral equation. The results provide a unified operator-theoretic framework for additive and multiplicative equations involving non-self mappings. Full article

(This article belongs to the Section Computational Engineering)

31 pages, 2423 KB

Open AccessArticle

Advanced Solvers for Nonlinear Heat Conduction with Generalized Constitutive Laws

by Madison Phelps and Malgorzata Peszynska

Computation 2026, 14(8), 187; https://doi.org/10.3390/computation14080187 - 14 Aug 2026

Abstract

In this paper, we discuss solvers for solving the discretized nonlinear heat equation with generalized constitutive laws. The model features multi-valued graphs, and the nonlinear functions are at best semi-smooth. Their solutions have low regularity. We identify appropriate global and local nonlinear solvers [...] Read more.

In this paper, we discuss solvers for solving the discretized nonlinear heat equation with generalized constitutive laws. The model features multi-valued graphs, and the nonlinear functions are at best semi-smooth. Their solutions have low regularity. We identify appropriate global and local nonlinear solvers and propose and evaluate a suite of accelerated iterative algorithms with additional enhancements that improve the convergence of schemes compared to those from the literature. We illustrate these solvers with numerical examples. Full article

(This article belongs to the Section Computational Engineering)

► Show Figures

Figure 1

29 pages, 670 KB

Open AccessArticle

Recovering Latent Association from Likert-Type Responses: A Monte Carlo Comparison of Pearson, Spearman, Kendall, and Polychoric Correlations Under Discretization, Response Perturbation, and Gumbel-Copula Dependence

by Diógenes de Jesus Ramirez-Ramirez, Osnamir Elias Bru-Cordero and Cristian David Correa-Álvarez

Computation 2026, 14(8), 186; https://doi.org/10.3390/computation14080186 - 13 Aug 2026

Abstract

Likert-type responses are often analyzed as continuous scores, although their categories are ordered rather than truly metric. This study uses Monte Carlo simulation to examine how ordinal discretization, response perturbation, and non-Gaussian dependence affect four association estimators: Pearson correlation, Spearman correlation, Kendall’s [...] Read more.

Likert-type responses are often analyzed as continuous scores, although their categories are ordered rather than truly metric. This study uses Monte Carlo simulation to examine how ordinal discretization, response perturbation, and non-Gaussian dependence affect four association estimators: Pearson correlation, Spearman correlation, Kendall’s

τ_{b}

, and latent-normal polychoric correlation. In the main latent-normal design, bivariate normal variables with known correlations were discretized into 3-, 5-, and 7-category scales using equal-probability thresholds, with

n = 300

and 200 replications per scenario. Responses were either left unchanged or modified by adjacent-category error and central-tendency shifts. A complementary Gumbel-copula extension used

n = 1000

, 1500 replications,

k = 5, \dots, 10

categories, and upper-tail dependence at Kendall’s

τ_{C} = 0.50, 0.70, 0.90

. In the unperturbed latent-normal setting, Pearson and Spearman underestimated the latent association, especially with three categories, whereas the polychoric estimator closely recovered the latent correlation. Under response perturbation, this advantage weakened; at

k = 5

,

ρ = 0.70

, and 20% adjacent error, polychoric bias increased to

- 0.072

. In the Gumbel design, the preferred estimator changed because the target changed: Kendall’s

τ_{b}

showed the smallest bias and mean squared error for copula-scale dependence, while polychoric correlation returned larger Pearson-type latent associations and overestimated

τ_{C}

. Overall, Pearson may be reasonable for many-category, symmetric observed-score questions; polychoric correlation is preferable when latent-normal assumptions are plausible; and Kendall’s

τ

is the natural choice for ordinal concordance or copula-scale dependence. Full article

(This article belongs to the Section Computational Engineering)

► Show Figures

Graphical abstract

22 pages, 3131 KB

Open AccessArticle

Reliability-Aware Gaussian Residual Counterpart Generation for Robust Multi-View Clustering with Noisy Correspondence

by Xin Liu, Lican Dai and Boyuan Zheng

Computation 2026, 14(8), 185; https://doi.org/10.3390/computation14080185 - 12 Aug 2026

Abstract

Multi-view clustering (MvC) aims to discover cluster structures by exploiting complementary information across views. Most existing MvC methods assume that same-index observations across views describe the same semantic instance. In practice, however, index-aligned observations can be semantically unrelated. This inconsistency between observed index-level [...] Read more.

Multi-view clustering (MvC) aims to discover cluster structures by exploiting complementary information across views. Most existing MvC methods assume that same-index observations across views describe the same semantic instance. In practice, however, index-aligned observations can be semantically unrelated. This inconsistency between observed index-level correspondence and underlying semantic correspondence is known as noisy correspondence (NC). Learning from such mismatched pairs imposes erroneous cross-view constraints and distorts clustering. Many existing methods only suppress unreliable pairs. This discriminative strategy avoids incorrect alignment but also excludes suspicious pairs from cross-view learning. To reuse these pairs without enforcing incorrect correspondence, we propose Reliability-Aware Gaussian Residual Counterpart Generation. Using reliability estimates derived from cross-view losses, the framework retains observed counterparts for reliable pairs and routes unreliable pairs to counterpart generation. For each unreliable pair, prototype-level semantic transport locates a matched target-view prototype. A Gaussian residual model estimated from reliable target-view samples captures variations around this prototype. The framework samples a residual from this model and adds it to the prototype center, yielding a semantically matched yet diverse counterpart. Random walk-based intra-view contrastive learning further preserves neighborhood structures. Experiments on Scene15, LandUse21, Reuters, and CCV20 achieve the best average ACC, NMI, and ARI across the evaluated NC ratios. Ablation and transfer studies further support the effectiveness of the proposed design. Full article

(This article belongs to the Special Issue Computational Methods for Multi-View Representation Learning)

► Show Figures

Figure 1

27 pages, 1578 KB

Open AccessArticle

Global Value Chain Reconfiguration and Circular Economy Transitions: A Mixed-Integer Linear Programming Model

by Hadi Zarea and Myriam Ertz

Computation 2026, 14(8), 184; https://doi.org/10.3390/computation14080184 - 12 Aug 2026

Abstract

Global value chains (GVCs) generate rising volumes of electronic waste (e-waste), of which only 22.3% is formally collected and recycled, and operationalizing circular economy principles within GVCs requires reverse logistics networks that existing optimization models only partially capture. This paper develops a multi-echelon [...] Read more.

Global value chains (GVCs) generate rising volumes of electronic waste (e-waste), of which only 22.3% is formally collected and recycled, and operationalizing circular economy principles within GVCs requires reverse logistics networks that existing optimization models only partially capture. This paper develops a multi-echelon mixed-integer linear programming (MILP) model for integrated forward–reverse e-waste network design that jointly optimizes facility locations, material flows, hybrid distribution–collection co-location, and the collection price offered to consumers. Returns follow uniformly distributed consumer reservation prices, and the resulting price-dependent return mechanism is linearized exactly through a discrete price menu, yielding a fully linear formulation without big-M constants; recyclable fractions re-enter manufacturing as secondary inputs, closing the material loop. The model is evaluated on thirty randomly generated instances of three sizes, with parameter ranges anchored to the literature, solved with the open-source HiGHS solver; the largest instances solve to within 0.1% of optimality in under two minutes. Endogenizing the collection incentive raises total profit by 4.5 to 20.1% over an exogenous-return baseline and lifts material recovery from roughly 25% to 36 to 49%, while co-location adds modest, scale-dependent value and the two mechanisms show a directionally consistent but not statistically significant tendency toward substitutability (Wilcoxon signed-rank test, p > 0.05 across all size classes). These figures characterize the calibrated synthetic instances studied here and should not be read as generalizable empirical estimates. Sensitivity analyses identify consumer responsiveness to incentives, rather than waste stream quality, as the binding determinant of achievable recovery. Full article

(This article belongs to the Section Computational Social Science)

► Show Figures

Figure 1

19 pages, 887 KB

Open AccessArticle

Numerical Modeling of a Boundary Value Problem for a Singularly Perturbed Differential Equation with Two Boundary Layers Using the Spectral-Grid Method

by Chori Begaliyevich Normurodov, Sardorbek Komil o’g’li Murodov, Muhriddin Amanturdiyevich Tilovov, Nasiba Turaxanovna Djurayeva, Mohira Majidovna Normatova and Elvira Erkin qizi Shakayeva

Computation 2026, 14(8), 183; https://doi.org/10.3390/computation14080183 - 11 Aug 2026

Abstract

This paper proposes a spectral-grid method based on Chebyshev polynomials of the first kind for the numerical solution of second-order singularly perturbed boundary value problems containing two boundary layers. The proposed method possesses several important advantages, including high numerical accuracy, computational efficiency in [...] Read more.

This paper proposes a spectral-grid method based on Chebyshev polynomials of the first kind for the numerical solution of second-order singularly perturbed boundary value problems containing two boundary layers. The proposed method possesses several important advantages, including high numerical accuracy, computational efficiency in terms of the number of arithmetic operations, reduced memory requirements, accurate localization and resolution of boundary layers, and applicability to singularly perturbed boundary value problems containing one, two, or multiple boundary layers. In the proposed approach, the computational domain is partitioned into several grid elements, and the solution on each element is approximated by a truncated series of Chebyshev polynomials. Continuity conditions for the solution and its derivatives are imposed at the interfaces between adjacent elements, resulting in a system of algebraic equations for the unknown expansion coefficients. The principal advantage of the method lies in its ability to accurately localize boundary layers by appropriately selecting the lengths of the grid elements and the degrees of the approximation polynomials. Numerical experiments for a wide range of perturbation parameters are presented in the form of tables and graphical illustrations and are compared with existing results available in the literature. The obtained results demonstrate that the proposed spectral-grid method provides highly accurate numerical solutions even for very small values of the perturbation parameter while significantly reducing the maximum absolute error. The convergence of the proposed method has been theoretically established, and its convergence rate has been analyzed. The numerical results confirm the accuracy, computational efficiency, robustness, and reliability of the proposed method for solving singularly perturbed boundary value problems. Full article

(This article belongs to the Section Computational Engineering)

► Show Figures

Figure 1

30 pages, 4989 KB

Open AccessReview

Computational Fluid Dynamics Simulations in Brain Arteriovenous Malformations: Application for the Study of Hemodynamic Alterations and Pre-Procedure Planning

by Salvatore Marrone, Carlotta Fontana, Luca Ruggeri, Carlo Giuseppe Licata, Giuseppe Emmanuele Umana, Michele Calì and Giuliana Baiamonte

Computation 2026, 14(8), 182; https://doi.org/10.3390/computation14080182 - 11 Aug 2026

Abstract

Brain arteriovenous malformations (AVMs) are complex cerebrovascular lesions characterized by abnormal direct connections between arteries and veins, resulting in altered hemodynamics and an increased risk of rupture. Following PRISMA, a comprehensive review on CFD-based modelling in brain AVMs was conducted across major scientific [...] Read more.

Brain arteriovenous malformations (AVMs) are complex cerebrovascular lesions characterized by abnormal direct connections between arteries and veins, resulting in altered hemodynamics and an increased risk of rupture. Following PRISMA, a comprehensive review on CFD-based modelling in brain AVMs was conducted across major scientific databases, including Pub-Med/MEDLINE, Scopus, Web of Science, Google Scholar, EBSCO Academic Search and IEEE Xplore, evaluating its role in hemodynamic analysis and pre-procedural planning. Twenty-three studies met the inclusion criteria and were analyzed through both qualitative synthesis and bibliometric approaches. Bibliometric analysis revealed a growing research interest in image-based modelling, 4D flow imaging and virtual embolization after 2021. Despite recent advances in study of hemodynamics simulations, the application of computational fluid dynamics (CFD) to brain AVMs remains challenging due to their complex vascular architecture and highly heterogeneous flow patterns. Nevertheless, CFD remains an important imaging modality for characterizing the lesion and guiding pre-interventional decision making. Full article

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

► Show Figures

Figure 1

31 pages, 1573 KB

Open AccessArticle

General Probabilistic Computational Framework Applied to Drake–Fermi–Brin Models of Technological Civilizations

by Matjaž Gams, Aleksander Kosanović and Julija Stopar

Computation 2026, 14(8), 181; https://doi.org/10.3390/computation14080181 - 7 Aug 2026

Abstract

Many scientific and engineering domains rely on simple models whose usefulness is limited by uncertain parameters and incompatible formulations. We present a computational framework that converts deterministic, semi-empirical, and heuristic equations into stochastic model families by using bounded parameter representations and Monte Carlo [...] Read more.

Many scientific and engineering domains rely on simple models whose usefulness is limited by uncertain parameters and incompatible formulations. We present a computational framework that converts deterministic, semi-empirical, and heuristic equations into stochastic model families by using bounded parameter representations and Monte Carlo sampling. An explicit semantic layer preserves differences in variable meaning, enabling comparison without forcing structural equivalence. The framework also supports supermodels—weighted mixtures of heterogeneous model families evaluated in a shared diagnostic space. The method is implemented as a reproducible pipeline for five structurally distinct Drake–Fermi–Brin models, with joint and marginal distribution analysis, exploratory clustering, parameter-importance diagnostics, and layered uncertainty decomposition. Results show that model structure and epistemic parameterization materially shape the induced distributions. Comparisons therefore remain conditional on the declared parameter bounds, sampling families, semantic bridges, and model-family weights. Nevertheless, ensemble integration can reveal behavior not visible within individual models. The space of possible supermodel configurations exceeds

10^{27}

, illustrating both the scale of the problem and the value of a structured probabilistic workflow. The framework provides an extensible basis for uncertainty propagation and comparison across incompatible models. Full article

(This article belongs to the Section Computational Engineering)

► Show Figures

Graphical abstract

21 pages, 2835 KB

Open AccessArticle

Evaluation and Benchmarking of a Bounded Data-Driven Correction for Compartmental Pharmacokinetic Models

by Hanan Al Lawati, Abdullah Al Lawati and Mohamed Al-Lawatia

Computation 2026, 14(8), 180; https://doi.org/10.3390/computation14080180 - 5 Aug 2026

Abstract

Background/Objectives: This study applies the previously introduced structure-preserving hybrid framework for compartmental pharmacokinetic models and extends its clinical evaluation using published clinical datasets. The framework combines a mechanistic pharmacokinetic backbone with a bounded data-driven correction. The aim was to assess predictive performance in [...] Read more.

Background/Objectives: This study applies the previously introduced structure-preserving hybrid framework for compartmental pharmacokinetic models and extends its clinical evaluation using published clinical datasets. The framework combines a mechanistic pharmacokinetic backbone with a bounded data-driven correction. The aim was to assess predictive performance in held-out subjects while keeping the main pharmacokinetic structure and avoiding a fully black-box model. Methods: This applied extension of the framework was tested in several numerical studies using published clinical pharmacokinetic datasets for polymyxin B, linezolid, and tacrolimus. For polymyxin B and linezolid, repeated subject-wise cross-validation with nested tuning was used to compare the mechanistic baseline with unconstrained and constrained hybrid corrections and a boosted-tree residual benchmark. The studies were designed to assess its behavior in a main application setting, across different drugs, under difficult fitting conditions, and under changes in correction strength and mechanistic parameters. Computational time was also assessed. Results: The results showed that the constrained correction remained close to the mechanistic baseline in the held-out analyses of polymyxin B and linezolid, but it did not significantly improve subject-level prediction. The unconstrained correction showed greater deterioration, while the boosted-tree benchmark gave mixed results and no significant subject-level improvement. The additional analyses showed that tighter correction bounds were generally selected and that the constrained hybrid still responds to changes in the mechanistic parameters in a sensible manner. Conclusions: Overall, the results suggest that the bounded data-driven correction can control the poorer performance seen with an unrestricted correction while keeping prediction close to the mechanistic baseline. It therefore provides a cautious way to combine mechanistic pharmacokinetic modeling with data-driven correction while preserving interpretability. Further external validation is still needed. Full article

(This article belongs to the Section Computational Biology)

► Show Figures

Figure 1

24 pages, 10078 KB

Open AccessArticle

Effect of Pleat Angle on Pressure Drop in H14 HEPA Filters: A Mathematical Analysis with Corrections for Real Filter Behaviour

by Raimundo Castillo, Marc Schmidt, Arisbel Cerpa-Naranjo and José O. Martínez

Computation 2026, 14(8), 179; https://doi.org/10.3390/computation14080179 - 4 Aug 2026

Abstract

The influence of pleat angle on the pressure drop of H14 HEPA filters was investigated through a mathematical model that represents the filter as a system of converging–diverging channels coupled with porous filtration media. The analysis was conducted for pleat angles ranging from [...] Read more.

The influence of pleat angle on the pressure drop of H14 HEPA filters was investigated through a mathematical model that represents the filter as a system of converging–diverging channels coupled with porous filtration media. The analysis was conducted for pleat angles ranging from 1° to 20° under a constant laminar airflow rate of 0.167 m³/s and 0.45 m/s velocity. The model combines Darcy–Forchheimer flow through the filtration media with laminar channel flow theory, enabling the total pressure drop to be expressed as a function of pleat geometry and subsequently optimised through analytical differentiation. The results show that the pressure drop contribution of the filtration media increases with the pleat angle, from 10.57 Pa at 1° to 213.49 Pa at 20°, whereas channel losses decrease sharply from 1121.71 Pa to 2.75 Pa over the same interval. The competing behaviour of these two mechanisms generates a minimum total pressure drop of 94.56 Pa at a pleat angle of approximately 6°, compared with 120 Pa for the current industrial configuration operating at 3.73°. This represents a pressure drop reduction of approximately 21.2%, implying a corresponding decrease in fan energy consumption without compromising filtration performance. The analysis further demonstrates that very small pleat angles (1–2°) are highly unfavourable, producing total pressure drops between 301 and 1132 Pa due to severe channel constriction, while for angles above 13–14°, the channel contribution becomes negligible, and the overall pressure drop is governed almost entirely by the filtration media. These findings provide quantitative design criteria for optimising HEPA, EPA, and ULPA filter geometries, highlighting pleat angle as a critical parameter for improving aerodynamic performance, flow uniformity, and energy efficiency in high-purity environments. The proposed model was further assessed using a commercially available H14 HEPA filter with 188 pleats, an effective filtration area of 10.618 m², and a nominal airflow rate of 600 m³/h, demonstrating its applicability to real industrial filter configurations. Full article

(This article belongs to the Section Computational Engineering)

► Show Figures

Graphical abstract

17 pages, 10307 KB

Open AccessArticle

Numerical Investigation Method for the Determination of Electro- and Plasma-Chemical Portions in Current Signals from Plasma Electrolytic Oxidation Processes

by Stephan Daniel Schwöbel, Frank Simchen, Thomas Mehner and Thomas Lampke

Computation 2026, 14(8), 178; https://doi.org/10.3390/computation14080178 - 4 Aug 2026

Abstract

The analysis of process signals is a key method for gaining experimental insight into the underlying layer formation mechanisms in plasma electrolytic oxidation (PEO). This is made possible by the simultaneous measurement of electrical and optical process signals with high temporal resolution. However, [...] Read more.

The analysis of process signals is a key method for gaining experimental insight into the underlying layer formation mechanisms in plasma electrolytic oxidation (PEO). This is made possible by the simultaneous measurement of electrical and optical process signals with high temporal resolution. However, according to the current state of the art, the interaction between these signals is primarily discussed in qualitative terms. Therefore, this article presents a robust methodology for analysing the current signal, which makes it possible to categorise the charge electro-chemical and plasma-chemical dominated subprocesses and to quantify their respective contributions. The evaluation is performed by taking additional process signals into account. The experimental setup for measuring process voltage, current, and photovoltage, as well as the measurement routine, are briefly described. This is followed by a detailed description of the numerical procedure. This includes the application of fundamental mathematical methods to the time-discrete measurement data, the automated selection of the pulse segment to be examined and the identification of discharge initiation to determine the interval boundaries of the electro- and plasma-chemically dominated pulse subsegments. The description of the routine is primarily intended for experimental scientists and is meant to provide them with a tool for extracting additional information from their process data. These can then be used to better understand electro-chemical side reactions and parasitic subprocesses in PEO. Full article

(This article belongs to the Section Computational Engineering)

► Show Figures

Graphical abstract

31 pages, 5051 KB

Open AccessArticle

Predicting Pornography Use Among Adolescents in Spain: Findings from Ensemble Tree Models and Explainable Machine Learning

by Jorge de Andrés-Sánchez, Ángel Belzunegui-Eraso, Inma Pastor-Gosálbez and Anna Sánchez-Aragón

Computation 2026, 14(8), 177; https://doi.org/10.3390/computation14080177 - 4 Aug 2026

Abstract

Pornography use during adolescence is a relevant issue from social, educational, and public health perspectives. As with other behaviours, its correlates may involve complex and nonlinear relationships. This study uses data from the 2023 Spanish Survey on Drug Use in Secondary Education (ESTUDES), [...] Read more.

Pornography use during adolescence is a relevant issue from social, educational, and public health perspectives. As with other behaviours, its correlates may involve complex and nonlinear relationships. This study uses data from the 2023 Spanish Survey on Drug Use in Secondary Education (ESTUDES), a major source for analysing potentially addictive behaviours among adolescents in Spain because of its large sample size (original sample: N = 42,208; complete-case analytical sample: N = 33,543). Pornography use was modelled as a binary outcome using logistic regression, XGBoost, LightGBM, and CatBoost. The models included sociodemographic characteristics, family-related factors, parental control, substance use, selected sexual behaviours, indicators of mental well-being, and addictive Internet use. The four methods showed similar predictive performance. CatBoost achieved the highest AUC and the lowest Brier score and log loss, whereas LightGBM obtained the highest accuracy, specificity, and precision. Logistic regression yielded the highest sensitivity, negative predictive value, balanced accuracy, and F1-score. SHAP analysis identified sex as the most influential predictor, followed by addictive Internet use, cannabis and alcohol use, family conflict, and selected risky sexual behaviours. The findings suggest that pornography use among adolescents forms part of a broader behavioural and psychosocial profile. Whereas sex operated mainly as a strong direct predictor, problematic Internet use showed substantial interactions with age and cannabis and alcohol use. This study demonstrates how explainable machine learning can complement conventional regression by identifying the main correlates, nonlinear patterns, and interactions associated with adolescent pornography use. Full article

(This article belongs to the Section Computational Social Science)

► Show Figures

Figure 1

27 pages, 7106 KB

Open AccessArticle

Lean GLASS: Efficient Edge-Deployable Visual Anomaly Detection with a MobileNetV2 Backbone and Learnable Feature-Stream Gating

by Muhammad Bilal

Computation 2026, 14(8), 176; https://doi.org/10.3390/computation14080176 - 4 Aug 2026

Abstract

Visual anomaly detection has achieved very high accuracy on standard benchmarks, yet state-of-the-art synthesis-based detectors such as GLASS rely on heavy backbones (e.g., WideResNet-50) that are ill-suited for deployment on resource-constrained edge platforms. This study investigates whether a lightweight convolutional neural network (CNN) [...] Read more.

Visual anomaly detection has achieved very high accuracy on standard benchmarks, yet state-of-the-art synthesis-based detectors such as GLASS rely on heavy backbones (e.g., WideResNet-50) that are ill-suited for deployment on resource-constrained edge platforms. This study investigates whether a lightweight convolutional neural network (CNN) backbone can retain such accuracy at much lower computational cost. To this end, the heavy backbone within the GLASS anomaly detection framework is replaced with the lightweight MobileNetV2 backbone. The experimental findings demonstrate that this particular choice of feature representation, namely the expanded depthwise features of MobileNetV2 rather than compressed bottleneck outputs, recovers the accuracy otherwise lost by a naive lightweight substitution. A lightweight learnable per-stream gating mechanism is further introduced, adaptively weighting feature streams on a per-category basis and yielding a measurable and consistent improvement at negligible parameter cost. On the MVTec AD benchmark, the proposed model attains a 0.992 mean image-level AUROC, matching or exceeding the ResNet-18 configuration reported by the GLASS authors under an identical training budget, while using 3.3× fewer backbone parameters and 4.7× fewer FLOPs. On a Jetson Nano, it runs 2.4× faster per frame than the ResNet-18 baseline (approximately 19 frames per second), confirming that the efficiency gains translate to usable speed on low-cost edge hardware. The findings are further corroborated on the more challenging VisA benchmark, where the proposed model matches the ResNet-18 configuration on image-level detection and improves pixel-level AUROC. Additionally, the proposed approach experimentally generalizes without modification to two further datasets from different domains, i.e., concrete crack and pharmaceutical pill inspection. A systematic negative result is additionally reported, demonstrating that the handcrafted complementary feature streams (PCA reconstruction-residual and wavelet high-frequency descriptors) do not improve accuracy, and the residual performance gap on difficult categories is attributed to the training schedule rather than to feature representation. This study therefore provides experimental evidence that careful backbone-feature selection, rather than architectural augmentation, is the key to efficient edge-deployable anomaly detection at a minimal cost in accuracy. These findings indicate that high-accuracy visual anomaly detection can be brought within reach of low-cost embedded hardware, lowering the barrier to automated inspection in smaller-scale industrial settings where a dedicated computing workstation is impractical. The source code is made publicly available to support this use. Full article

(This article belongs to the Section Computational Engineering)

► Show Figures

Figure 1

26 pages, 2365 KB

Open AccessArticle

A Number-Theoretic Generalization of the ElGamal Cryptosystem with Applications to Digital Signatures and Image Encryption

by Hayder R. Hashim

Computation 2026, 14(8), 175; https://doi.org/10.3390/computation14080175 - 3 Aug 2026

Abstract

In this paper, we propose an image encryption and digital signature scheme based on a modification of ElGamal cryptosystem over a large modulus represented by

2 p^{n}

, where p is a large prime number and

n > 2

is a dynamic [...] Read more.

In this paper, we propose an image encryption and digital signature scheme based on a modification of ElGamal cryptosystem over a large modulus represented by

2 p^{n}

, where p is a large prime number and

n > 2

is a dynamic exponent derived from a shared secret established by the communicating participants through a modification of the Diffie–Hellman Key Exchange Protocol. The proposed scheme combines a modified ElGamal framework with image encryption and digital signature mechanisms to support confidentiality, image authentication and integrity verification while preserving the standard security assumptions of discrete-logarithm-based cryptography. Empirical performance and image-statistical evaluations, including histogram analysis, entropy, correlation coefficients, NPCR and UACI, are conducted on applying the procedures of the proposed scheme on four standard images. The experimental results demonstrate the correctness and favorable empirical image-statistical behavior of the proposed scheme for grayscale image encryption and authentication under the classical DLP assumptions. Furthermore, comparisons with representative symmetric and asymmetric cryptographic schemes are provided to evaluate the performance of the proposed scheme. 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