The main aim of this study was to evaluate the optimum conditions for extracting the total reducing sugar content for bioethanol production using spirulina algae. The spirulina algae was pretreated using microwave-assisted acid hydrolysis, and the parameters were optimized using response surface methodology (RSM). The selected independent parameters were microwave power (250-350 W), sulfuric acid concentration (1-7%), and time duration (1-5min). The results showed that a maximum reducing sugar concentration of 3.8mg/mL was produced at optimum conditions. ANOVA and R-squared (R2) value (99.87%) show the model was significant (p value is < 0.0001). Additionally, a study on optimization and modeling was conducted utilizing response surface methodology (RSM) as well as artificial neural networks (ANN) to evaluate the impact of temperature (30-40°C), concentration of inoculum (1-5g/L), and fermentation duration (12-45h). This comparative assessment showed that the highest ethanol concentration of 1.824g/L was achieved under optimal conditions of 30°C, 5g/L inoculum concentration, and 28.5h duration, as determined by the high-performance liquid chromatography method. Finally, it is suggested that the RSM approach demonstrated superior performance with a higher R2 value (97.42%), p value is < 0.0001 (significant), and a lower mean square error (MSE) of 0.0065 compared to the ANN model.

This study experimentally investigates the effects of adding different amounts (1-5 wt.%) of Al2O3 particles on the wear behavior of glass fiber-reinforced epoxy composites to improve their tribological performance. Composite laminates produced using the hand-lay up method were subjected to wear tests using a ball-on-disc test setup under dry sliding conditions. Among all tested compositions, the composite containing 3 wt.% Al2O3 exhibited the highest wear resistance. Compared to the neat composite, the specific wear rate was reduced by up to 70%. In contrast, 4% and 5% Al2O3 additions resulted in a decrease in wear resistance due to particle agglomeration. While the highest specific wear rate was 260×10⁻⁶ mm³/Nm, this value decreased to 80×10⁻⁶ mm³/Nm in the 3% added sample. Furthermore, wear rate predictions were performed using models such as artificial neural network and different machine learning regressors. Random Forest (17.62%), Ridge regressor (18.46) and artificial neural network (19.92%) achieved the lowest MAPE values, indicating strong predictive performance for Al2O3-reinforced glass fiber composites. The artificial neural network model optimized with grid search achieved a mean squared error of 0.90 and a coefficient of determination of 0.92, while the random forest regressor demonstrated strong generalization with a coefficient of determination of 0.91. The results demonstrated the critical roles of both particle ratio and data-driven models in wear performance analysis.

Ensuring the authenticity of meat products is a critical issue for consumer protection, regulatory compliance, and the integrity of local food systems. In this study, attenuated total reflectance Fourier-transform infrared (ATR-FTIR) spectroscopy combined with chemometric and machine learning models was applied to differentiate Maltese from non-Maltese pork. Spectral datasets were subjected to a range of preprocessing techniques, including Savitzky–Golay first and second derivatives, detrending, orthogonal signal correction (OSC), and standard normal variate (SNV). Linear methods such as principal component analysis–linear discriminant analysis (PCA-LDA), the soft independent modeling of class analogy (SIMCA), and partial least squares regression (PLSR) were compared against nonlinear approaches, namely support vector machine regression (SVMR) and artificial neural networks (ANNs). The results revealed that derivative preprocessing consistently enhanced spectral resolution and model robustness, with the fingerprint region (1800–600 cm−1) yielding the highest discriminative power. While PCA-LDA, SIMCA, and PLSR achieved high accuracy, SVMR and ANN models provided a superior predictive performance, with accuracies exceeding 0.99 and lower misclassification rates under external validation. These findings highlight the potential of FTIR spectroscopy combined with nonlinear chemometrics as a rapid, non-destructive, and cost-effective strategy for meat authentication, supporting both consumer safety and sustainable food supply chains.

ABSTRACT This study responds to the complex relationships between Virtual Reality (VR), Augmented Reality (AR), Perceived Environmental Responsibility (PER) and Ecotourism Destination Characteristics (EDC). Conceptualization of the relationship is based on the Stimulus-Organism-Response (SOR) framework and Value-Belief-Norm Theory. To address the developmental issues in sustainable tourism, data were collected from experienced ecotourists using a structured questionnaire consolidated from established scales. After statistical purifications 457 responses were processed to final analysis via Partial Least Squares Structural Equation Modelling (PLS-SEM) and Artificial Neural Networks (ANN). Results revealed that adoption of VR and AR technologies significantly enhances PER, which in turn positively influences sustainable practices in ecotourism. The mediation analysis confirmed that the impact of VR and AR technologies on sustainable practices is fully transmitted through these two constructs. Furthermore, the ANN model exhibited superior predictive accuracy. The findings highlight the theoretical, practical, and social potential of immersive technologies in driving sustainable behaviour and improving destination management in ecotourism, while suggesting avenues for future research to explore additional contextual factors and long-term impacts.

Water quality and quantity affect crop productivity, with surface water quality having a significant impact. The amount of surface water being used for drinking is gradually rising. Thus, assessing surface water quality and related hydro-chemical characteristics is essential for surface water resource management in Mahanadi River Basin, Odisha. The current study examined surface water quality and appropriateness for drinking and agriculture, utilizing several techniques such as Weighted Arithmetic (WA) Water Quality Index (WQI), Multivariate models namely Pearson Correlation, Cluster Analysis (CA) and Principal Component Analysis (PCA), six multiple machine learning (ML) techniques like, gaussian process regression (GPR), linear regression (Stepwise), fit binary tree (FBT), support vector regression, SVM (linear and polynomial kernels), and artificial neural network (ANN) to predict the WQI, for sustainable use of the surface water resources. Thirteen physicochemical parameters were used to analyse eleven surface water samples, which indicating that the primary cation and anion concentrations were as follows: Mg2+ > Ca2+ > K+ > Na+, and HCO3- > Cl- > SO42- > NO3-, respectively. The best input combination for WQI model prediction was identified using subset regression analysis. These eight input combinations had high R2, ranging from 0.975 to 1, and high Adjusted R2 amounts to 0.974-1. The WAWQI range is divided into five categories: excellent (18.18%), good (18.18%), poor (27.27%), very poor (27.27%), and unsuitable (9.09%). The study discovered that increased turbidity concentration, carbonate weathering, and the growth of agricultural and urban-industrial sectors regulate the geographical variance in surface water quality. The correlation results depict that the significant positive correlation has been found between TDS and TH (0.87), Mg2+ with turbidity (0.84) and coliform (0.78), Ca2+ and coliform (0.72), Cl- and HCO3- (0.83), and K+ and Na+ (0.7). Owing to the correlation study, these ions are enriched in the surface water by major anthropogenic activity. While, in the present study, CA and PCA has been used to determine the surface water's governing factors. Differentiation of three clusters based on the sources, hydrogeochemical environment, and reactions between chemical variables by utilizing CA and the results of PCA shows that the first three primary components (PCs) account for 84.76% of the overall variation. Hence, CA and PCA shows the several processes that are the main sources of the ions, such as carbonate, silicate weathering, and evaporate dissolution. Pursuant to the stepwise fitting model, bicarbonate was a non-significant variable for the WQI, whereas turbidity, pH, and coliform were the most significant factors. With a high correlation of 1 and low errors, the results demonstrated that the GPR, stepwise linear regression, and ANN models outperformed the others during the training and testing phases.In contrast, during the training and testing stages, the SVM and FBT models showed the lowest performance. Therefore, the GPR, stepwise regression, and ANN models exhibited low mistakes and a strong correlation during the training and testing phases. In conclusion, the combination of physicochemical characteristics, WQI, CA, PCA, and ML tools to assess the surface water suitability for drinking and irrigation and their regulating variables are beneficial and provides a clear picture of water quality. Future research should improve the data accuracy to increase model precision and extend its applicability to various geographical and environmental settings.

In order to improve the competitiveness of smart tourist attractions in the tourism market, this paper selects a scenic spot in Shenyang and uses big data technology to predict the passenger flow of the scenic spot. Firstly, this paper introduces the big data-driven forecast model of scenic spot passenger flow. Based on the traditional autoregressive integral moving average model and artificial neural network model, it builds a big data analysis and forecast model. Through the analysis of data source, model building, scenic spot passenger flow accuracy, and modeling time comparison, it affirms the advantages of big data analysis in forecasting scenic spot passenger flow. Finally, it puts forward four commercial operation optimization strategies: adjusting the ticket pricing of scenic spots, upgrading the catering and accommodation services in scenic spots, planning and designing play projects, and formulating accurate scenic spot marketing strategies, in order to provide references for the optimization and upgrading of smart tourist attractions in the future.

Understanding the relationship between environmental variables and gas concentrations from livestock production is essential for evaluating the impact of pollutants on animal housing and surrounding areas. This study investigates the use of ANNs to predict NH3 concentrations in a Mediterranean dairy barn under seasonal conditions-namely, hot, cold, and transitional weather. A Multi-Layer Perceptron (MLP) structure was employed, trained using Levenberg-Marquardt and Bayesian Regularization algorithms. The input dataset included ten variables related to internal and external environmental conditions, NH3 concentrations, and time of day. The models were evaluated using R2, R, MAE, MSE, and RMSE as performance metrics. Results showed strong predictive capabilities, with R2 values ranging from 0.75 to 0.96 and RMSE values between 0.47 and 0.80 due to the number of input data (different days) and environmental conditions. These findings highlight the potential of ANNs as effective tools for real-time pollutant prediction, supporting Precision Livestock Farming (PLF) strategies.

The application of automatic technology based on artificial intelligence is one of the important elements in supportingthe transformation towards Industry 4.0, especially in the palm oil industry sector. This study aims to develop a CrudePalm Oil (CPO) level monitoring system in a bulking tank using an ultrasonic sensor and automatically regulate thepump speed based on the classification of the Neural Network (NN) results implemented on an Arduino microcontroller.The designed system utilizes an HC-SR04 ultrasonic sensor to read the CPO surface height, then the data is processedthrough an artificial neural network model with one hidden layer containing three neurons and six output classes todetermine the pump speed level. The results of this conversion become a PWM signal which is used to control the pumpmotor through the motor driver. Testing was carried out 10 times with variations in liquid height, showing that the sensorhas a high level of accuracy with an average error of 0.13 cm. The NN model produces a classification accuracy of 100%on the test data, and the motor speed control runs proportionally to the liquid level. This system has proven to beresponsive and capable of controlling fluids efficiently and in real-time. The results of the study indicate that thisapproach is feasible to be applied for intelligent and adaptive automation of filling and emptying CPO tanks, in line withthe principles of Industry 4.0

Abstract Over the past decade, the Global Positioning System (GPS) has been extensively employed in land vehicle navigation systems. Inertial Navigation Systems (INSs) are utilized in scenarios where GPS fails to deliver consistent and reliable navigation solutions. It is well-known that the low-cost INS sensors can exhibit significant errors. To mitigate these errors, the measurements of position and velocity sensors from GPS are integrated and fused. In this study, an intelligent technique has been proposed to optimize the benefits of GPS and INS while minimizing their drawbacks. This study employed an artificial intelligence model to predict the instantaneous INS error based on instantaneous de-noised measurements and readings. Different artificial neural network (ANN) models have been tested, examined and compared. The results revealed that a considerable improvement has been reached in terms of ability to predict INS error during different GPS outages. However, a promising result have been found using RNN model. As compared to other ANN models, the RNN model gives better performance in terms of standard deviations of position and velocity errors. The standard deviations of prediction position errors resulting from RNN are equal to 0.0864, 0.0726, and 0.1297 min the sense of x, y and z channels, respectively, while the standard deviations of predicted velocity errors are 0.0025, 0.0029, and 0.0049 m s−1 on the x, y, and z axis, respectively.

An important development in the modelling of exchange rate volatility is the use of artificial neural networks (ANN) to create enhanced generalised autoregressive conditional heteroskedasticity (GARCH) models. Conventional GARCH models are good at capturing the clustering of volatility in financial time series, but they have trouble understanding complex linkages and non-linear patterns in the data. This paper aims to investigate the hybrid approach in modelling exchange rate volatility of two currency pairs: South African Rand against Brazilian Real (ZAR/REAL) and South African Rand against Chinese Yuan (ZAR/YUAN) using monthly observations over the period of January 1996-March 2024. The paper introduced ANN as an additional factor to both symmetric and asymmetric GARCH models that capture most common stylised facts about exchange rate volatility and leverage effects. The GARCH (1,1)-ANN, EGARCH (1,1)-ANN, and GJR-GARCH (1,1)-ANN models were used, and their performance was assessed using mean squared error (MSE), mean absolute error (MAE), and mean absolute percentage error (MAPE). The results revealed that the EGARCH (1,1)-ANN model had the best overall performance when compared to the other hybrid models based on all three evaluation measures for both the currency pairs’ data. The paper recommends further similar studies to predict future exchange rate trends and also incorporating other nonlinear methods.

The Mediterranean fig (Ficus carica L.) is a dioecious fruit tree of high nutritional and economic value in the Mediterranean basin. In northern Morocco, phenological desynchronization between male and female fig trees limits pollination and production. This study aimed to characterize the phenological stages of indigenous fig and caprifig varieties using the BBCH scale and to evaluate the predictive capacity of artificial neural networks (ANNs). This study was conducted in the Bni Ahmed region over two consecutive years (2021 and 2022) at two sites. At each site, a total of 80 female fig trees were selected. Caprifig trees were selected in accordance with their availability (37 trees/site 1; 24 trees/site 2). Local meteorological data were incorporated into the analysis to evaluate the influence of climatic conditions on phenological stages. Our results revealed significant effects of temperature, humidity, and rainfall on phenological dynamics, along with a clear inter-varietal variability and pronounced desynchronization between male and female fig trees. Early-ripening caprifig varieties showed limited pollination efficiency, whereas late-ripening varieties were better synchronized with the longer receptivity period of female fig trees. Importantly, the ANN model demonstrated exceptional predictive performance (R2 up to 0.985, RMSE &lt; 1 day), serving as a robust and practical tool for forecasting key phenological stages and minimizing potential yield losses. These findings demonstrate the value of combining phenological monitoring with AI-based modeling to improve adaptive management of fig orchards under Mediterranean climate change. This is the first study in Morocco to implement such an integrated approach to fig and caprifig trees.

Portal hypertension (PHT) is pivotal in managing decompensated cirrhosis. In clinical practice, hepatic venous collaterals are frequently present, often leading to failure or reduced accuracy of hepatic venous pressure gradient (HVPG) measurements, thereby making HVPG an imperfect surrogate for the portal pressure gradient (PPG). Artificial neural networks (ANNs) have shown potential in integrating multidimensional clinical variables and predicting complex disease states; however, their value in the assessment of PHT remains insufficiently validated. The study compared ANN-predicted PPG with measured PPG and HVPG in two cohorts: Group A (all patients), reflecting routine clinical practice, and Group B (excluding cases with coefficient of variation (CV) > 30%, most with venous collaterals), approximating optimized conditions. Subgroup analyses in Group B further assessed differences by etiology and Child–Pugh class. We retrospectively included 164 patients with decompensated cirrhosis who underwent TIPS between June 2014 and July 2024, with intra-procedural HVPG and PPG measurements. An ANN model predicted PPG based on INR, WBC, and portal vein diameter. Group A included all patients (n = 164), reflecting real-world conditions where HVPG may be affected by collaterals. Group B represented a strict quality-control cohort (n = 101), in which cases with a measurement CV > 30% were excluded; retrospective review indicated that most of these excluded patients exhibited hepatic venous collaterals thereby approximating an “ideal condition” without the influence of collaterals. Statistical analyses included paired t tests, Pearson correlations, Steiger’s Z-tests, and Bland–Altman analysis. Subgroup analyses were conducted by etiology and Child–Pugh class. In the overall cohort (Group A, n = 164), HVPG showed negligible correlation with PPG (r = 0.014), whereas ANN-predicted PPG demonstrated moderate correlation (r = 0.437, P < 0.001) with significantly narrower LoA. In the quality-controlled cohort (Group B, n = 101), both HVPG and ANN-predicted PPG correlated moderately with PPG (r = 0.457 vs. 0.476) with comparable agreement. Subgroup analyses indicated that ANN outperformed HVPG in hepatitis B and Child–Pugh C patients, while HVPG was slightly better in alcohol-related cirrhosis; both methods performed poorly in autoimmune liver disease. HVPG remains the gold standard for assessing portal pressure but is limited by hepatic venous collaterals, advanced liver dysfunction, and the need for invasive measurement. ANN-predicted PPG showed favorable correlation and agreement with measured PPG, providing a noninvasive, simple, and reproducible complement to HVPG for clinical assessment and follow-up.

Sheath blight of rice caused by Rhizoctonia solani is a major disease that causes substantial yield loss ‎of production globally. Development and use of resistant cultivars is a cost-effective strategy to ‎manage the disease. The present study encompasses 43 rice germplasms that were evaluated for sheath ‎blight resistance during 2022-2024 in Ranadevi farm of Centurion University of Technology and ‎Management, Paralakhemundi, Odisha. Evaluation was conducted under natural field conditions ‎supplemented with artificial inoculation of the pathogen. Disease severity was measured using a 0-5 ‎disease grading scale. The classification of genotypes was carried out using Percent Disease Index ‎‎(PDI), Area Under Disease Progress Curve (AUDPC), and Genotypic Category (GC). Disease ‎modeling was established through the Gompertz model, RF (Random Forest), and ANN (Artificial ‎Neural Network) models. Some of the genotypes like Swarna Subhagya, CR 1017, and NLR 33892 ‎indicated consistent resistance or moderate resistance between seasons, having lower AUDPC ‎value (&lt;400), whereas, number of famous varieties (&gt;15), such as IR 64, MTU 1010 and CR Dhan ‎‎308, recorded high susceptibility with AUDPC of more than 1500 and GC score of 5. Also, disease ‎progression was well fitted within the different categories of resistance (R2 = 0.9747 - 0.9963) ‎using the Gompertz model where whereas in the machine learning model, resistance responses were ‎categorized. Further, compared to the test accuracy of the (Random Forest) RF classifier, which was ‎‎76.92 per cent, the ANN (Artificial Neural Network) model produced an accuracy of 81 per cent‎.

Legionella is an environmental bacterium capable of causing severe respiratory infections, with outbreaks posing significant public health challenges in developed countries. Understanding public awareness of Legionella transmission, risk perception, and preventive behaviors is crucial for reducing exposure and guiding health education strategies. This study aimed to evaluate the Portuguese population’s knowledge of Legionella infections and their readiness to adopt preventive measures. A structured questionnaire was developed and administered to 239 participants aged 18–76 years across Portugal, collecting socio-demographic data and assessing literacy through statements organized into domains related to Legionella risk, control measures, and public health impact. The results indicate that participants possess moderate to high awareness of Legionella severity, transmission routes, and preventive strategies, yet gaps remain in understanding key risk factors, optimal water system maintenance, and the influence of temperature on bacterial growth. Age, educational attainment, and occupational status were associated with differences in self-assessed literacy levels. Artificial neural network models were applied to classify literacy levels, achieving a near 90% accuracy and demonstrating higher confidence in low and moderate categories. These findings provide insights for designing tailored educational programs, improving public health communication, and enhancing preventive behaviors to reduce Legionella infection risks.

Design of experiments (DOE) is a crucial tool for improving product or process quality. Accurately modelling the relationship between process variables and quality characteristics is essential. However, traditional DOE methods have limitations when dealing with complex data and cannot effectively fit the data. Data-driven methods can handle complex data and construct high-precision empirical models, but they often lack interpretability. Therefore, in this article, we incorporate explainable artificial intelligence (XAI) into DOE for the purpose of constructing high-precision models, improving interpretability, and identifying important experimental factors. First, we use artificial neural networks (ANN) to model DOE data and construct a high-precision empirical model. Then, since engineers usually focus only on the optimal solution and its nearby local space, we use the local interpretable model-agnostic explanation (LIME) method to interpret the ANN model and enhance the local interpretability of the model. Finally, we design an experiment and obtain the response using the ANN model to identify the important experimental factors. At the end of the article, we validate the accuracy of the ANN model through a simulation study, and illustrate the proposed method using two datasets from designed experiments.

Integrating AI-based neural network modeling with experimental characterization for Cd(II) ion adsorption using Sargassum fusiforme biosorbent.

In underground tunnel construction for mining, the drilling and blasting method is widely used due to its advantages, such as low cost, simple calculation and implementation, and applicability in various geological and hydrogeological conditions. The drilling and blasting method is also suitable for tunnels with different cross-sectional shapes. One parameter that significantly influences the effectiveness of the drilling and blasting method is the area of the tunnel face after blasting. In this study, 136 datasets of influencing parameters and the tunnel face area after blasting from the DeoCa tunnel construction project were used to develop an artificial neural network (ANN) model capable of predicting the tunnel face area after blasting. The paper developed an ANN model and proposed a hybrid model based on the ANN model combined with a genetic algorithm (GA) to predict the area of the tunnel face after blasting. The input variables for the models included the designed tunnel face area (Sd), the specific charge (SC), the average borehole length (L), and the rock mass rating (RMR) of the rock mass on the tunnel face. This paper demonstrates that the hybrid GA-ANN model provides more accurate calculations and predictions for the tunnel face area after blasting than the ANN model alone.

Background: The study of non-Newtonian fluids in thin channels is crucial for advancing technologies in microfluidic systems and targeted industrial coating processes. Nanofluids, which exhibit enhanced thermal properties, are of particular interest. This paper investigates the complex flow and heat transfer characteristics of a Sutterby nanofluid (SNF) within a thin channel, considering the combined effects of magnetohydrodynamics (MHD), Brownian motion, and bioconvection of microorganisms. Analyzing such systems is essential for optimizing design and performance in relevant engineering applications. Method: The governing non-linear partial differential equations (PDEs) for the flow, heat, concentration, and bioconvection are derived. Using lubrication theory and appropriate dimensionless variables, this system of PDEs is simplified into a more simplified system of ordinary differential equations (ODEs). The resulting nonlinear ODEs are solved numerically using the boundary value problem (BVP) Midrich method in Maple software to ensure accuracy. Furthermore, data for the Nusselt number, extracted from the numerical solutions, are used to train an artificial neural network (ANN) model based on the Levenberg–Marquardt algorithm. The performance and predictive capability of this ANN model are rigorously evaluated to confirm its robustness for capturing the system’s non-linear behavior. Results: The numerical solutions are analyzed to understand the variations in velocity, temperature, concentration, and microorganism profiles under the influence of various physical parameters. The results demonstrate that the non-Newtonian rheology of the Sutterby nanofluid is significantly influenced by Brownian motion, thermophoresis, bioconvection parameters, and magnetic field effects. The developed ANN model demonstrates strong predictive capability for the Nusselt number, validating its use for this complex system. These findings provide valuable insights for the design and optimization of microfluidic devices and specialized coating applications in industrial engineering.

Effective delineation of Agricultural Drought Hazard (ADH) zones is crucial for mitigating crop losses and ensuring water security in semi-arid regions. Conventional agricultural drought assessment methods, reliant on single-index approaches or static multi-criteria frameworks, struggle to capture the non-linear interactions between geo-environmental drivers that govern drought severity in semi-arid, rainfed basins. This study introduces a Machine Learning (ML)-geospatial framework integrating satellite-derived indices with soil-hydrological parameters to overcome the limitations of conventional drought assessment methods. Four popular ML models, Random Forest (RF), Artificial Neural Network (ANN), Support Vector Machine (SVM), and Adaptive Regression (AR), are utilized for this purpose, considering eight geo-environmental input variables. Model performance was rigorously evaluated in the Upper Dwarakeshwar River Basin (UDRB), a drought-prone, rainfed catchment in eastern India, using a suite of standard statistical approaches. The RF model excelled with a 97.8% area under the curve-receiver operating characteristic (AUC-ROC) curve and root mean square error (RMSE) of 0.26, followed by the SVM model (94.6%, 0.28). The ANN model, too, yielded promising results (93.8%, 0.32), while the AR model exhibited the least performance (90.0%, 0.31). Based on the outputs from all four ML models, ADH mapping for UDRB revealed that 24.85-44.35% of its area was identified as very high and 16.96-22.86% as high ADH regions. From a practical application point of view, the findings of this study and ADH maps are helpful in various aspects, ranging from early drought warning to emergency preparedness, advancing precision agriculture in rainfed basins, where 60-80% of livelihoods depend on climate-vulnerable farming.

Recent studies have demonstrated a growing global interest in utilising agricultural waste to remediate wastewater. This stems from growing apprehensions about high levels of heavy metals, especially Pb2+ ions, in wastewater produced by industrial processes such as mining, paint production, oil refining, smelting, and electroplating. This study examined apple pomace's Pb2+ ions adsorption from wastewater. Response Surface Methodology (RSM) was employed, utilising the central composite face-centred design (CCFD) with three variables: initial concentration (1–50 mg/L), adsorbent dosage (0.1–1 g), and particle size (75–425 µm) to formulate a mathematical model for the biosorption of Pb2+ ions on apple pomace. An artificial neural network (ANN) was developed using data generated from the RSM design. The CCFD and ANN models showed considerable efficacy in the adsorption process, exhibiting correlation coefficient values of 0.9921 and 0.9999, respectively. The isotherm and kinetic studies were performed, and the Freundlich Isotherm model best fitted the equilibrium data, with a correlation coefficient of 0.972 and a qe of 5.145 mg/g. Additionally, the pseudo-second-order model proved to be the most appropriate for the kinetic data, with an R2 of 0.9996. These results confirm that apple pomace functions as an effective, low-cost, and environmentally and sustainably biosorbent for the removal of Pb2+ ions from wastewater. Both RSM and ANN models exhibited high predictive capability for the biosorption process. While ANN provides more flexibility in modelling complex non-linear relationships, it is prone to overfitting, particularly with limited datasets, and this was addressed through a 5-fold cross-validation technique.