Search Results (254)

The transition towards sustainable manufacturing necessitates complex optimization that integrates economic goals with environmental factors, such as energy consumption and greenhouse gas emissions. This research addresses the critical challenge of optimizing the Incoming Quality Control (IQC) policy for raw material batches. The primary objective is formulated as a multi-criteria control problem that jointly minimizes the weekly final product cost, carbon footprint, and energy consumption. To handle sequential decision making under uncertainty, we adopt a scalarized reinforcement learning (RL) reward that combines these objectives into a single value function and explores different trade-offs through alternative weight configurations. To effectively handle the uncertainty in incoming quality and the sequential decision making required for dynamic control, the optimization problem is modeled as a Bayesian Adaptive Markov Decision Process (BAMDP). To maintain computational tractability despite the continuous belief space inherent in the BAMDP formulation, we employ a Deep Q-Network (DQN) architecture acting as an approximate dynamic programming solver. The Bayesian framework represents model uncertainty explicitly, updates beliefs as new inspection evidence becomes available, and allows prior domain knowledge on supplier quality to be incorporated into the learning process. The BAMDP formulation is used to learn a set of adaptive inspection policies that adjust the IQC strategy over time to achieve conflicting goals: reducing inspection costs while maintaining standard quality, minimizing energy consumption, and lowering CO₂-equivalent emissions. The goal is to find robust policies that balance these trade-offs under different quality and demand conditions. This methodology aligns with the principles of Industry 5.0 by leveraging advanced artificial intelligence (AI) methods, such as reinforcement learning (RL), coupled with a stochastic simulation of the production system, based on a geometric/physical model of the component’s tolerance chains, to support decision-makers in designing and assessing sustainable IQC strategies. Comparative simulations on the case study, including a benchmark against ISO 2859-1 sampling plans, confirm that this dynamic and risk-aware optimization paradigm can reduce overall cost, energy use, and environmental impact across various quality conditions, while preserving outgoing quality. Full article

This study proposes a novel and improved numerical approximation of the simulation of Gaussian process autoregressive models. As a Bayesian nonparametric regression method, Gaussian process models offer the unique advantage of providing closed-form uncertainty quantification. When Gaussian process models are used for autoregressive models, the validation procedure requires the model’s simulation or multi-step-ahead prediction. However, simulating dynamical Gaussian process models is complex due to the intractable propagation of uncertain inputs through the nonlinear model. Numerical approximation, namely Monte Carlo simulation, is one of the most frequent options for simulating dynamical models based on Gaussian processes. The computational burden of Monte Carlo simulation algorithms increases cubically with data size, representing a challenge. This paper introduces a unified simulation framework invariant to sparse and variational approximations to obtain a static sample from the pseudo-point posterior. Furthermore, we propose an innovative method for simulating Gaussian process dynamical models. A single parameter is proposed to regulate the trade-off between computational complexity and algorithmic accuracy. This innovation demonstrates the potential to replace the conditionally independent Monte Carlo method with no additional computational burden, thereby enhancing estimates of latent responses. The proposed simulation method is demonstrated using two synthetic examples and a realistic case study. Full article

Geopolitical risks—including export controls, entity listings, and end-use restrictions—have become a major source of disruptions in semiconductor supply chains. The impact of such disruptions depends not only on the policy trigger itself but also on the vulnerability of cross-regional partnerships between supply chain partners. Specifically, under the same policy regime, firms with weak partnerships suffer far greater disruption than those with strong partnerships. Apart from risk propagation, this vulnerability also propagates through the supply chain: when an upstream supply channel has weak partnerships, its downstream stages also become more exposed to disruptions. We call this phenomenon vulnerability propagation. Existing Bayesian Network (BN) frameworks portray risk propagation through fixed parameters that do not reflect partnership vulnerability and cannot capture vulnerability propagation. To fill this gap, we propose a Dependency-Robust Bayesian Network (DeRBN) that conditions risk propagation parameters on the partnership vulnerability. A robust worst-case oriented evaluation method is developed to assess the disruption risk under data scarcity. Computational experiments on a typical semiconductor supply chain network show that (i) moving from all-strong to all-weak partnerships increases the worst-case risk by approximately 24%, (ii) the dependency-induced risk amplification is unevenly distributed across supply channels, with the most influential channel contributing approximately 2.2 times the marginal risk of the least influential one, and (iii) the relative ranking of vulnerability profiles remains perfectly stable under varying levels of data uncertainty. These results suggest that DeRBN has the potential to serve not only as a risk assessment tool but also as a diagnostic instrument for identifying and prioritizing the most vulnerable supply channels for targeted risk mitigation. Full article

This paper addresses the estimation of the multi-component stress–strength reliability when both the strength variables and the stress variable follow the one-parameter Garhy distribution. Data are assumed to arise from a unified hybrid censoring scheme, which generalizes both Type-I and Type-II hybrid censoring. A closed-form expression for the reliability parameter $R_{m,k}=P(\mathrm{at}\mathrm{least}m\mathrm{of}(X_1,\dots ,X_k)>Y)$ is derived, enabling efficient computation. Three estimation procedures are developed: maximum likelihood estimation (MLE), Bayesian inference using Markov chain Monte Carlo (MCMC) with non-informative priors, and the Tierney–Kadane Laplace-type approximation for posterior moments. For each method, we provide complete mathematical derivations, including the likelihood function under unified hybrid censoring, the posterior conditionals, and the asymptotic distribution of the reliability via the Delta method. Furthermore, Bayesian estimation is extended to asymmetric loss functions, and posterior propriety is formally proven. To check the suitability of the proposed methods, a real data application on generator failure times in power systems is presented. Full article

Temples are tourist attractions that represent the history and culture of Thailand, especially in Chiang Mai province—a city with a rich history that has become a prominent destination attracting visitors from around the world. Many temples remain undiscovered yet are ready for tourists to visit; however, due to unfamiliarity, tourists tend to visit only the well-known temples, as other visitors do, missing great opportunities to engage with new cultural heritage tourism experiences. To address this issue, we propose a Hybrid Three-Stage Route Planning Recommendation (HTS-RPR), a novel method for tourist route planning that delivers recommended routes based on tourists’ preferred constraints. This model contains three-stage route recommendations providing an optimal single-day route with mandatory and recommended points of interest (POIs) through a metaheuristic integrating Mixed Integer Programming (MIP), heuristic-based POI recommendation filtering, and Genetic Algorithm route optimization with Bayesian reward and peak-time awareness, ensuring that users can effectively travel cultural routes with high popularity and satisfaction while avoiding attractions during periods of high traffic. To validate the efficacy of the proposed model, experiments with three baseline methods were conducted. The results demonstrate that HTS-RPR achieves the best fitness score in 55 out of 60 scenarios and the best reward in 54 out of 60 scenarios, with a median fitness score 28.34% and 103.67% higher than the Genetic Algorithm and Multi-Start Simulated Annealing baselines, respectively, and a median total reward exceeding all three baselines by up to 40.74%. Although HTS-RPR’s median execution time is approximately 2.6 times that of the Genetic Algorithm, it remains 84.5% faster than the Multi-Start Simulated Annealing baseline, offering a favorable trade-off between solution quality and computational cost. Moreover, the framework’s pluggable reward function enables destination managers to configure recommendation priorities, including the promotion of undiscovered tourist attractions, while the peak-time-aware optimization mitigates congestion at specific POIs. Full article

The success of large neural networks trained for sequential prediction via log-loss minimization over massive and diverse datasets has sparked debate regarding the fundamental limits of this paradigm. While these models are not explicitly programmed to perform planning and search, their behavior increasingly resembles complex reasoning and adaptive problem-solving. This paper reviews a series of theoretical and empirical works, aiming to bridge the gap between the practical success of LLMs and formal theories of computation and intelligence—that is, algorithmic information theory and Universal Artificial Intelligence. Grounded in the framework of memory-based meta-learning, the main argument is that training sequence models to predict the next token across diverse tasks implicitly meta-trains them to perform algorithmic compression, thereby performing (amortized) Bayesian inference over the task in-context. Consequently, when pretrained on a sufficiently rich data distribution, the resulting neural networks behave as if compressing by inferring the generative algorithm producing the observed data. We discuss recent theoretical and empirical evidence demonstrating that this approach can approximate Solomonoff induction in the theoretical limit, match exact Bayesian inference on complex sources in practice, achieve strong compression on out-of-distribution data, and synthesize complex in-context algorithms like chessboard evaluations. As models become more capable and general, the theoretical understanding through the lens of algorithmic information theory, including hard theoretical limits and how far practical models are from them, becomes increasingly relevant. We thus conclude our paper by outlining a number of open research questions to further bridge the gap from well-understood theory to modern machine learning practice. Full article

This paper investigates the estimation of the stress–strength reliability parameter $R=P(Y<X)$ when both stress and strength follow independent Garhy distributions under progressive Type-II censoring schemes. A closed-form expression for $\mathit{R}$ is explicitly derived, enabling effective and precise calculation without numerical integration. The Garhy distribution, a flexible one-parameter lifetime model with an increasing hazard function, is confirmed by full-scale goodness-of-fit diagnostics. A Bayesian estimation model is trained on non-informative priors (normal and extended Jeffreys priors) under squared error loss. The posterior expectations are analytically intractable; we adopt two complementary methods of computation: (i) Markov Chain Monte Carlo (MCMC) using the Metropolis–Hastings algorithm and (ii) the Tierney–Kadane (TK) approximation, which provides extremely precise analytical estimates with significantly reduced computational burden. Monte Carlo simulations are large-scale and compare the proposed estimators under different censoring schemes, sample sizes, and parameter configurations in terms of bias and mean squared error (MSE). The methodology is further applied to a real medical dataset comprising kidney dialysis patient survival times, demonstrating its practical relevance in clinical reliability assessment. Results consistently indicate that Bayesian methods, particularly with the extended Jeffreys prior, outperform classical MLEs in terms of stability and accuracy, especially under heavy censoring. Moreover, the TK approximation yields estimates virtually identical to MCMC while requiring only a fraction of the computational effort. We further extend the TK framework to approximate the posterior variance of R and the expected log-likelihood, providing a fully analytical alternative to MCMC for comprehensive Bayesian inference. Full article

A three-year (January 2020–December 2022) daily dataset of 16 polycyclic aromatic hydrocarbons (PAHs) collected in parallel with PM_2.5 and a suite of meteorological variables at a coastal Mediterranean urban site in southern Italy (Pomigliano d’Arco, Campania) is presented and analysed. Raw PAH time series were decomposed into a long-term trend component (LT), a seasonal component (ST), and a residual component (RT) using an iterative missing-value-robust Kolmogorov–Zurbenko (KZ) moving-average filter. Spearman rank correlations between PAH concentrations and four meteorological predictors (mean temperature, relative humidity, mean wind speed, and maximum wind speed) were computed for each congener. Diagnostic molecular ratios—Fla/(Fla + Pyr), BaP/BghiP, Indeno[1,2,3-cd]pyrene/(IcdP + BghiP), and BaA/(BaA + Chr)—were evaluated seasonally and interpreted jointly with an information-theoretic Bayesian mixture modelling procedure (SNOB/MML) and with the documented susceptibility of some PAH ratios, especially BaP-containing ratios, to atmospheric ageing, phase repartitioning and summer photodegradation. Total PAH concentrations (sum of 16 congeners) ranged from <1 ng m⁻³ in summer to 46 ng m⁻³ during winter high-pollution episodes, with BaP peaking at ≈6.7 ng m⁻³. Because BaP was measured in the PM_2.5 fraction, comparisons with the EU annual target value of 1 ng m⁻³ established for PM₁₀-bound BaP are treated as indicative context only, not as formal compliance statements. Pronounced seasonal variability was driven primarily by residential heating emissions, and the incremental lifetime cancer risk (ILCR) for inhalation exposure reached $1.03\times {10}^{-4}$ (95% CI: $0.88$–$1.20\times {10}^{-4}$) during the heating season under a continuous outdoor-exposure worst-case scenario. The absolute ILCR magnitude is conditional on the selected TEF scheme and on the adopted BaP unit-risk coefficient; under an additional indoor-dominated scenario (16 h day⁻¹, infiltration factor 0.6), the corresponding risk remained above the conventional ${10}^{-6}$ benchmark. An anomalous near-background PAH signal during spring 2020 is attributed to the COVID-19 national lockdown, which reduced total PAH concentrations by approximately 85% relative to the seasonal component predicted by the iterative moving-average filter for the same calendar window. Source apportionment via diagnostic ratios identifies residential/biomass combustion as the dominant cold-season source and vehicular emissions as the prevailing warm-season source. These results provide a novel characterisation of PAH pollution dynamics in the undersampled southern Mediterranean and provide evidence to support targeted abatement policies. Full article

An adaptive progressive first-failure censoring scheme is used to enhance the efficiency of statistical analyses and minimize test time in life-testing experiments. This paper focuses on statistical inferences for the unknown parameters, survival, and hazard rate functions of the Burr XII distribution under this censoring scheme. Since the maximum likelihood estimates for the model parameters and reliability characteristics cannot be obtained explicitly, the Newton–Raphson method is employed for numerical derivation. The delta method is used to determine the variances of reliability characteristics and is applied to construct confidence intervals. Bayesian estimates of the unknown parameters and reliability characteristics are derived under the squared error and linear exponential loss functions. As these estimates are not explicitly obtainable, the Lindley and Markov chain Monte Carlo methods are used as approximation techniques. Additionally, asymptotic confidence intervals and highest posterior density credible intervals are developed for the parameters and reliability characteristics. A Monte Carlo simulation is performed to evaluate the proposed estimators, and the methodology is validated through a real dataset analysis on arthritic patients. Full article

Blade Tip Timing (BTT) is an essential non-contact technique for monitoring vibrations in rotating machinery, but its practical accuracy is often degraded by noise, undersampling, and spectral leakage. This paper proposes a multi-stage robust Bayesian high-resolution identification framework that systematically addresses these challenges. A recursive digital algorithm based on Kalman filtering estimates the rotational speed without requiring once-per-revolution probes, effectively suppressing sensor noise. An attention-enhanced dynamic convolutional autoencoder then generates channel-specific window functions to minimize spectral leakage. The core identification algorithm extracts phases via all-phase FFT and employs sub-bin interpolation to overcome the resolution limitation of conventional FFT. A Tukey-biweight-based robust aggregation strategy is used to suppress the influence of abnormal or unequal-quality sensor channels during multi-channel phase fusion. A Bayesian prior distribution over the vibration order guides the estimation toward physically plausible values under noisy conditions. Finally, a coarse-to-fine multi-stage search strategy drastically reduces computational burden while preserving accuracy. Experiments on a rotor-blade test bench at constant and variable speeds show that the method reduces the noise floor by about 60 dB, achieves a maximum frequency identification error of 7.84%, and accelerates the search by approximately 48.6% compared to exhaustive search. The proposed method provides a reliable and efficient solution for blade health monitoring. Full article

This study addresses the critical need for sustainable energy by optimizing diaphragm-type piezoelectric elements for efficient waste vibration energy harvesting. Traditional experimental optimization of complex, non-linear design parameters including applied load, tapper diameter, and support structures is often resource-intensive and time-consuming. To overcome these limitations, we developed a novel hybrid machine learning framework that seamlessly integrates an Artificial Neural Network (ANN) with a Modified Grey Wolf Optimization (mGWO) algorithm. The ANN was rigorously trained on experimental data using Bayesian Regularization, establishing itself as a robust and high-fidelity surrogate model capable of accurately predicting voltage output based on diverse input parameters, evidenced by an R-value close to 1. This predictive model subsequently served as the fitness function for the mGWO algorithm, which incorporated a non-linear control parameter to efficiently explore the multi-dimensional design space and effectively balance exploration with exploitation. The framework successfully identified the optimal configuration for maximizing voltage output, predicting a theoretical maximum of approximately 70.67 V. This optimal setup notably involved a high applied load of 100 N, the 6CA multi-pointed tapper configuration, and the three-support boundary condition, which is consistent with the experimentally validated results. The computational findings demonstrated excellent agreement with empirical results while providing significantly higher resolution for design insights. This validated, predictive tool offers a substantial advancement for the future scaling and design optimization of piezoelectric energy harvesters, minimizing the need for extensive physical prototyping and ensuring efficient stress transfer without mechanical failure. Full article

In this study, variational Bayesian inference (VBI) with Gaussian mixture models is applied to update models of nonlinear structures, and then, the calibrated model is employed to estimate the failure probability of structures using a subset simulation (SS) algorithm. To improve the computation efficiency of probabilistic nonlinear model updating, a Gaussian Process (GP) model is used to construct a surrogate likelihood function in Bayesian inference using an active learning algorithm, and then, Gaussian mixture models (GMMs) are employed to approximate the unknown posterior probabilistic density functions (PDFs) of model parameters. The optimized hyperparameters of GMMs can be obtained by maximizing the evidence lower bound (ELBO), and the stochastic gradient search method is used to solve this optimization problem. Based on the optimized hyperparameters, the posterior distributions of model parameters can be approximated using a combination of multiple Gaussian components. Subsequently, the SS algorithm is used to calculate the earthquake-induced failure probability of structures based on the calibrated nonlinear model. To verify the feasibility and effectiveness of the proposed method, a numerical simulation of a two-span bridge structure subjected to seismic excitations was developed. Moreover, the proposed strategy is further applied to estimate the failure probability of a scaled monolithic column structure subjected to bi-directional earthquake excitations. Both numerical and experimental results indicate that the proposed method is feasible and effective for probabilistic nonlinear model updates, and the updated model can significantly enhance the accuracy of structural failure probability predictions. Full article

Accurate prediction of protein sidechain conformations is a fundamental challenge in structural biology, with diverse applications ranging from protein structure determination to computational drug design. The performance of backbone-dependent rotamer libraries is often limited by discrete binning artifacts and difficulties handling sparse conformational regions. In this work, we present a Bayesian framework for rotamer prediction that addresses these limitations through Dirichlet priors and spatial smoothing. Our approach models rotamer probabilities as continuous functions of backbone dihedral angles, using circular Gaussian convolution, to make the most of statistical strength from neighboring conformations while respecting the periodic nature of angular data. We constructed rotamer libraries through structural clustering of sidechain conformations rather than chi angle binning, ensuring that each rotamer represents a distinct three-dimensional geometry. We evaluated and compared our framework against the state-of-the-art libraries on two independent test sets. Our Dirichlet model achieved chi angle prediction accuracy of 59–60%. Notably, our method produced consistently lower angular errors, an approximate 13% reduction in mean deviation, suggesting that the continuous probability distributions better capture subtle conformational preferences. Further, we explored the incorporation of non-sequential context by including the identity of nearby non-neighboring residues as an example of extensibility of our framework. Full article

Accurate multi-target filtering is crucial for low-altitude surveillance, where point and group targets often coexist. Poisson multi-Bernoulli mixture (PMBM) filters provide a unified Bayesian framework for the joint filtering of point and group targets under the assumptions of independent target dynamics and standard measurement models. However, in practical scenarios, group targets may generate new targets through member separation, while point targets may produce multiple measurements due to multi-beam sensing and micro-Doppler signatures. These phenomena violate the assumptions of existing PMBM filters and lead to degraded state estimation and target-type inference. To address these challenges, this paper proposes a modified PMBM filter with group target spawning and generalized measurement models for coexisting point and group targets. Specifically, a group-dependent spawning model is incorporated into the prediction step to enable timely detection of newly spawned targets. In addition, a generalized update function is developed to support point-target density updates with measurement sets of arbitrary cardinality, and a measurement-rate-based correction factor is introduced to improve target-type estimation under nonstandard measurement conditions. Furthermore, an efficient Poisson multi-Bernoulli approximation is derived to reduce computational complexity. The effectiveness of the proposed filter is verified through simulation and experimental results. Full article

The continuous-flow technologies in organic synthesis for the production of active pharmaceutical ingredients (APIs) are nowadays more and more applied. In-silico process design is a powerful tool able to support organic synthesis in the field of scale-up and process development. Process design feasibility and reliability depend on the availability of a well-defined chemical reaction kinetic scheme, information which is usually derived from experimental datasets collected on purpose. The latter approach is time-consuming and demanding in terms of resources. Different possibilities are here proposed to valorize widely available experimental data from explorative works with different approaches, depending on the nature, richness, and structure of the datasets. The kinetic parameters (i.e., reaction order, kinetic constant, and activation energy) of some interesting organic reactions have been approximately estimated by applying different computational methodologies, thanks to built-in experimental databases. The numerical algebra approach dealing with linear and non-linear regression analysis for the kinetic parameters has been initially considered and related to the database information for oseltamivir synthesis. The Bayesian statistic was applied to the ibuprofen case through the application of the Markov Chain Monte Carlo (MCMC) method for reaction order estimation. At last, a Machine Learning (ML) approach has been applied to the Rolipram and Pregabalin case study. The in-house developed T-ReX experimental kinetic constant database was exploited, with application of the k-Nearest neighbor algorithm for classification and regular expression pattern recognition. Advantages and limitations of the three approaches are discussed. Full article