Abstract As products become more sophisticated and quality demands grow, manufacturing systems are becoming increasingly complex. In particular, for Small and Medium Enterprises (SMEs) operating complex, high-mix, low-volume production systems, manufacturing layout significantly impacts productivity and operational costs. Given this impact, accurate prediction of production performance becomes essential for optimizing layout configurations and resource allocation. In this paper, the objective is to develop a decision-support framework for selecting the best production layouts in labor-intensive SME manufacturing systems. The framework uses Discrete Event Simulation (DES) with Cellular Manufacturing System (CMS) models to generate datasets. And, Gaussian Process Regression (GPR) is applied to provide probabilistic predictions for key performance variables (KPVs) while accounting for inherent uncertainties. When GPR-based production volume predictions for different layouts result in overlapping prediction ranges, Fuzzy Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) is then utilized to systematically rank the alternatives using multiple key variables as evaluation criteria. A case study demonstrates the framework's effectiveness by comparing three alternative layouts with GPR-based production volume predictions and Fuzzy TOPSIS-based decision making. The multi-criteria evaluation systematically ranks the alternatives using KPVs including adjusted cycle time, total buffer capacity, average manpower efficiency, and number of cells.

ABSTRACT Creep-feed grinding of WC–Co cemented carbide is a crucial roughing process in tool manufacturing, where surface integrity directly influences tool performance. This study combines single-factor and orthogonal experiments to quantify how grinding depth, feed rate, and wheel speed affect residual stress, microhardness, and subsurface cracking, and to develop a multi-output predictive framework. Surface roughness Ra increased from 0.507 to 0.595 μm as grinding depth increased and from 0.525 to 0.671 μm as feed rate increased, whereas it decreased from 0.664 to 0.501 μm with increasing wheel speed. The maximum subsurface crack depth increased from 2.06 to 6.33 μm with increasing grinding depth. ANOVA indicated that feed rate most strongly affected Ra, grinding depth dominated residual stress and subsurface damage, and wheel speed exerted the greatest influence on microhardness. A multi-output Gaussian process regression model achieved an overall test-set prediction accuracy of 94.25%.

Cell growth rates exhibit cell-intrinsic cell-to-cell variability, which influences cell fitness and size homeostasis from bacteria to cancer. It remains unclear whether this variability arises from stochasticity in cell growth or division processes, or from cell-size-dependent growth regulation. To separate these potential sources of growth variability, single-cell growth rates need to be examined across different timescales. Here, we study cell size and growth regulation by tracking lymphocytic leukemia cell mass accumulation with high precision and minute-scale temporal resolution along long ancestral lineages. We first show that correlations between growth rates and cell-size nor asymmetric divisions explain cell-to-cell growth variability. We then isolate growth fluctuations by smoothing and detrending the growth rate dynamics using a Gaussian process regression. We find that these growth fluctuations drive cell-to-cell growth variability within ancestral lineages despite being independent of cell divisions, cell cycle, and cell size. Overall, our results provide a quantitative framework for understanding single-cell growth rates, and indicate that cell-intrinsic long-term patterns in growth are a byproduct of short-term growth fluctuations.

Development of Machine-Learning Models for Predicting Escherichia coli O157:H7 Inactivation on Fresh-Cut Lettuce during Chlorine Washing.

Mechanism-driven life Prediction of lithium-ion batteries via a coupled mechanical-electrochemical degradation model and Gaussian process regression

Accurate characterization of thermal diffusion and blackbody radiation on the detector foil is crucial in infrared imaging video bolometry (IRVB) for reliably inferring the spatial distribution of plasma radiation. This paper presents a new inference framework for modeling blackbody radiation and thermal diffusion power densities using Gaussian process regression. This method is validated with both synthetic and experimental IRVB data, producing reliable results without the need for temporal or spatial averaging. In addition, the effects of noise level and foil material are analyzed, and both the limitations of this framework and strategies for improving its performance are identified.

Unveiling sulfonamide adsorption on biochar using explainable machine learning.

Inferring cell division kinetics in Chlamydomonas reinhardtii from flow cytometry with Gaussian process regression.

Accurately predicting International Roughness Index (IRI) is essential for effective pavement maintenance and long-term network sustainability. This study evaluates several advanced machine learning models for IRI prediction in Continuously Reinforced Concrete Pavement (CRCP) using a comprehensive dataset from Long-Term Pavement Performance (LTPP) program. Support Vector Machines, Artificial Neural Networks, Regression Trees, Ensemble Trees, and Gaussian Process Regression (GPR) were developed and assessed using Root Mean Square Error (RMSE) and R-squared (R²). The Matern 5/2 GPR model achieved the best performance, with R² = 0.97 and RMSE = 0.0776. Feature importance analysis using Random Forest identified initial IRI, construction number, layer thicknesses and temperature as the strongest predictors. Sensitivity analysis confirmed the influence of age, climate, and traffic on IRI. Using only the top ten variables produced nearly identical accuracy, improving computational efficiency. Overall, the study demonstrates the strong potential of ML for reliable and sustainable IRI prediction in rigid pavements.

Explainable hybrid modeling of nitrous oxide emissions in wastewater treatment: Integrating mechanistic knowledge with uncertainty-aware machine learning.

Multivariate Gaussian process regression for characterization of geo-data spatial variability from limited and non-co-located measurements

American option pricing with model constrained Gaussian process regressions

Gaussian process regression for modeling computational and experimental mineral processing data

Accurately forecasting the breakdown voltage of insulating oils is a prerequisite for the reliable design and operation of high-voltage equipment. The present work focuses on developing data-driven artificial intelligence (AI) models to predict the breakdown voltage of transformer oil as a function of temperature and electrode spacing. Two different machine learning algorithms are applied and compared: Gaussian Process Regression (GPR) and Radial Basis Function (RBF) neural network. The experimental data for electrode distances of 5 mm and 20 mm are used to train, test, and validate the models using a 60/20/20 data-splitting scheme. The predictive capacity of the models is evaluated using the three metrics: mean absolute error (MAE), root mean square error (RMSE), and coefficient of determination (R). Experimental results confirm that the model predictions are in excellent agreement with the measurements at short electrode distances for both models. Nevertheless, at longer distances, the differences between the two performances become quite substantial. The GPR method is more reliable and generalises better, particularly at 20 mm, where it yields lower validation errors than the RBF approach. In addition, as a probabilistic method, GPR enables the estimation of predictive uncertainty, which is essential for applications oriented toward safety and dependability. Overall, the present work has demonstrated GPR's capability to determine the breakdown voltage of insulating oils and its potential for high-voltage insulation diagnostics and design.

Crop yield estimation, particularly early-season yield prediction, is highly important for global food security and disaster mitigation. In this study, we utilized deep learning models combined with remote sensing data to develop in-season crop yield estimation models, enabling immediate yield prediction. We employed a convolutional neural network (CNN) for spatial feature extraction and a long short-term memory network (LSTM) for temporal patterns, complemented by Gaussian process regression (GP) that introduced geographical coordinates. Three groups of in-season yield prediction experiments were designed, utilizing four-phase, two-phase, and single-phase data, respectively. The results indicated that under the two-phase training scheme, the LSTM_GP model achieved the highest performance in the sixth period, with an R2 value of 0.61 and a root mean square error (RMSE) value of 983.38 kg/ha. When trained on single-phase data at the twelfth phase (approximately mid-to-late July), the LSTM_GP model also performed best, attaining an R2 value of 0.62 and an RMSE value of 969.06 kg/ha. The single-phase prediction model outperformed time-series models in yield prediction accuracy. The periods from mid-to-late July to early-to-mid August represent critical crop growth stages were essential for accurate yield prediction. From our research, we found that adding GP can improve the prediction accuracy, especially for LSTM. Moreover, the proposed single-phase prediction model realized reliable crop yield prediction as well as the silking to early grain-filling stage (mid-to-late July), providing a critical lead time of approximately 2–2.5 months before harvest to support pre-harvest agricultural decision-making.

Research on eco-friendly and energy-efficient machining processes has gained significant importance within the domain of sustainable production. This study is focused on enhancing the energy performance and sustainability of the milling process. Four machine learning (ML) models, namely, multiple linear regression (MLR), support vector regression (SVR), Gaussian process regression (GPR), and adaptive network-based fuzzy inference system (ANFIS), were proposed to estimate specific energy consumption (SEC) in the milling of Ti6-Al4-V under two eco-benign cooling conditions: cryogenic and minimum quantity lubrication (MQL). Several statistical metrics, including normalized mean absolute error (nMAE), mean absolute percentage error (MAPE), normalized root mean square error (nRMSE), maximum absolute percentage error (maxAPE), coefficient of determination (R2), and Willmott’s index of agreement (IA), were employed to validate the performances of the ML models. A high level of agreement between the predicted and experimental SEC data for both the training and test datasets supports the reliability of the proposed ML models. Although the MLR model performed well, the results revealed that the other ML models demonstrated better overall performance. According to the statistical metrics, the models’ predictive performance improved in the following sequence: MLR, SVR, GPR, and finally ANFIS, which demonstrated the highest predictive capability.

Abstract Radial pumps and compressors are used in various engineering applications, including rocket turbopumps, automotive turbochargers, and refrigeration systems. Several physical effects, including viscous losses, flow separation, compressibility, and rotational dynamics, dominate radial impeller flow, making flow field prediction very difficult and requiring computationally expensive CFD. However, designers typically only require information at specific positions, resulting in most simulation data being unused. Accurately predicting the impeller exit flow field is often key to impeller design. Recent advances in reduced-order modeling and machine learning show promise for a priori flow field prediction. In this study, nine different reduced-order models (ROMs) were created to predict the dimensionless exit flow field of radial flow impellers in real time. The ROMs consist of various linear and nonlinear dimensionality techniques paired with different regressors. Inputs include parameterized, dimensionless impeller geometry based on Bezier control points, number of blades, and dimensionless operating conditions. The ROMs were trained using over 1800 flow fields from high-fidelity CFD simulations representing a large radial flow compressor design space. ROMs were evaluated on relative error, training time, and evaluation time. Principal component analysis coupled with Gaussian process regression (PCA-GPR) emerged as the preferred ROM, training within two seconds, evaluating hundreds of cases in real time, and matching the accuracy of computationally demanding nonlinear alternatives. PCA-GPR predictions show pressure, density, and velocity profiles within 5% average of CFD results. The ROM was validated through four test cases probing robustness across different operating conditions and geometries.

This paper evaluates and compares four data-driven methods (Gaussian Process Regression (GPR), echo state network (ESN), gated recurrent unit (GRU), and long short-term memory (LSTM)) for lithium-ion capacity prognostics adapted to electric vehicle conditions. This comparison aims to find the most efficient prognosis method considering two constraints: the limitation of computational power and the unavailability of on-board capacity measurement that requires full charge and discharge conditions. The machine learning models are trained using capacity values estimated under vehicle conditions. The ageing data is collected from cycling tests of two battery chemistries, Lithium Fer Phosphate (LFP) and Nickel Manganese Cobalt (NMC), with different ageing trends. The prognosis algorithms are tuned with three different percentages of the data, allowing for the evaluation of the methods at different ageing stages. The comparison and analysis of the results show that ESN outperforms other methods; it has the lowest prediction error (mean absolute percentage error less than 1.4% at initial ageing of the cells) and the shortest training time, making it the most appropriate method for automotive applications.

The increasing demand for functional foods has stimulated the creation of healthier bakery items. Riceberry flour, sourced from Thai purple rice, is abundant in antioxidants and dietary fiber, rendering it a viable substitute for wheat flour in donut recipes. Nonetheless, its incorporation can profoundly modify sensory characteristics including texture, hue, and general palatability. Hydrocolloids such as Hydroxypropyl Methylcellulose (HPMC) and Methylcellulose (MC) are frequently included to enhance moisture retention and diminish oil absorption; nevertheless, their interactions with Riceberry flour are intricate and nonlinear. In this study, the use of machine learning approaches for predicting the sensory qualities of Riceberry flour-formulated functional donuts is investigated. Three regression models—Gaussian Process Regression (GPR), Support Vector Regression (SVR), and Random Forest Regression—were assessed for their efficacy in predicting Overall Liking, Taste, and Texture scores based on constituent composition. The findings indicated that GPR consistently surpassed the other models, attaining the lowest MAE and RMSE values for all sensory targets. SVR offered competitive performance, however Random Forest displayed elevated prediction errors, especially for Texture. These findings underscore the efficacy of kernel-based models in elucidating nonlinear ingredient-sensory connections within tiny, noisy datasets. This study illustrates the viability of using machine learning into nascent product development, facilitating enhanced precision and efficiency in sensory optimization for functional food design.

This paper presents a novel computational framework for the design of closed-loop guidance and control laws in highly uncertain environments, possibly characterized by unmodeled dynamics and external disturbances. The proposed methodology integrates the use of Gaussian process regression (GPR) to model uncertainties within a stochastic optimal control framework. GPR provides a probabilistic approach to disturbance estimation using simulated or observed data to characterize noise as a Gaussian process. Convexification techniques are used to transform the original nonconvex stochastic control problem into a sequence of deterministic convex problems that can be efficiently solved by state-of-the-art interior-point algorithms. As a study case, the docking maneuver between an active chaser spacecraft and a passive target spacecraft in the presence of differential drag, with uncertainty in the ballistic coefficient, is considered. The obtained numerical results suggest the effectiveness of the proposed approach in compensating for the disturbances and driving the system to the target final distribution.