Accelerated simulation and design optimization of elastic rod networks with a spline-based least-squares formulation

Radiation shielding performance of CeF3-Doped Li2O-PbO-GdF3-SiO2 glasses: Monte Carlo simulation MCNPX and PHY-X/PSD study.

Simultaneous optimization of chromatographic resolution and analysis time constitutes a significant analytical challenge in multicomponent pharmaceutical analysis, as resolution-driven optimization strategies may improve peak separation without providing explicit control over retention time, often resulting in unnecessarily prolonged runtimes. To address this limitation, an optimization strategy based on a Box-Behnken experimental design was implemented in conjunction with the Improved Chromatographic Response Function (ICRF), which integrates separation quality and analysis time within a single mathematical objective function. This strategy describes the development of a rapid, chemometrically optimized UPLC-PDA method for the simultaneous determination of dorzolamide hydrochloride (DH) and timolol maleate (TI) in a commercial ophthalmic preparation. A Box-Behnken experimental design and optimization approach was employed in combination with an Improved Chromatographic Response Function (ICRF), which integrates resolution, peak overlap, peak width, and runtime into a single composite objective function. This strategy enabled short runtime (or short retention time of analytes in a chromatogram) while preserving adequate peak separation. Under the optimized conditions, complete chromatographic separation was achieved within 3 min using a BEH C18 column and a mobile phase consisting of acetonitrile and 4 × 10-4 M CCl3COOH (60:40, v/v) at a flow rate of 0.32 mL/min with detection at 275 nm. The method demonstrated excellent linearity over the range of 5.0-40.0 µg/mL (r > 0.999), with limits of detection of 0.51 µg/mL for DH and 0.61 µg/mL for TI. Mean recoveries were 99.9% and 99.5% for DH and TI, respectively, with satisfactory precision and robustness. The proposed ICRF-assisted optimization approach provided high-resolution separation within minimal runtime and was successfully applied to the routine quality-control analysis of a commercial ophthalmic formulation. The study demonstrates the effectiveness of composite response-based chemometric optimization in enhancing analytical efficiency in pharmaceutical drug analysis.

Daily order planning is the core of dynamic decision-making that ensures the reliability and cost-effectiveness of manufacturing system delivery. However, order planning faces significant challenges due to the interplay of load sequence-dependent degradation, human errors, and imperfect inspection. The interplay of these factors introduces significant uncertainty, often leading to overestimated system performance and suboptimal planning outcomes. Thus, a comprehensive order planning framework is developed in this paper to optimise order scheduling. First, an Extended Stochastic Flow Manufacturing Network (ESFMN) model is introduced to characterise the dynamic interactions among processing machines, inspection machines, buffers, operators, and heterogeneous feedstocks (qualified and unqualified feedstocks) in manufacturing systems with load sequence-dependent degradation. Second, a semi-Markov model is employed to evaluate system mission performance that combines human errors and imperfect inspection. Third, a Clustering-Enhanced NSGA-II (CNSGA-II) algorithm is developed to tackle the order planning problem through the simultaneous optimisation of mission performance and cost. Comparative experimental results demonstrate that proposed order planning framework outperforms several advanced algorithms, including Memetic-NSGA-II, Reinforcement Learning-NSGA-II, Greedy algorithm-NSGA-II, and NSGA-II. Through a case study of a servo valve spool manufacturing system, the framework's effectiveness in practical applications is validated, confirming its ability to achieve reliable and cost-effective order planning.

The design of proteins that bind to small molecules has been challenging because it requires simultaneous optimization of the protein sequence, protein structure and ligand conformation1-7. Current deep-learning algorithms have struggled to navigate this landscape, precluding the zero-shot design of binders. Here we show that by combining two neural networks in an iterative design algorithm, small-molecule binding proteins can be created from scratch with high accuracy. We trained a graph neural network-ligand-aware sequence engineering message-passing neural network (LASErMPNN)-to designcompatible protein sequences for an inputprotein backbone and docked ligand. We paired LASErMPNN with a structure predictor that models a three-dimensional protein-ligand complex for an input protein sequence and ligand identity. The closed-loop iteration of these reciprocal networks optimized sequence-structure-ligand compatibility, and outperformed a comparable design loop using a physics-based energy function. We used our strategy, termed neural iterative selection-expansion (NISE),to design proteins that, using different folds, specifically bind to two chemically distinct small-molecule drugs, exatecan and apixaban, with success rates of 100% and 83%, respectively. The tightest NISE binders had nanomolar-to-picomolar affinities, surpassing those of the next-leading method by 70-fold for exatecan and nearly 10,000-fold for apixaban. LASErMPNN then suggested two amino-acid substitutions that improved the affinity of thetightestexatecan binder by 100-fold without any experimental input. The optimized binder protected the labile lactone ring of exatecan from hydrolysis for days. Our work describes a general recipe for using neural networks to automate the design of small-molecule binding proteins for applications in drug delivery, sensing and catalysis.

Background/Objectives: In gas chromatography-mass spectrometry (GC-MS) library-based compound identification, spectrum preprocessing and associated tuning parameters critically influence identification performance. These parameters are conventionally optimized using grid search, which requires predefined parameter spaces and becomes computationally inefficient as dimensionality increases, often failing to identify optimal values because of discretization. Differential evolution (DE), a population-based metaheuristic optimization algorithm, provides a flexible alternative through efficient global exploration of the parameter space. This study compared the performance of DE and grid search for optimizing compound identification. Methods: Cosine similarity was applied to the NIST GC-MS library. DE was used to maximize either cross-validated accuracy or mean reciprocal rank (MRR). Results were compared with those from a grid search over five equally spaced parameter values. Identification performance was evaluated using accuracy, MRR, and area under the receiver operating characteristic curve (AUC). Results: When all four parameters were optimized simultaneously, DE achieved slightly higher cross-validated accuracy and MRR than grid search, although the absolute differences were modest. More pronounced differences were observed in specific unidimensional tuning scenarios, particularly for the intensity weight factor. Simultaneous multidimensional parameter optimization yielded better performance than isolated parameter tuning. Conclusions: Grid search may be computationally advantageous when the parameter space is known and limited, whereas DE provides a more flexible approach for unknown or high-dimensional search spaces. Overall, DE achieved comparable identification performance to grid search, with modest improvements observed in some optimization settings. A command line Julia-based tool, MSTune, was developed for spectrum preprocessing parameter optimization and is publicly available on GitHub.

The rapid transition from 5G to 6G wireless technologies has led to a substantial increase in electromagnetic interference (EMI), posing significant challenges to device reliability, signal integrity, and human health. There is an urgent need for lightweight, flexible, and efficient absorption-dominant EMI shielding materials. In this study, we propose a rationally designed coaxial electrospinning strategy to fabricate hierarchical MXene/reduced graphene oxide/polyacrylonitrile core-shell nanofibers. This design integrates a conductive, MXene-rich core surrounded by a reduced graphene oxide/polyacrylonitrile outer shell, encouraging simultaneous optimization of electrical conductivity, impedance matching, and polarization loss to achieve superior EMI shielding performance and thermal blocking. Interestingly, when the reduced graphene oxide to MXene ratio was optimized to be 1:2, the resultant composite nanofiber revealed an EMI shielding effectiveness of 74.38 dB at X-band, a shielding efficiency of 99.99999635%, a high electrical conductivity (169.81 S·cm-1), and an outstanding specific shielding effectiveness of 2.54 × 104 dB·cm2·g-1. Additionally, the sample exhibited effective thermal insulation, with the surface temperature rising by only 41 °C after 1 min of exposure to a 75 °C heat source. These findings highlight the synergy between the high electrical conductivity of MXene, which enables strong internal reflection and ohmic loss, and the interfacial polarization of reduced graphene oxide, which fosters multiple internal reflections that enhance absorption. This optimized, absorption-dominated mechanism provides valuable guidance for designing next-generation lightweight and flexible EMI shielding materials for applications in aerospace, 5G/6G communications, and advanced electronic devices.

This study presents a sustainable machining optimization framework for waste-derived Al-Cu-Mg alloys reinforced with zirconium dioxide (ZrO₂). A hybrid Fuzzy Analytic Hierarchy Process-Weighted Aggregated Sum Product Assessment framework was applied to optimize turning parameters, including cutting speed of 11-42m/min, feed rate of 0.05-0.15mm/rev, and depth of cut of 0.5-1.0mm. Multiple machining responses were evaluated, including surface roughness, material removal rate, tool wear, power consumption, dimensional accuracy, specific cutting energy, and chip volume. Three die-cast alloys were fabricated from recovered metallic waste streams: Sample A with 91% Al, 4.5% Cu, 1.8% Mg, and 2% ZrO₂; Sample B with 87.3% Al, 4.9% Cu, 1.8% Mg, and 6% ZrO₂; and Sample C with 82.5% Al, 5.5% Cu, 2% Mg, and 10% ZrO₂. The experimental trials were planned using a D-optimal Response Surface Methodology design, and the machining alternatives were ranked using the hybrid Fuzzy AHP-WASPAS model. The final ranking identified Alternative 8, corresponding to Sample B with 6 wt.% ZrO₂, as the overall optimum condition because it provided the best compromise among surface quality, productivity, tool wear, energy demand, dimensional accuracy, and material utilization. Compared with the unoptimized baseline condition, the optimum condition improved surface finish by 25%, increased material removal rate by 30%, and reduced tool wear by 15%. The superior performance of Sample B is attributed to balanced ZrO₂-induced grain refinement, improved reinforcement dispersion, reduced agglomeration, and stable tool-workpiece interaction. Although Run 3 showed the best energy-specific response, it was not the overall optimum because the final decision was based on simultaneous multi-response optimization. The proposed framework demonstrates a practical route for converting waste-derived aluminum alloy streams into value-added machinable composites while supporting energy-efficient and circular manufacturing aligned with SDG 9 and SDG 12.

Radon (222Rn) and its alpha-emitting progeny are major contributors to natural background radiation and a leading cause of lung cancer in non-smokers. This work presents the design and simulation of a portable alpha detector based on cadmium telluride (CdTe) technology for environmental radon monitoring. The system combines a compact, CAD-modeled housing with an optimized radon diffusion inlet and a low-noise CSA-CR-RC readout chain. Coupled Geant4 particle transport and SPICE electronics simulations were performed under realistic conditions (50-1000Bq/m³, 0-40°C, 20-90% RH). Results show a sensitivity of 0.15 cps/(Bq/m³), energy resolution better than 2% FWHM at 5.49MeV, and < 15min response for significant concentration changes. In contrast to previous CdTe-based alpha detection studies, which primarily focused on detector characterization under laboratory conditions, the present work introduces an instrument-level design framework that couples Geant4 particle transport, SPICE electronics modeling, and CAD-based housing optimization specifically for portable radon monitoring. This integrated approach enables simultaneous optimization of sensitivity, response time, and environmental robustness within a compact, battery-operated platform. The resulting design not only offers portability, robustness, and high-resolution alpha spectroscopy, but also establishes a transferable methodology for developing next-generation portable radon monitors for residential, occupational, and outdoor environments.

Gout and hyperuricemia, linked to purine metabolism abnormalities or impaired uric acid excretion, are rising with lifestyle changes. Effective self-management and health literacy are crucial for gout management. While large language models (LLMs) show promise in enhancing health management, their potential on gout patient education remains underexplored. This study aimed to evaluate the accuracy and readability of responses generated by three LLMs, including DeepSeek-V3, DeepSeek-R1, and GPT-5, to questions based on the gout and hyperuricemia guidelines published by the American College of Rheumatology (ACR). Based on the ACR gout guidelines, a set of 42 questions was curated and submitted to three LLMs. Their responses were independently rated on a 5-point Likert scale by three expert gout and hyperuricemia specialists against the guidelines. Accuracy was rated as either an average score of ≥ 4 (low threshold) or 5 (high threshold). The readability of the responses was assessed by Microsoft Word, which provided metrics including the word count, character count, Flesch Reading Ease (FRE) score, Flesch-Kincaid Grade Level (FKGL), and Automated Readability Index (ARI) were calculated. Our findings reveal that response accuracy was significantly higher for GPT-5 compared to DeepSeek-V3 (P < 0.001), with no significant difference between GPT-5 and DeepSeek-R1 (P > 0.05). In terms of readability, GPT-5 produced the most complex responses (FKGL: 12.89 ± 2.22, ARI: 14.87 ± 2.40), while DeepSeek-R1 generated the longest outputs. LLMs show potential in generating responses that are consistent with clinical guidelines for gout management. The deployment of LLMs in gout patient education and clinical decision support necessitates the simultaneous optimization of both accuracy and readability. Key Points • This study is the first to systematically evaluate the agreement with ACR gout guidelines and readability of three state-of-the-art LLMs (DeepSeek-V3, DeepSeek-R1, GPT-5) in gout management, filling the gap of LLM benchmarking for gout patient education. • Response accuracy was significantly higher for GPT-5 compared to DeepSeek-V3, with no significant difference between GPT-5 and DeepSeek-R1. However, GPT-5 produced texts with the lowest readability. • The study innovatively combined expert-rated accuracy (5-point Likert scale by gout and hyperuricemia specialists) and objective readability metrics (FRE, FKGL, ARI), providing a comprehensive framework for assessing LLM utility in chronic disease self-management. • Findings confirm LLMs' potential for gout patient education but emphasize the need for simultaneous optimization of medical accuracy and health literacy, guiding future LLM refinement for clinical application.

Optimizing biochar rate and particle size for enhanced wheat yield and water productivity using response surface methodology

This study develops an innovative optimization framework for designing hybrid variable-stiffness composite structures featuring spatially varying fiber orientations with inter-regional blending constraints, which enables significant weight reduction in aerospace applications. The methodology addresses the complex design challenge involving mixed discrete variables (the numbers of plies and layer thicknesses) and continuous variables (fiber orientation angles), while ensuring manufacturing feasibility through strictly enforced fiber path continuity requirements. To realize simultaneous optimization of such mixed design variables, a ground laminate concept with redundant plies guides the optimization process and establishes the initial design space. A multi-fidelity surrogate modeling technique incorporating exponent-based adaptive correction enhances computational efficiency, and a hybrid optimization strategy combines genetic algorithms for global exploration with sequential quadratic programming for local refinement. A novel layer-sharing and mutation scheme, integrated between optimization phases, guarantees blending constraint compliance throughout the design process. Two representative numerical examples, i.e., a two-patch plate and a T-shaped composite structure, demonstrate the framework’s effectiveness in optimizing the composite layup while satisfying all manufacturing constraints. The developed methodology provides a systematic and practical solution for designing hybrid variable-stiffness composite structures, showing superior performance compared to conventional approaches.

Thermoelectric materials enable direct thermal-to-electrical energy conversion for waste heat recovery, yet their figure of merit ZT is constrained by intrinsic transport trade-offs. Herein, we target CuCrTi2Se6, a new narrow-bandgap semiconductor and quaternary chalcogenide derived from two-dimensional transition-metal dichalcogenides (TMDs), and tune its coordination via Ag doping, driving its peak ZT to 1.0 at 773 K and single-leg efficiency to 6.3% at ΔT = 500 K. The larger atomic mass and size of Ag generate local stress fields that drive Cu migration from octahedral to tetrahedral interlayer sites, shortening Cu-Se bonds, enhancing bond covalency, and increasing carrier mobility from 28 to 35 cm2·V-1·s-1. Concurrently, Ag doping reduces Cu-vacancy formation energy, increasing hole concentration while elevating valence band degeneracy to enhance the Seebeck coefficient. On the phononic side, weak Ag-Se bonds induce lattice softening, and strong point-defect scattering from Ag-Cu mass/strain fluctuations synergistically reduces lattice thermal conductivity from 0.46 to 0.31 W·m-1·K-1. Benefiting from the simultaneous optimization of electronic and phononic transport, Cu0.95Ag0.05CrTi2Se6 achieves a peak ZT nearly 70% higher than pristine CuCrTi2Se6. This work establishes coordination environment regulation as an effective strategy for tuning chemical bonding and achieving coupled optimization of thermoelectric transport in layered materials.

A new integrated extraction strategy was established by combining natural solvent systems (NADES), microwave-assisted extraction, and machine learning techniques for optimizing the extraction of bioactive compounds from fermented watermelon rind. The results showed that solid-state fermentation significantly improved extraction rates, compared to non-fermented controls. The fermentation pretreatment comparison was conducted under identical extraction conditions, with non-fermented watermelon rind serving as the internal control. Machine learning models optimized using Bayesian optimization were developed to describe and predict three important response variables: TPC, TFC, and antioxidant activity, as functions of four extraction variables (microwave power, temperature, time, and solid-liquid ratio). The ensemble models showed outstanding predictive capabilities with test R² values of 0.9147, 0.9088, and 0.9252 for TFC, TPC, and DPPH activity, respectively, with very low overfitting (ΔR² < 0.06). Importance analysis of the features showed temperature as the most important parameter in the extraction of bioactive compounds (importance: 0.842-0.885), followed by solid-liquid ratio. Simultaneous optimization using the validated models showed the optimal extraction conditions to be 62.5°C, 27.7min, 300W, and 30mg/mL, predicting values of 1.656mg CE/g (TFC), 19.80mg GAE/g (TPC), and 71.27% (DPPH activity). Solid-state fermentation enhanced extraction yields, increasing total phenolics (16.9 to 19.1mg GAE/g), flavonoids (0.61 to 0.74mg CE/g), and DPPH activity (66% to 75%). The developed ensemble models showed high accuracy (R² = 0.91-0.93; RMSE = 0.0812mg/g, 0.51mg/g, and 0.74%). Experimental validation results have confirmed the high accuracy of the model with relative errors of less than 3% for all responses. The combination of fermentation pretreatment, green extraction, and machine learning indicates promise for sustainable valorization of agricultural waste resources as sources of bioactive compounds for nutraceutical applications.

This study predicts octanol/water partition and distributioncoefficients(logP and logD) for 18 triazine macrocycles, which serve as indicatorsof how drug-like compounds partition between lipid and aqueous environments.Using DFT-computed solvation free energies (ωB97X-D/6-311++G­(2d,p)/6-31G­(d)/SMD),we compare predicted values with experimentally measured logD. Ourapproach models solvent-dependent conformations and ensemble distributionsfor a family of 24-atom triazine macrocycles exhibiting well-definedhinge motions. Because experimental determination of these propertiesis time-consuming and costly, reliable computational predictions areessential. Simple additive models (AlogP) fail to accurately capturethe behavior of these macrocycles. In contrast, their well-definedstructures and conformational flexibility make them promising candidatesfor therapeutic development. Incorporating intramolecular hydrogenbonding (IMHB) and environment-dependent conformational changes intocomputational models enables simultaneous optimization of solubilityand permeability. Our quantum mechanical method, combined with a linearcorrection, predicts logP and logD for macrocycle A with root-mean-squaredeviations of 0.9 and 0.8 log units, respectively, slightly outperformingAlogP. Careful consideration of conformational dynamics, protonationstates, polarity, and IMHB significantly improves prediction accuracy.However, AlogP remains an effective high-throughput screening metricdue to its minimal computational cost.

Abstract The performance of cable-stayed bridges is drastically conditioned by the design of their cable-supporting system. Finding the optimal combination of cross-sectional areas and prestressing forces for a given cable-supporting system configuration is crucial to ensure safety and cost-effectiveness. Despite the clear benefits of structural optimization techniques for this goal, their implementation in the industry and the research realm when addressing natural hazards is quite limited due to their high computational demands and implementation challenges. This study proposes and compares four different optimization strategies aiming at reducing the computational burden, laying the foundation for their effective formulation and implementation. The first strategy is a traditional nested approach, which considers all cable cross-section areas as design variables and calculates the prestress values in each evaluation by solving a system of linear equations to achieve zero dead load displacement. The second method, simultaneous optimization, involves both area and prestress values as design variables, while solving the zero-displacement constraint by imposing dead load displacement constraints. Moreover, two new sequential optimization methods divide the optimization into two stages: area optimization and prestress values calculation. This second phase is solved by a standalone optimization process, or by solving the zero-displacement system of equations. Comparisons conducted using three application examples revealed the advantages and limitations of each method, demonstrating the strong reliability of simultaneous optimization and the potential of the sequential method to significantly alleviate the computational burden in multivariate design problems.

This paper proposes an innovative, hybrid approach to task scheduling and virtual machine (VM) placement in fog computing environments that aims to optimize energy efficiency and task completion for Internet of Things (IoT) applications. The proposed method combines Gorilla Troops Optimizer (GTO), a bio-inspired metaheuristic, with a resource-aware virtual machine placement strategy that allows for simultaneous optimization of task scheduling and virtual machine allocation. This integrated approach addresses key challenges in the allocation of IoT tasks by considering multiple objectives, such as latency, power consumption, and load balancing. A dynamic exploration-exploitation strategy and innovative fitness functionality have been employed to efficiently map tasks to fog nodes while minimizing task failure and suspension periods. Extensive simulations performed with iFogSim2 demonstrate the effectiveness of the proposed method and have achieved significant improvements over existing algorithms. The proposed approach is 18% better than ant colony optimization (ACO), 15% better than improved multi-objective differential evolution (IMODE), and 13% better than genetic and simulated annealing (GASA). These results underscore the effectiveness of the hybrid method in optimizing task scheduling and VM placement for dynamic and latency-sensitive applications in IoT. This work provides a scalable solution for fog computing systems that significantly improves service quality by optimizing resource consumption and reducing energy consumption, providing a promising approach to real-world IoT environments.

Steric-hindrance engineering-induced formation of closed pores enhances sodium storage capacity in starch-derived hard carbon.

Experimentation is inherently difficult because most methods require substantial refinement, calibration, and validation before high-quality, reliable data can be collected. In most cases, experimental outcomes are impacted by multiple variables, thus requiring their simultaneous optimization for single and multi-objective targets. Traditional experimental approaches rely on trial-and-error methods guided by rational decision making, but these become increasingly inefficient and ineffective as complex interactions between inputs limit our ability to capture underlying trends using conventional statistical approaches. Machine learning and active learning (ML/AL) combined with automation represents an approach that can bolster future laboratory productivity. However, a steep initial learning curve and high costs of instrumentation pose substantial barriers to adoption. To democratize access, we herein comprehensively cover both the computational skills and hardware implementation necessary for self-driven experimental workflows. The accompanying open-source, low-cost liquid handling platforms offer practical templates for researchers adopting self-driving lab (SDL) methodologies. Complete tutorials and build guides are provided at https://gormleylab.github.io/SDLGuide.

For compact spectrometers, the enduring technical challenge lies in the simultaneous optimization of high optical performance and miniaturization. Here we report a solid immersion grating microspectrometer with wedge prism correction (SIG-W-µSPEC), which realizes both high spectral resolution and a highly compact structure across a broad spectral range. The SIG-W-µSPEC integrates a solid immersion grating (SIG), planar reflectors, an optical wedge, a convex lens, a silicon microslit, and a CMOS linear array sensor into a single optical module. The SIG endows the system with high angular dispersion but inevitably induces focal plane tilt; the optical wedge not only compensates for this tilt with high precision but also effectively compresses the optical path length. Following high-precision encapsulation, the SIG-W-µSPEC achieves an average spectral resolution of 4.4 nm in the 600-1000 nm wavelength range, with an overall volume of merely 0.63 cm3. We applied this device to collect the reflectance spectra of fruits for the prediction of soluble solids content (SSC), yielding a high correlation coefficient (R2=0.941). This ultra-compact spectrometer demonstrates broad application prospects in diverse fields, including agricultural quality monitoring, biomedical sensing, and real-time non-invasive on-site detection.