Principal Component Analysis Technique Research Articles

Validity and reliability of a questionnaire developed to explore quality assurance components for teaching and learning in vocational and technical education

Most vocational and technical education (VTE) educational systems invest in quality assurance for teaching and learning to improve the skills of the workforce that would contribute to socio-economic development. Quality assurance provides confidence in educational services to meet the quality desires of teaching and learning settings. Yet it has been reported that there are challenges in teaching and learning effectiveness due to a lack of technological breakthroughs among students, resulting in an insufficiently skilled workforce. The primary aim of this study is to develop and validate a questionnaire for quality assurance for teaching and learning in vocational and technical education. The questionnaire was developed based on the data obtained by item generation through an interview with quality managers, administrators, and lecturers and a literature review. The content validity of the quality assurance for teaching and learning (QATL) questionnaire was validated using expert judgment. The principal component analysis (PCA) and orthogonal varimax techniques examined the structures and validity of the quality components. After exploratory factor analysis on the quality components under input, process, and output domains, the input quality factors explained 78.81% of the cumulative variance, the process quality factors explained 72.679% of the cumulative variance, and the output quality factors explained 75.027% of the cumulative variance. The analysis used Cronbach’s alpha to test the internal consistency reliability, and the alpha values ranged from 0.835 to 0.963, indicating very high reliability. Present findings suggest that the QATL is a valid and reliable instrument for assessing quality components of quality assurance for teaching and learning in vocational and technical education. These quality components enable optimal VTE objective achievement if implemented in teaching and learning.

Martian Atmospheric Spectral End‐Members Retrieval From ExoMars Thermal Infrared (TIRVIM) Data

AbstractKey knowledge about planetary composition can be recovered from the study of thermal infrared spectral range datasets. This range has a huge diagnostic potential because it contains diagnostic absorptions from a planetary surface and atmosphere. The main goal of this study is to process and interpret the dataset from the Thermal Infrared channel (TIRVIM) which is part of the Atmospheric Chemistry Suite of the ExoMars2016 Trace Gas Orbiter mission to find and characterize dust and water ice clouds in the atmosphere. The method employed here is based on the application of principal component analysis and target transformation techniques to extract the independent variable components present in the analyzed dataset. Spectral shapes of both atmospheric dust and water ice aerosols have been recovered from the analysis of TIRVIM data. The comparison between our results with those previously obtained on Thermal Emission Spectrometer (TES) data and with previous analysis on TIRVIM data, validates the methodology here applied, showing that it allows to correctly recover the atmospheric spectral endmembers present in the TIRVIM data. Moreover, comparison with atmospheric retrievals on PFS, TES and IRIS data, allowed us to assess the temporal stability and homogeneity of dust and water ice components in the Martian atmosphere over a time period of almost 50 years.

Development of an Index to Assess COVID-19 Hospital Care Installed Capacity in the 450 Brazilian Health Regions.

The article seeks to assess the Brazilian health system ability to respond to the challenges imposed by the coronavirus disease 2019 (COVID-19) pandemic by measuring the capacity of Brazilian hospitals to care for COVID-19 cases in the 450 Health Regions of the country during the year 2020. Hospital capacity refers to the availability of hospital beds, equipment, and human resources. We used longitudinal data from the National Register of Health Facilities (CNES) regarding the availability of resources necessary to care for patients with COVID-19 in inpatient facilities (public or private) from January to December 2020. Among the assessed resources are health professionals (certified nursing assistants, nurses, physical therapists, and doctors), hospital beds (clinical, intermediate care, and intensive care units), and medical equipment (computed tomography scanners, defibrillators, electrocardiograph monitors, ventilators, and resuscitators). In addition to conducting a descriptive analysis of absolute and relative data (per 10,000 users), a synthetic indicator named Installed Capacity Index (ICI) was calculated using the multivariate principal component analysis technique to assess hospital capacity. The indicator was further stratified into value ranges to understand its evolution. There was an increase in all selected indicators between January and December 2020. It was possible to observe differences between the Northeast and North regions and the other regions of the country; most Health Regions presented low ICI. The ICI increased between the beginning and the end of 2020, but this evolution differed among Health Regions. The average increase in the ICI was more evident in the groups that already had considerably high baseline capacity in January 2020. It was possible to identify inequalities in the hospital capacity to care for patients affected by COVID -19 in the Health Regions of Brazil, with a concentration of low index values in the Northeast and North of the country. As the indicator increased throughout the year 2020, inequalities were also observed. The information here provided may be used by health authorities, providers, and managers in planning and adjusting for future COVID-19 care and in dimensioning the adequate supply of hospital beds, health-care professionals, and devices in Health Regions to reduce associated morbidity and mortality. We recommend that the ICI continue to be calculated in the coming months of the pandemic to monitor the capacity in the country's Health Regions.

Investigation and Discrimination of Ballpoint Pen Inks by Analytical Techniques and Chemometrics.

A population study has been performed for Pakistani ballpoint pen inks of blue, black, red, and green colors (a total of four colors) commercially used in Pakistan. Ballpoint pen inks have been investigated and discriminated by using UV/Vis spectroscopy and FTIR spectroscopy. We have calculated and compared the results in terms of discriminating power (DP). The statistical techniques of principal component analysis and cluster analysis have been applied on obtained data. By visual comparison, the best DP is obtained for green ballpoint pen inks, i.e., 0.866 by using UV/Vis spectroscopy and FTIR techniques. Black and red ballpoint pen inks showed the highest DPs by using UV/Vis spectroscopy; however, blue ballpoint pen inks got the highest DP by using FTIR spectroscopy. DP has been improved by using chemometric techniques and higher DPs are obtained as compared to visual examination.

Evaluation of factors affecting the quality of citizen science rainfall data in Akaki catchment, Addis Ababa, Ethiopia

Citizen Science can fulfill the quest for high-quality and sufficient environmental data, such as rainfall. However, the factors affecting the quality of rainfall data collected by the citizen scientists are not well understood. In this study, we examined the effect of citizen scientists’ attributes on the quality of rainfall data. For this purpose, Principal Component Analysis (PCA), stepwise regression and Multiple Linear Regressions (MLR) were used. A quality control procedure was developed and applied for daily observed rainfall data collected in the summer rainy season of 2020. Attributes of the citizen scientists' were gathered for those who collected rainfall data in the urban and peri-urban Akaki catchment which is located in the Upper Awash sub-basin, Ethiopia. We found that easy-to-detect errors, which were identified during the initial stage of quality control, formed most of the errors in the rainfall data. The PCA and the stepwise regression results revealed that four dominant attributes (education level, gauge relative location, use of smartphone app, and supervisor’s travel distance) highly affected the rainfall data quality. The MLR model using these four prominent dominant variables performed very well with R2 value of 0.98. The k-fold cross validation result showed that the developed model can be used to predict the relationships between data quality and attributes of citizen scientists with high accuracy. Hence, the PCA technique, stepwise regression and MLR model can provide useful information regarding the influence of citizen scientists’ attributes on rainfall data quality. Therefore, future studies should carefully consider citizen scientists' attributes when engaging and supervising citizen scientists, with a comprehensive data quality control while monitoring rainfall.

Distinguishing tumor from healthy tissue in human liver ex vivo using machine learning and multivariate analysis of diffuse reflectance spectra.

The aim of this work was to evaluate the capability of diffuse reflectance spectroscopy to distinguish malignant liver tissues from surrounding tissues and to determine whether an extended wavelength range (450-1550 nm) offers any advantages over using the conventional wavelength range. Furthermore, multivariate analysis combined with a machine learning algorithm, either linear discriminant analysis or the more advanced support vector machine, was used to discriminate between and classify freshly excised human liver specimens from 18 patients. Tumors were distinguished from surrounding liver tissues with a sensitivity of 99%, specificity of 100%, classification rate of 100% and a Matthews correlation coefficient of 100% using the extended wavelength range and a combination of principal component analysis and support vector techniques. The results indicate that this technology may be useful in clinical applications for real-time tissue diagnostics of tumor margins where rapid classification is important.

Development of Input Training Neural Networks for Multiple Sensor Fault Isolation

This paper considers the problem of inhibiting smearing effects for multiple sensor fault isolation. Although the reconstruction-based approach has received considerable attention and achieved excellent results, the investigation into nonlinear systems has been relatively limited. A new reconstruction-based input training neural network aided by the bound and branch method (BAB-RBITNN) is proposed for the nonlinear systems so as to achieve the desired performance of multiple sensor fault isolation. In BAB-RBITNN, a hypothetical faulty direction is naturally integrated into the objective testing function as residual weights for reducing the negative impacts of inaccurate sensor data, taking advantage of the non-static ITNN testing process. The weighted objective function is then considered as the fitness function to obtain the minimum predicted residual by the covariance matrix adaptation evolutionary strategy algorithm. The corresponding optimized fitness is regarded as the reconstruction-based index, which is the critical indicator of fault isolation in the proposed method. Besides, the BAB method is implemented to handle the combinatorial optimization problem instead of a brute search for intelligently pinpointing the real faulty sensors. The efficiency of the proposed method is shown through a validation example and two industrial examples. Comparisons with other methods, including reconstruction-based principal component analysis and reconstruction-based auto-associative neural network techniques, are also presented.

Drought over Southeast Asia and Its Association with Large-Scale Drivers

Abstract In this study, the spatiotemporal variability of drought over the entire Southeast Asia (SEA) region and its associations with the large-scale climate drivers during the period 1960–2019 are investigated for the first time. The 12-month Standardized Precipitation Evapotranspiration Index (SPEI) was computed based on the monthly Global Precipitation Climatology Centre (GPCC) precipitation and the monthly Climate Research Unit (CRU) 2-m temperature. The relationships between drought and large-scale climate drivers were examined using the principal component analysis (PCA) and maximum covariance analysis (MCA) techniques. Results showed that the spatiotemporal variability of drought characteristics over SEA is significantly different between mainland Indochina and the Maritime Continent and the difference has been increased substantially in recent decades. Moreover, the entire SEA is divided into four homogeneous drought subregions. Drought over SEA is strongly associated with oceanic and atmospheric large-scale drivers, particularly El Niño–Southern Oscillation (ENSO), following by other remote factors such as the variability of sea surface temperature (SST) over the tropical Atlantic, the Pacific decadal oscillation (PDO), and the Indian Ocean dipole mode (IOD). In addition, there exists an SST anomaly dipole over the Pacific Ocean, which modulates the atmospheric circulations and consequently precipitation over SEA, affecting drought conditions in the study region.

Multipopulation mortality modelling and forecasting: the weighted multivariate functional principal component approaches

ABSTRACT Human mortality patterns and trajectories in closely related populations are likely linked together and share similarities. It is always desirable to model them simultaneously while taking their heterogeneity into account. This article introduces two new models for joint mortality modelling and forecasting multiple subpopulations using the multivariate functional principal component analysis techniques. The first model extends the independent functional data model to a multipopulation modelling setting. In the second one, we propose a novel multivariate functional principal component method for coherent modelling. Its design primarily fulfils the idea that when several subpopulation groups have similar socio-economic conditions or common biological characteristics such close connections are expected to evolve in a non-diverging fashion. We demonstrate the proposed methods by using sex-specific mortality data. Their forecast performances are further compared with several existing models, including the independent functional data model and the Product-Ratio model, through comparisons with mortality data of ten developed countries. The numerical examples show that the first proposed model maintains a comparable forecast ability with the existing methods. In contrast, the second proposed model outperforms the first model as well as the existing models in terms of forecast accuracy.

Heat Vulnerability Index Mapping: A Case Study of a Medium-Sized City (Amiens)

Urbanization, anthropogenic activities, and social determinants such as poverty and literacy rate greatly contribute to heat-related mortalities. The 2003 strong heat wave (Lucifer) in France resulted in catastrophic health consequences in the region that may be attributed to urbanization and other anthropogenic activities. Amiens is a medium-sized French city, where the average temperature has increased since the year 2000. In this study, we evaluated the Heat Vulnerability Index (HVI) in Amiens for extreme heat days recorded during three years (2018–2020). We used the principal component analysis (PCA) technique for fine-scale vulnerability mapping. The main types of considered data included (a) socioeconomic and demographic data, (b) air pollution, (c) land use and cover, (d) elderly heat illness, (e) social vulnerability, and (f) remote sensing data (land surface temperature (LST), mean elevation, normalized difference vegetation index (NDVI), and normalized difference water index (NDWI)). The output maps identified the hot zones through comprehensive GIS analysis. The resultant maps showed that high HVI exists in three typical areas: (1) areas with dense population and low vegetation, (2) areas with artificial surfaces (built-up areas), and (3) industrial zones. Low-HVI areas are in natural landscapes such as rivers and grasslands. Our analysis can be implemented in other cities to highlight areas at high risk of extreme heat and air pollution.

The Multi-Omics Landscape and Clinical Relevance of the Immunological Signature of Phagocytosis Regulators: Implications for Risk Classification and Frontline Therapies in Skin Cutaneous Melanoma.

Simple SummaryIn this study, we focused on exploring phagocytosis regulators’ expression and mutational characteristics in skin cutaneous melanoma samples and delineating two molecular subtypes based on expression characteristics. We determined the relationship between phagocytosis regulators and survival by survival analysis of molecular subtypes. We then constructed a survival model (PRRS) to further quantify the criteria. Moreover, we combined pathway analysis, immune infiltration analysis, and mutation analysis to deeply explore the effects of phagocytosis regulators on skin cutaneous melanoma samples.Tumor-associated macrophages (TAMs) have gained considerable attention as therapeutic targets. Monoclonal antibody treatments directed against tumor antigens contribute significantly to cancer cell clearance by activating macrophages to phagocytose tumor cells. Due to its complicated genetic and molecular pathways, skin cutaneous melanoma (SKCM) has not yet attained the expected clinical efficacy and prognosis when compared to other skin cancers. Therefore, we chose TAMs as an entrance point. This study aimed to thoroughly assess the dysregulation and regulatory role of phagocytosis regulators in SKCM, as well as to understand their regulatory patterns in SKCM. This study subtyped prognosis-related phagocytosis regulators to investigate prognostic differences between subtypes. Then, we screened prognostic factors and constructed phagocytosis-related scoring models for survival prediction using differentially expressed genes (DEGs) between subtypes. Additionally, we investigated alternative treatment options using chemotherapeutic drug response data and clinical cohort treatment data. We first characterized and generalized phagocytosis regulators in SKCM and extensively examined the tumor immune cell infiltration. We created two phagocytosis regulator-related system (PRRS) phenotypes and derived PRRS scores using a principal component analysis (PCA) technique. We discovered that subtypes with low PRRS scores had a poor prognosis and decreased immune checkpoint-associated gene expression levels. We observed significant therapeutic and clinical improvements in patients with higher PRRS scores. Our findings imply that the PRRS scoring system can be employed as an independent and robust prognostic biomarker, serving as a critical reference point for developing novel immunotherapeutic methods.

Mitigation of Black-Box Attacks on Intrusion Detection Systems-Based ML

Intrusion detection systems (IDS) are a very vital part of network security, as they can be used to protect the network from illegal intrusions and communications. To detect malicious network traffic, several IDS based on machine learning (ML) methods have been developed in the literature. Machine learning models, on the other hand, have recently been proved to be effective, since they are vulnerable to adversarial perturbations, which allows the opponent to crash the system while performing network queries. This motivated us to present a defensive model that uses adversarial training based on generative adversarial networks (GANs) as a defense strategy to offer better protection for the system against adversarial perturbations. The experiment was carried out using random forest as a classifier. In addition, both principal component analysis (PCA) and recursive features elimination (Rfe) techniques were leveraged as a feature selection to diminish the dimensionality of the dataset, and this led to enhancing the performance of the model significantly. The proposal was tested on a realistic and recent public network dataset: CSE-CICIDS2018. The simulation results showed that GAN-based adversarial training enhanced the resilience of the IDS model and mitigated the severity of the black-box attack.

Identification of errors-in-variables ARX models using modified dynamic iterative PCA

Identification of autoregressive models with exogenous input (ARX) is a classical problem in system identification. This article considers the errors-in-variables (EIV) ARX model identification problem, where input measurements are also corrupted with noise. The recently proposed Dynamic Iterative Principal Components Analysis (DIPCA) technique solves the EIV identification problem but is only applicable to white measurement errors. We propose a novel identification algorithm based on a modified DIPCA approach for identifying the EIV-ARX model for single-input, single-output (SISO) systems where the output measurements are corrupted with coloured noise consistent with the ARX model. Most of the existing methods assume important parameters like input-output orders, delay, or noise-variances to be known. This work’s novelty lies in the joint estimation of error variances, process order, delay, and model parameters. The central idea used to obtain all these parameters in a theoretically rigorous manner is based on transforming the lagged measurements using the appropriate error covariance matrix, which is obtained using estimated error variances and model parameters. Simulation studies on two systems are presented to demonstrate the efficacy of the proposed algorithm.

Classification of Dysphonic Voices in Parkinson's Disease with Semi-Supervised Competitive Learning Algorithm.

This article proposes a novel semi-supervised competitive learning (SSCL) algorithm for vocal pattern classifications in Parkinson’s disease (PD). The acoustic parameters of voice records were grouped into the families of jitter, shimmer, harmonic-to-noise, frequency, and nonlinear measures, respectively. The linear correlations were computed within each acoustic parameter family. According to the correlation matrix results, the jitter, shimmer, and harmonic-to-noise parameters presented as highly correlated in terms of Pearson’s correlation coefficients. Then, the principal component analysis (PCA) technique was implemented to eliminate the redundant dimensions of the acoustic parameters for each family. The Mann–Whitney–Wilcoxon hypothesis test was used to evaluate the significant difference of the PCA-projected features between the healthy subjects and PD patients. Eight dominant PCA-projected features were selected based on the eigenvalue threshold criterion and the statistical significance level (p < 0.05) of the hypothesis test. The SSCL algorithm proposed in this paper included the procedures of the competitive prototype seed selection, K-means optimization, and the nearest neighbor classifications. The pattern classification experimental results showed that the proposed SSCL method can provide the excellent diagnostic performances in terms of accuracy (0.838), recall (0.825), specificity (0.85), precision (0.846), F-score (0.835), Matthews correlation coefficient (0.675), area under the receiver operating characteristic curve (0.939), and Kappa coefficient (0.675), which were consistently better than those results of conventional KNN or SVM classifiers.

Influence of Upper and Lower Body Anthropometric Measures on An Aggregate Physical Performance Score in Young Elite Male Soccer Players: A Case Study

Background: The present study aimed to determine the association of anthropometry-based characteristics with an aggregate score (AS) of physical performance in young elite soccer players. Methods: Sixteen under 15 elite players were enrolled. Among numerous anthropometrics variables, upper arm contracted (UACC) and relaxed circumference (UARC), corrected arm muscle area (AMAcorr), arm muscle circumference (AMC), thigh muscle circumference (TMC) and suprapatellar girths were also employed in this study. Players’ physical performance was assessed by the change of direction (COD), 10 m and 20 m sprint, countermovement jump (CMJ) test, sprint with 90∘ turns (with ball), and yo-yo intermittent recovery test level 1 (Yo-Yo IRT1). The AS was computed by Principal Components Analysis technique with one component on normalized performance results. A stepwise regression analysis was conducted to assess potential association between anthropometry-based variables and AS. Results: Large negative correlations (r &lt; –0.68) of AS with UACC, UARC, AMAcorr, and AMC were detected. UACC and TMC permits to accurately estimate AS explaining 60% of the total variance (p &lt; 0.001). Conclusions: These findings demonstrated the importance of including anthropometry-based measures of both upper and lower body to the physical performance potential expressed by AS in elite youth soccer.

Pepper‐rosmarin essential oil (Lippia sidoides Cham.) as an antioxidant additive for PBAT – poly (butylene adipate co‐terephthalate) films and its application for active packaging

SummaryActive antioxidant films additivated with Pepper‐rosmarin essential oil (PRO) and poly (butylene adipate co‐terephthalate) – PBAT were produced from the solvent casting technique in different concentrations (5%, 10%, and 15%) and the antioxidant activity of the film was evaluated by the oxidation of olive oil. Homogenous films were produced and the incorporation of PRO was confirmed by the principal component analysis (PCA) technique. The antioxidant capacity of the PRO analysed indicated high activity. There were no changes in the degradation temperatures of the films by thermogravimetric analysis (TGA). All films have become more permeable with the increase in PRO content. The addition of 15% PRO made the film more resistant. In the olive oil oxidation test, by the PCA technique, an inhibition of oxidation with higher PRO concentrations was observed. Therefore, it is possible to confirm the potential of an active antioxidant packaging of PBAT added with PRO to minimise oxidative effects on food, maintaining its quality.

A Non-Intrusive Approach for Enhancing Power-System Frequency Stability

This paper presents a novel protection strategy for assuring power-system frequency stability. After many years of use, the existing (static) under-frequency load-shedding concept has proven to be reliable and fast. However, it does not adapt well to volatile operating conditions, which might often lead to a system collapse. This developed into a major obstacle, since operational conditions of modern power systems are persistently subjected to fast and unpredictable changes. Therefore, there is a need to introduce a new capability in under-frequency load-shedding that would be able to fine-tune the mismatch in active power in any conditions. From the practical aspect, a systematic implementation of improvements should not threaten system stability. In terms of cost, a minimum intervention in the existing concept is desirable. This is why we developed an effective upgrade of the existing concept by introducing a specialized group of intelligent electronic devices supported by a principal component analysis technique. This group should be operated alongside the existing solution to monitor the actual frequency conditions and to take over power-balancing when fine-tuning is required. This supports the existing load-shedding and improves the process of frequency stabilization.

Detection of Uniform Gearbox Wear in Induction Motors Based on the Analysis of Stray Flux Signals Through Statistical Time-Domain Features and Dimensionality Reduction Techniques

Gearboxes are core elements in power transmission systems. Although gearboxes are reliable and high-efficiency components, the occurrence of different faults is frequent since they are often subjected to adverse operating conditions. Classical gearbox condition monitoring approaches are based on the analysis of vibration and current motor signals and rely on the identification of specific fault-related frequency patterns. In this regard, this article proposes a novel diagnosis methodology based on the analysis of stray flux signals for detecting uniform wear in the gear teeth. The proposed methodology is based on the processing of the stray flux signals through feature calculation and extraction stages that lead to a high-performance signal characterization by estimating a set of statistical time domain-based features and then reducing the dimensionally by means of the principal component analysis and linear discriminant analysis techniques. Additionally, an automatic fault diagnosis is achieved through a neural network-based classifier for the detection and identification of uniform wear in a gearbox. The obtained results prove the potential of the proposal for its incorporation in condition maintenance programs in the industry, becoming an excellent alternative to classical approaches.

Genetic Diversity in Tomato (Solanum lycopersicum L.) based on Quantitative Traits

Present investigation was conducted at Institute of Agricultural Sciences, Banaras Hindu University (BHU), Varanasi, Uttar Pradesh state, India during 2014−15 to assess the genetic diversity among 66 accessions of tomato maintained at Institute of Agricultural Sciences, Banaras Hindu University. The techniques of principal component analysis and D2 analysis were used for genetic diversity assessment. The analysis revealed significance of four principal components representing 70.12% of the total variability. Some of the important yield traits like average fruit weight, fruit yield plant-1 and number of locules fruit-1 were the most important traits in PC1 indicating that selection for diversity must be based on these traits. The correlation plot depicted that the traits plant height, number of fruits cluster-1 and number of fruits plant-1 were closely related to each other. Based on the scores of 1st and 2nd principal component, the genotypes were distributed in biplot. Most of the genotypes were congregated near the centre and revealed narrow genetic diversity existing among the germplasm accessions. The coordinates of the genotype EC 538380 indicated it as the most unique. The entries EC 538155, Pant T3, EC 521069, EC 538434 and EC 168283 were standing separate from the majority of entries. Cluster analysis also indicated congregation of most of the genotypes in one group. Overall, the study indicated limited genetic diversity in the test material.

A Novel Combination of PCA and Machine Learning Techniques to Select the Most Important Factors for Predicting Tunnel Construction Performance

Numerous studies have reported the effective use of artificial intelligence approaches, particularly artificial neural networks (ANNs)-based models, to tackle tunnelling issues. However, having a high number of model inputs increases the running time and related mistakes of ANNs. The principal component analysis (PCA) approach was used in this work to select input factors for predicting tunnel boring machine (TBM) performance, specifically advance rate (AR). A reliable and precise forecast of TBM AR is desirable and critical for mitigating risk throughout the tunnel building phase. The developed PCAs (a total of four PCAs) were used with the artificial bee colony (ABC) method to predict TBM AR. To assess the created PCA-ANN-ABC model’s capabilities, an imperialist competitive algorithm-ANN and regression-based methods for estimating TBM AR were also suggested. To evaluate the artificial intelligence and statistical models, many statistical evaluation metrics were evaluated and generated, including the coefficient of determination (R2). The findings indicate that the PCA-ANN-ABC model (with R2 values of 0.9641 for training and 0.9558 for testing) is capable of predicting AR values with a high degree of accuracy, precision, and flexibility. The modelling approach utilized in this study may be used to other comparable studies involving the solution of engineering challenges.