Actual Response Research Articles

Nonlinear Radiative Response to Patterned Global Warming due to Convection Aggregation and Nonlinear Tropical Dynamics

Abstract Climate responses to global warming exhibit large dependence on the spatial pattern of sea surface temperature (SST) changes. A Green’s function (GF) approach—predicting the global climate response to a complex SST pattern change as the linear superposition of the response to finite-area SST perturbations evaluated in isolation—has been used to systematically evaluate and understand the top of atmosphere (TOA) radiation response to patterned warming. However, in general, the linear GF approach fails for TOA radiation reconstruction under future global warming scenarios in coupled models. Here, we show that the linear superposition of global mean TOA radiation responses to large SST warming perturbations at multiple patches (the GF approach prediction) overestimates the actual response to simultaneous multipatch perturbation. This is because the linear superposition overestimates tropical large-scale convection aggregation strengthening upon localized heating that enhances longwave radiative cooling, which is further explained by the overestimation of circulation response and associated horizontal water vapor transport. The nonadditivity of TOA radiation response is caused by the nonadditivity of convection aggregation, ultimately rooted in nonlinear tropical dynamics. We also demonstrate that the prediction error of the GF approach grows with decreasing patch size (equivalently, increasing patch number). We conclude that using the GF approach to predict future climate change could overestimate longwave radiative cooling and underestimate (effective) climate sensitivity. Numerical experiments may be used to identify specific perturbation patterns where the errors are smaller than the signal. Our research also highlights that an increase in the degree of large-scale convection aggregation has a nonlinear and negative feedback for mean warming.

CT radiomics to predict pathologic complete response after neoadjuvant immunotherapy plus chemoradiotherapy in locally advanced esophageal squamous cell carcinoma.

To develop and validate a CT-based radiomics model to predict pathologic complete response (pCR) after neoadjuvant immunotherapy plus chemoradiotherapy (NICRT) in locally advanced esophageal squamous cell carcinoma (ESCC). A total of 105 patients with locally advanced ESCC receiving NICRT from February 2019 to December 2023 were enrolled. Patients were randomly divided into the training cohort and the test cohort at a 3:1 ratio. Enhanced CT scans were obtained before NICRT treatment. The 2D and 3D regions of interest were segmented, and features were extracted, followed by feature selection. Six algorithms were applied to construct the radiomics and clinical models. These models were evaluated by area under curve (AUC), accuracy, sensitivity, and specificity, and their respective optimal algorithms were further compared. Forty-eight patients (45.75%) achieved pCR after NICRT. The AUC values of three algorithms in 2D radiomics models were higher than those in the 3D radiomics model and clinical model. Among these, the 2D radiomics model based on eXtreme Gradient Boosting (XGBoost) exhibited the best performance, with an AUC of 0.89 (95% CI, 0.81-0.97), accuracy of 0.85, sensitivity of 0.86, and specificity of 0.84 in the training cohort, and an AUC of 0.80 (95% CI, 0.64-0.97), accuracy of 0.77, sensitivity of 0.84, and specificity of 0.69 in the test cohort. Calibration curves also showed good agreement between predicted and actual response, and the decision curve analysis further confirmed its clinical applicability. The 2D radiomics model can effectively predict pCR to NICRT in locally advanced ESCC. Question Can CT-based radiomics predict pathologic complete response (pCR) after neoadjuvant immunotherapy plus chemoradiotherapy (NICRT) in locally advanced esophageal squamous cell carcinoma (ESCC)? Findings The model based on eXtreme Gradient Boosting (XGBoost) performed best, with an AUC of 0.89 in the training and 0.80 in the test cohort. Clinical relevance This CT-based radiomics model exhibits promising performance for predicting pCR to NICRT in locally advanced ESCC, which may be valuable in personalized treatment plan optimization.

‘Give us the numbers!’: sovereign capture and the undermining of shared situational awareness in the UK’s COVID-19 response

ABSTRACT Shared situational awareness is advanced as a solution to the problem of collaborative decision-making in multi-agency emergency responses. By harnessing information and communication technologies to construct and distribute a common understanding of the event, shared situational awareness promises to optimise, distribute and devolve decision-making, thus improving the overall speed and efficiency of emergency responses. However, in the context of actual emergencies shared situational awareness is found to routinely breakdown. This paper draws upon 41 interviews conducted with emergency responders on their experience of using ResilienceDirect to manage shared situational awareness in the UK COVID-19 response. We argue that while shared situational awareness attempts to technologise the problem of sovereign decision-making in emergencies, UK emergency response frameworks contain a latent potential for sovereign capture. As the COVID-19 response demonstrates, an atmosphere of distrust; desire for secrecy and control; and the ubiquity of hierarchies throughout UK Civil Contingencies permitted the reactualisation of sovereign modes of control to frustrate and displace bottom-up modes of emergency organisation meant to guide UK emergency responses. Elucidating the impediments to achieving shared situational awareness within actual response raises urgent questions regarding the robustness of shared situational awareness to political interference within major emergency events.

Modern alternative research methods in genetic toxicology (literature review)

The review represents the current principles of assessment of chemicals genotoxicity. The main attention is paid to alternative research methods. The international experience of the application of alternative approaches and prospects of their use for regulatory purposes are discussed. The data for this review were collected from the Russian and foreign literature, as well as Internet resources, concerning the development of the new alternative methods for testing chemicals for genotoxicity. The OECD database, Scopus, Medline, Google Scholar, RISC, CyberLeninka were used for the information retrieval. Although the evaluation of genotoxicity of chemical substances is the well-established and based on the battery of validated methods, the studies for improving the existing tests and developing new technologies, including the alternative approaches, continue unabated up to now. In general, three trends of development of genetic toxicology can be outlined, including creating of new methods based on the whole-genome sequencing and the application of genome editing technologies; implementation of quantitative system of effects assessment in addition to the existing qualitative approach (mutagenic/non-mutagenic) and testing of various combinations of genotoxicity evaluation methods to identify a battery of tests with a greater predictive activity regarding carcinogenic effects. To use the developed alternative models for regulatory purposes, it is necessary to provide convincing evidence that the data obtained are good predictors of the organism’s actual response to the effects of toxicants/genotoxicants, validation of methods, standardization, and harmonization of research protocols, and changes to the existing regulatory framework are required.

Attention-based hybrid network for structural nonlinear response prediction under long-period earthquake

To develop structures that can respond to severe seismic events, models and tools are essential for quickly and precisely assessing and controlling the structure's performance. Long-period ground motions may pose potentially significant hazards to the safety of high-rise buildings located in basins or alluvial plains. However, traditional regression-based methods and pure data-driven machine learning methods are insufficient to address this problem of response prediction, which is characterized by non-stationarity, nonlinearity, and scarcity of samples. To tackle this issue, we propose a novel semi-explicit physics-informed deep learning method. The method incorporates the physical information into the model, thereby achieving better prediction accuracy. Through numerical and experimental validations, including hysteresis experimental data, the actual seismic response record of a six-story hotel, as well as data from single-degree-of-freedom structures across various periods and multi-story buildings in Los Angeles, Seattle, and Boston, we verified the proposed method significantly outperforms traditional regression-based methods and data-driven machine learning methods in terms of accuracy, efficiency, and robustness.

Optimal iterative learning PI controller for SISO and MIMO processes with machine learning validation for performance prediction

The multivariable process plays a significant role in industrial applications, and designing a controller for the Multi-Input Multi-Output process is challenging due to dynamic process changes and interactions between system variables. Traditionally, the PI family of controllers has been used for its simple design, easy tuning, and quick deployment. However, these processes require complex control actions due to multiple loops in process plants. Thus, this paper proposes an Iterative Learning Controller Dead-time compensating PI, which utilizes the newly developed hybrid Simulated Annealing-Ant Lion Optimization algorithm for Single-Input Single-Output process simulation and real-time experimentation on the Quadruple Tank System. To validate the effectiveness of the developed controller, Machine Learning techniques such as regression and ensemble trees are used to accurately predict the actual system response using error values from respective processes. The simulation and experimental results demonstrate that the proposed controller achieved better performance. The regression and ensemble tree algorithm models effectively predicted the actual response. The obtained data shows that the proposed controller improved system stability and robustness by minimizing nearly half of the overshoot and improving settling time, with an average of 29.9596% faster than the other controller in the SISO process and 14.6116% in the MIMO process.

Optimizing PID controller parameters for robust automatic voltage regulator system through indirect design approach-2

Automatic voltage regulators (AVR) are designed to manipulate a synchronous generator’s voltage level automatically. Proportional integral derivative (PID) controllers are typically used in AVR systems to regulate voltage. Although advanced PID tuning methods have been proposed, the actual voltage response differs from the theoretical predictions due to modeling errors and system uncertainties. This requires continuous fine tuning of the PID parameters. However, manual adjustment of these parameters can compromise the stability and robustness of the AVR system. This study focuses on the online self-tuning of PID controllers called indirect design approach-2 (IDA-2) in AVR systems while preserving robustness. In particular, we indirectly tune the PID controller by shifting the frequency response. The new PID parameters depend on the frequency-shifting constant and the previously optimized PID parameters. Adjusting the frequency-shifting constant modifies all the PID parameters simultaneously, thereby improving the control performance and robustness. We evaluate the robustness of the proposed online PID tuning method by comparing the gain margins (GMs) and phase margins (PMs) with previously optimized PID parameters during parameter uncertainties. The proposed method is further evaluated in terms of disturbance rejection, measurement noise, and frequency response analysis during parameter uncertainty calculations against existing methods. Simulations show that the proposed method significantly improves the robustness of the controller in the AVR system. In summary, online self-tuning enables automated PID parameter adjustment in an AVR system, while maintaining stability and robustness.

Validation of AISI Design of Cold-Formed Steel Beams Using Non-Linear Finite Element Analysis

In the local building industry of Pakistan, pre-engineered steel building manufacturers mainly employ their own self-developed software and Excel sheets. These systems are based on empirical formulas mentioned in the AISI manual. Under this scenario, a need was found to validate AISI flow charts using commercial software like CUFSM 5.04 and ABAQUS R2019x. This study presents a validation of the CUFSM software and the American Iron and Steel Institute (AISI) Direct Strength Method (DSM) results of channel section flexural members using the non-linear finite element method employing ABAQUS. In this study, eight standard cold-formed channel-section (C-section) steel members were modeled and analyzed using ABAQUS to simulate realistic behavior under four-point loading conditions. The non-linear finite element models incorporated material and geometric non-linearities to capture the actual response of the steel elements. The results obtained from ABAQUS were compared with those predicted by the CUFSM and DSM, focusing on critical parameters such as nominal strength, buckling modes, and deformation patterns. During this study, it was observed that out of the selected sections, the AISI charts predict conservative and even unsafe flexural capacities in some of the cases concerning other methods, with a maximum difference of 14.03%. The differences obtained using DSM and ABAQUS when compared with the results of the AISI charts varies on both the plus and minus sides. This study will not only affect the industry in terms of innovative designs for efficient structures but also the community in regards to low-budget construction.

Optimization of nano-filler and silane treatment on mechanical performance of nanographene hybrid composites using RSM and ANN technique

This study utilized response surface methodology (RSM) and artificial neural networks (ANN) to optimize the formulation of polymer composites with three key variables: silane concentration, silane dipping time, and the number of nanoparticles. RSM and analysis of variance were employed to explore the effects of these variables on the composite’s mechanical properties, while ANN was used for analyzing complex interactions between parameters. The results reveal a significant correlation between the predicted and actual response values derived from the models. Notably, silane treatment of fiber was identified as the primary factor influencing the composite’s flexural strength. The study found that all fiber-related properties substantially affected both flexural strength and hardness, with silane dipping time being the most critical factor. Additionally, incorporating nanoparticles improved the fiber matrix’s strength by enhancing agglomeration. ANN achieved a 95% accuracy in predicting flexural strength and hardness. Comparisons among experimental data, the regression model, and ANN confirmed the robustness of the ANN predictions. The optimal composition identified for flexural strength was 5 wt % nanoparticles, 10 wt % silane treatment, and a 20-min dipping time. For hardness, the ideal formulation was 5 wt % nanoparticles, 15% silane treatment, and a 20-min dipping time in silane. This comprehensive approach offers a refined method for developing high-performance polymer composites with tailored properties.

A UAV-based method for efficiently measuring the spatial pattern of multiple air pollutants in street canyons

Street canyons are fundamental units of urban spaces where individuals engage in social activities. The air quality within street canyons directly impacts the health and behavior of residents. Therefore, it is crucial to comprehensively assess and depict the actual distribution of multiple air pollutants (SO2, CO, NO2, O3, PM2.5, and PM10) in street canyons, for both investigation and validation purposes. However, an efficient and accurate method for doing so is rarely available. To address this issue, indoor and field measurements were conducted to determine the response times of these pollutants in various polluted environments, identify the optimal stay duration at individual sampling points, and optimize the sampling layout. A method based on Unmanned Aerial Vehicle (UAV) was developed to efficiently measure the spatial pattern of pollutants in street canyons and verified through field application. The results showed that the actual response times of pollutants in heavily and slightly polluted environments were less than 30 s. The recommended stay duration at an individual sampling point in an outdoor environment was determined to be 2 min, with uncertainty of 2 %. For the sampling layout, it is recommended to take 12 samples to obtain a spatial pattern of a street canyon. These samplings should be arranged at the air inlet, middle, and air outlet along the middle of the street canyon, as well as at their bottom, 20 %–30 %, 50 %–60 %, and top in the vertical direction, respectively. The UAV-based method can efficiently complete the sampling process in just 24 min.

Characterization of inter-panel connections in CLT floors using finite element model updating

Large cross-laminated timber (CLT) floors are inevitably composed of multiple CLT panels assembled on the construction site. Due to the sheer lack of adequate numerical characterization of the inter-panel connections, CLT floors are normally modelled as monolithic structures or independent “piano keyboards”, yielding errors between simulated and the actual vibration response. This paper presents a benchmark study for a deterministic finite element (FE) model updating of so called “an equivalent elastic strip” representation of the inter-panel connections. Fitting modelling parameters of the strip, i.e. its width and material properties, featured FE modelling and modal testing of two full-scale CLT floors composed of one and two panels connected with a half-lapped joint. The floors were made of (almost) identical panels and placed on elastic mats along their shorter edges. First, the single-panel floor was used to fit material properties of the timber and the boundary conditions. Then, the FE model updating (FEMU) of two-panel floor yielded the actual values of the strip modelling parameters. Finally, these were successfully verified by direct comparison between numerically calculated and experimentally measured modal properties of a full-scale model of a simply supported three-panel floor. The results implied that the rotational stiffness of the inter-panel connection depends strongly on its vertical vibration displacement. Consequently, values of the rotational stiffness derived from rare static tests published so far cannot be used reliably in a much subtler vibration response analysis. Future studies should feature other types of the inter-panel connections in various floor layouts and a probabilistic FEMU.

Responsiveness and Reliability of a Sipping Device to Measure Motivation in Normal-Weight Individuals and Bariatric Surgery Patients.

There is an urgent need to measure the motivation to taste a sweet fluid in order to determine the influence of sweet tastes on the potential choices and consumption of beverages in patients with obesity. Current methods utilize either survey instruments or arbitrary operant tasks. The sipometer enables the participant to utilize an actual ingestive behavioral response to measure motivation during access to beverages on either ad libitum (AL) or progressive time ratio (PR) schedules. We determined the sipometer's responsiveness and reliability as a test of change in motivation for sweet tastes after bariatric surgery. Participants (58 patients and 28 controls, BMI: 18.5-24.9 kg/m2) sham-consumed an aspartame-sweetened (S) and non-sweetened (N) beverage under AL and PR schedules at a pre-surgery/baseline and a 3-month and 24-month visit (patients only). Cumulative pressure (CumPres), a measure of effort, was the sum of the pressures exerted during sipping under each condition. Baseline CumPres for PRS was higher than ALS and ALN in patients (p < 0.03) and higher than PRN in controls (p = 0.009). At 3 months, CumPres did not differ amongst conditions in patients, but CumPres for PRS was higher than all other conditions in controls (p < 0.0005). There were no baseline group differences; however, patients' CumPres for PRS was lower than controls' at 3 months (p = 0.002). Patients' CumPres for PRS decreased non-significantly between the baseline and 3 months but increased at 24 months compared to 3 months (p = 0.025) and was no different from baseline. Controls' CumPres for PRS increased at 3 months (p = 0.0359), but CumPres for all conditions was correlated between visits (p's < 0.038). The sipometer is a reliable and sensitive measure of motivation to consume sweet beverages and may reflect changes in post-operative energy intake.

Study of vibration patterns and response transfer relationships in walnut tree trunks

Fruit tree trunk is the support structure of the whole tree, carrying the task of vibration transmission to all levels of branches, fruit tree trunk vibration response to the design and optimization of the vibration harvesting device to provide guiding value. In this paper, the vibration response and harvesting test was carried out for four walnut fruit trees with different morphology and size, and the vibration transmission relationship between the excitation device and the fruit tree as well as inside the fruit tree was analyzed and studied according to the acceleration response curves at each position point under different frequencies. At the same time, the theoretical vibration pattern of the fruit tree and the measured spatial three-dimensional vibration pattern were constructed, and the response transfer law between each point of the fruit tree was further analyzed in terms of the vibration pattern and direction of each position point on the tree. The results show that the excitation response acceleration is basically no attenuation inside the device, but there is some attenuation in the clamping part due to the presence of damping pads. The vibration response in the fruit tree trunk is greater in amplitude farther from the excitation point. The phase relationship between each position point is basically the same in the theoretical vibration mode and the measured vibration mode, but there is damping in the actual fruit tree, which leads to the accumulation of phase hysteresis in the transmission of excitation parameters. The actual response trajectory is a narrow ellipse, and the different position points show different rotational relationships along the ellipse trajectory due to the growth of the fruit tree.

Physics-informed neural network classification framework for reliability analysis

Reliability analysis is crucial for quantitatively evaluating structural safety amidst uncertainties, laying the foundation for reliability-based design optimization aimed at augmenting structural reliability and reducing economic expenditures, thereby offering substantial engineering benefits. However, performing reliability analysis on intricate structures, especially those requiring time-intensive finite element models, presents a formidable challenge. This study develops an innovative physics-informed neural network classification (PINNC) model to tackle this issue. In the PINNC model, the loss associated with the structural output state (i.e., safety or failed state) is defined as classification loss. Additionally, the structural output value (i.e., actual structural response) is treated as critical physical information, with its associated loss termed as physical loss. To facilitate a separate calculation of physical loss and classification loss, a parametric sigmoid activation function is used, establishing a link between the structural output value and the structural output state. The total loss is calculated as a weighted sum of both physical and classification losses. Unlike conventional neural network classification models that solely focus on classification loss, the PINNC model integrates both physical and classification losses, markedly improving the model’s efficacy in structural reliability analysis. Moreover, an adaptive framework is developed to incrementally include samples proximal to the limit state surface as new training data, thereby reinforcing the PINNC model’s computational precision. The effectiveness of the proposed PINNC framework in overcoming reliability analysis challenges is demonstrated through various applications.

Response of seed vigor to experimental warming in a double rice cropping system

Climate warming affects rice seed vigor during ripening, which plays a crucial role in seed quality. However, the actual response of rice seed vigor to warming is still unclear. In this study, seeds after warming treatment in a double rice cropping system were used to determine seed vigor and related physiological traits during germination. Warming treatment significantly improved the germination index (GI), seed vigor index (VI), and seedling dry weight (SDW) for the late-season rice seeds but had no effect on hull thickness, grain weight, and starch and protein contents for both early- and late-season rice seeds, and these parameters were highly associated with germination rate, GI, VI, and SDW. Warming treatment increased gibberellin (GA) content and α-amylase and β-amylase activities in endosperm and coleoptile in both seasons during the later stage of germination, reaching a significant level on the 7th d. Moreover, indole-3-acetic acid (IAA) content was consistently increased in the coleoptile but decreased in the endosperm in response to warming, and warming did not affect zeatin content. These results suggest that future global warming will improve rice seed vigor by regulating the synthesis of endogenous hormones and amylases, especially in the late-season rice.

Probabilistic evaluation method for the stability of large underground cavern considering the uncertainty of rock mass mechanical parameters: A case study of Baihetan underground powerhouse project

For deep large underground cavern projects under complex geological conditions, the determination of rock mass mechanical parameters and stability analysis is fraught with a great deal of uncertainty. This paper proposes a set of methods for estimating rock mass mechanical parameters and probabilistic stability assessment suitable for construction characteristics of large underground caverns. On the one hand, according to the Hoek-Brown criterion and Monte Carlo simulation, a dynamic estimation method for rock mass mechanical parameters is proposed by using Rock Mass Rating (RMR), the uniaxial compressive strength (UCS) of intact rock, and the material constant (mb) as input parameters. On the other hand, considering the uncertainty of rock mass mechanical parameters, a probabilistic evaluation method applicable to the unloading response characteristics of rock mass (including displacement and fracturing zones) around the large cavern is put forward, combining the point estimation method and numerical simulation. The proposed methods were applied to an underground powerhouse, with an excavation span of 34 m and height of 88.7 m, at the Baihetan hydropower station in the southwest of China. Based on extensive field investigations and tests, the probability distributions of key rock mass mechanical parameters were obtained. Furthermore, the high sensitivity of input parameters in the H-B criterion for estimating rock mass mechanical parameters were revealed. Through dynamic simulation, the probability distributions of the displacement and fracturing zones in surrounding rock during the layer-by-layer excavation of the cavern were presented. Comprehensive on-site tests showed that the simulated results were in basic agreement with the actual unloading response behavior during the cavern excavation, validating the correctness and applicability of the proposed methods. The study provides a relatively comprehensive approach for estimating rock mass mechanical parameters and stability evaluation in similar underground engineering projects, and also has guiding significance for predicting and preventing engineering rock mass hazards.

Data simulation of the impact of ball strikes on the bottom of electric vehicle battery packs based on finite element analysis

As a key component, electric vehicle battery packs may suffer serious consequences from external impacts, but there is currently a lack of comprehensive data and simulation studies to evaluate the impact resistance of battery pack structures. Therefore, this article establishes a three-dimensional model of the electric vehicle battery pack through finite element analysis method, taking into account the elastic and geometric nonlinear characteristics of the material. In the ball hitting simulation, appropriate impact dynamic loads were selected to simulate the actual response and deformation of the battery pack bottom under impact, in order to ensure the accuracy and reliability of the simulation results. By analyzing the simulation results, the deformation, stress, and strain distribution at the bottom of the battery pack under ball impact were obtained, as well as the related variation patterns. It was observed that the battery pack underwent significant deformation under impact load, and stress concentration also occurred in certain areas. By analyzing the stress and strain distribution obtained, the weak areas of the battery pack can be identified, and corresponding improvement measures can be proposed to enhance its impact resistance and structural strength. These simulation results also provide reference for the optimization design of the battery pack, helping to further improve its overall working performance.

Multi-Fidelity Physics-Informed Generative Adversarial Network for Solving Partial Differential Equations

Abstract We propose a novel method for solving partial differential equations using multi-fidelity physics-informed generative adversarial networks. Our approach incorporates physics supervision into the adversarial optimization process to guide the learning of the generator and discriminator models. The generator has two components: one that approximates the low-fidelity response of the input and another that combines the input and low-fidelity response to generate an approximation of high-fidelity responses. The discriminator identifies whether the input–output pairs accord not only with the actual high-fidelity response distribution, but also with physics. The effectiveness of the proposed method is demonstrated through numerical examples and compared to existing methods.

Consumer acceptance of robotic surgeons in health services

Over the course of the preceding half-century, specific advancements in robotic technologies have been assimilated into the continuum of human life in a seamless manner, demonstrating the profound impact of robotics on society. Nevertheless, the actual consumer response to service robots remains a subject of ongoing research, with scant attention paid to it thus far. This study focuses on the potential factors influencing the acceptance of robots in the realm of health services. Specifically, a model has been developed to explain individuals’ willingness to use a robot as a surgeon to correct visual impairment in one eye. The Partial Least Squares Structural Equation Modeling (PLSc-SEM) technique is deployed to validate the proposed hypotheses. The model proposed exhibits a robust explanatory power concerning the intention to utilize the robot surgeon, as evidenced by a high R-squared value of 0.817. The findings show the influence of effort expectancy, performance expectancy, social influence, and perceived risk on the intention to adopt robot services. However, the emotional dimensions, specifically pleasure and arousal, were not observed to exert any significant impact on the intention to employ the proposed robot surgeon. The proposed and tested model serves as a roadmap for future research and holds significant practical implications for the industry, paving the way for a more robot-friendly future in health services.

Investigating and predicting tribological characteristics of AZ31 alloy composites reinforced with nano-Al2O3 and micro-Sn particles: a comparative analysis using CCD-RSM and ANN models

In this research, the central composite-based response surface methodology was adopted to select the dominant optimal input factors on wear behaviour and coefficient of friction of an AZ31-microtin/2 wt% nano-Al2O3 composite prepared through a stir casting process with different wt% of Sn. The input factors, such as wt% of Sn reinforcement, sliding distance, sliding speed, and applied load, were selected to determine their significant effects on the coefficient of friction and wear behaviour with 30 trial runs. Analysis of variance (ANOVA) results indicated that Sn reinforcement plays a significant role in the wear behaviour of the nanocomposites, followed by applied load and sliding distance. In addition, an enhancement in wear resistance was witnessed by the addition of Sn reinforcement with AZ31/nano-Al2O3 composites. The optimal process parameters as per the desirability approach were found to be a weight percentage of Sn: 8%, load: 20 N, sliding speed: 2 m s−1, and sliding distance: 1000 m. According to the ANN results, the predicted data is perfectly acceptable with the actual experimental response value. The R values for the training, validation, and testing phases are 0.96166, 0.96801, and 0.98914 for COF, and 0.97688, 0.99247, and 0.99331 for wear rate, indicating a robust correlation between predicted and actual values. The worn-out pin samples were used to examine the worn surface morphology and analyze the wear mechanism.