Significant Influencing Factors Research Articles

Answering the call to action: an empirical analysis of SDG performance in global banks

Purpose Following Agenda 2030, the integration of Sustainable Development Goals (SDGs) into banking strategies has become essential for promoting global sustainability and responsible financial practices. This study aims to examine how the financial characteristics of banks influence their efforts towards achieving the SDGs. Design/methodology/approach A fixed-effect panel regression is used to analyse a sample of 646 publicly listed banks across 61 countries, observed from 2018 to 2023. The data are sourced from the Refinitiv Eikon database, with the dependent variable pertaining to the SDGs calculated as a composite score. Findings Empirical results show that larger and less profitable banks reach better performance in terms of contribution to the SDGs. Instead, the level of leverage does not appear to be a significant influential factor in fostering the support of the SDGs. Originality/value Theoretically, this work enriches the academic literature on SDG performance and confirms the relevance of stakeholder theory. Managerially, this study provides valuable insights for banks integrating the SDGs into their business models and for policymakers identifying strategies to encourage sustainable practices.

Scenario-based simulation training, occupational burnout and psychological capital in hospital-based nursing instructors: A cross-sectional study

AimThis study aimed to scrutinize the extent of engagement in the implementation of scenario-based simulation training within hospital settings and the status of occupational burnout and psychological capital among nursing instructors. MethodsThis research spanned from August 2022 to June 2023. A total of 172 nursing instructors from 67 hospitals participated in an electronic survey, which included knowledge, attitudes, and practices questionnaires related to scenario-based simulation training, the Maslach burnout inventory-general survey for assessing occupational burnout, and the 24-item Psychological Capital Questionnaire. ResultsThe findings revealed that the average scores for knowledge and attitude in scenario-based simulation training among the 172 nursing instructors were (54.85 ± 13.26) and (76.55 ± 28.82), respectively. Out of these, 105 instructors (61.0%) have engaged in scenario-based simulation training, with an average practice score of (64.0 ± 25.4). Psychological capital emerged as a significant influencing factor on the attitudes and practices of nursing instructors in scenario-based simulation training (p = .046, p = .004, respectively). Positive correlations were observed between knowledge, attitude, and practice, while a positive association was noted between occupational burnout and psychological capital. ConclusionNursing instructors exhibited a relatively low proficiency in knowledge acquisition, positive attitudinal disposition, and inadequacy in engagement concerning scenario-based simulation training. The study advocated for the augmentation of standardized training programs tailored specifically for nursing instructors in scenario-based simulation training. Additionally, the study underscored the importance of prioritizing interventions that could enhance the psychological capital of nursing instructors as a pivotal strategy to ameliorate the overall quality of simulation training.

Prediction model for elastic modulus of recycled concrete based on properties of recycled coarse aggregate and cementitious materials

The supplementary cementitious materials (SCMs) are commonly used in recycled coarse aggregate (RA) concrete to improve the performance of RA concrete or reduce the amount of cement used. However, the diverse sources and compositions of RA and SCMs pose a challenge for predicting the elastic modulus of concrete containing both RA and SCMs. In this study, the main influencing factors on the elastic modulus of recycled concrete were analyzed, and the statistical significance of the influencing factors on the elastic modulus was quantitatively analyzed using factor analysis techniques, either individually or in interaction. The results of the factor analysis reveled that the content of recycled coarse aggregates and the reactivity index of the cementitious materials have the most significant impact on the elastic modulus, followed by the water absorption of recycled coarse aggregates and the ratio of aggregates to cementitious materials (A/C). Furthermore, the interaction between A/C ratio and the content of recycled coarse aggregates, as well as the interaction between the water absorption and the content of recycled coarse aggregates, were also found to be significant influencing factors on the elastic modulus. A unified prediction model for elastic modulus was proposed by considering the hydraulic and pozzolanic activities of OPC and SCMs, performance and content of recycled coarse aggregate simultaneously, based on a statistical analysis of the experimental database. The statistical fitting results indicted that the elastic modulus can be accurately predicted using proposed model, thus promoting the application of recycled aggregate and green binder in structures.

Psychological Effects of Noise Exposure on Personnel in Central Sterile Supply Department in China

Psychological Effects of Noise Exposure on Personnel in Central Sterile Supply Department in China

Exploring Disease Manifestations and Influencing Factors in Acute and Chronic Hepatitis B

Scope and Aims: The investigation of disease manifestations and influencing factors in both acute hepatitis B (AHB) and chronic hepatitis B (CHB) remains limited, with varying results. This study aimed to explore the factors influencing AHB and CHB and their disease manifestations within a Ghanaian population, with the goal of developing a control strategy. Methods: A retrospective study was conducted on 569 admitted hepatitis B cases. Demographic data and disease manifestations were compared between AHB and CHB patients. Logistic regression and correlation analyses were employed to identify the factors influencing the progression of the disease. Results: Significant differences were observed between AHB and CHB patients in terms of median age and hospitalization duration. Variations in age, gender, education level, and occupational distributions were statistically significant (P < 0.05) between the two groups. Symptoms such as fever, nausea, polydipsia, palpitation, anicteric presentation, anorexia, and itching were less common in CHB patients (P < 0.05), while abdominal pain, jaundice, and enlarged liver were more frequent (P < 0.05). CHB patients exhibited significantly higher levels of aspartate aminotransferase, viral load, bilirubin, prothrombin time, partial thromboplastin time, HBeAg, albumin, abdominal ultrasound findings, and globin (P < 0.05), while HBsAg and liver function test levels were significantly lower (P < 0.05) in CHB patients compared to AHB patients. Logistic regression identified age, gender, occupation, education level, and hospitalization duration as significant influencing factors. Conclusion: Males and the adult population represented a higher proportion of CHB patients, with a significant association between CHB and elevated clinical and laboratory characteristics. Age, gender, occupation, education level, and hospitalization duration were established as key influencing factors in the progression of hepatitis B.

Coupling Coordination Analysis Between Farmland Transfer and Agricultural Eco-Efficiency in China

The transfer of farmland refers to the process of transferring land contracting and management rights, during which various changes occur in production entities, management behaviors, and management methods. This process fosters positive changes in grain yield and farmers’ income. However, there is no consensus on the ecological impacts of farmland transfer. The objective of this study is to explore the coupling and coordination relationship between the process of farmland transfer and regional agricultural eco-efficiency across 30 provincial units in China. The following conclusions can be drawn from the results: (1) The level of farmland transfer in China exhibits an overall upward trend, primarily influenced by the average farmland area per household in each region and the number of large grain producers, with Heilongjiang consistently maintaining the highest level. (2) The agricultural eco-efficiency value remains relatively stable at approximately 1.1, with high-efficiency areas concentrated in Beijing, Qinghai, Hainan, Shanghai, Fujian, and Guizhou, where lower agricultural carbon emissions are a significant influencing factor. (3) The coupling degree between farmland transfer and agricultural eco-efficiency in various provinces of China is predominantly in a state of adjustment, with the coupling coordination level improving year by year. However, the coupling coordination level does not demonstrate a clear “high-high” or “low-low” synchronous state with grain yield per unit area, indicating that the levels of farmland transfer and agricultural eco-efficiency have not yet established a robust interactive relationship with farmland production efficiency.

초등학교 1학년 아동의 여가시간 활용에 대한 잠재프로파일분석

Objectives The purpose of this study is to identify the latent profile of leisure time utilization among first grade children and to test the differences in its influencing factors and subjective well-being. Methods Using data from the eighth year of the Korean Child Panel (2015), we analyzed latent profiles (LPA) using the Mplus 8.3 program for 927 children in the first grade and their guardians. The R3STEP method was used to verify influencing factors and the DCON method was used to verify differences in subjective well-being. Results ‘Book-reading group’, ‘Going out group’, ‘TV/Media group’ were extracted as potential groups according to leisure time utilization, and gender, simulated controlling parenting behavior, parental warmth parenting behavior, and parental final education were identified as significant influencing factors. In the case of subjective happiness, ‘book-reading group’, ‘going out group’, and ‘TV and media group’ were found to be higher, and there were significant differences between the groups. Conclusions Discussed ways to support leisure activities that can improve children's subjective happiness.

카데바 실습에 참여한 간호대학생의 생명의료윤리의식 영향 요인

Objectives The purpose of this study was to confirm the relationship between consciousness of biomedical ethics, empathy, and occupational values of nursing students who participated in cadaver practice, and to identify factors influencing consciousness of biomedical ethics. Methods This study was a cross-sectional descriptive research study, and data were collected through a structured questionnaire from June 1 to 10, 2024, targeting nursing students who participated in cadaver practice at a local university in J, a total of 121 students. Data collected from was used for analysis. Data were analyzed using SPSS Win 25.0, and descriptive statistics, t-test, ANOVA, Scheffe' test, Pearson correlation, and hierarchial regression analysis were used. Results There was a positive correlation between the consciousness of biomedical ethics, empathy (r=.29, p=.002), and occupational values (r=.22, p=.039) of nursing students who participated in cadaver practice, with multiple As a result of regression analysis, the experience of death of family and relatives (β=-.27, p=.002) and empathy (β=.23, p=.014) were significant influencing factors on nursing students' consciousness of biomedical ethics, and biomedical ethics Ethical awareness was explained by 14%(F=7.06, p=<.001). Conclusions It was confirmed that the empathy of nursing students who participated in cadaver practice was a factor that influenced their awareness of biomedical ethics. In order to improve the consciousness of biomedical ethics of nursing students, it is necessary to develop and apply an educational program that considers empathy.

Performance Verification and Evaluation of Semi-Active Actuator System Using Quarter Car Lab Simulation with RLDA Data

This study outlines a methodology for determining the durability specifications of electronically controlled dampers by examining performance degradation observed on actual driving roads. It identifies areas of performance decline and the primary causes affecting major actuator dampers. Traditionally, the durability performance of automobile parts has been assessed by calculating damage based on load profiles. However, analyzing actual road conditions is essential because the control commands of electronically controlled suspension systems change in real time according to load conditions. Simulations based on Road Load Data Acquisition (RLDA) use statistically independent representative road surfaces to assess damper deterioration performance. Following this analysis, a rig test of the damper is performed to establish the durability specifications for damper actuator products. The primary form of performance degradation observed was a change in the tensile damping force, which was more substantial than the degradation observed on the compression side. Oil leakage and cavitation were identified as significant influencing factors from a Failure Mode and Effects Analysis (FMEA) perspective. The study concludes that additional design research is necessary, focusing on damper oil and leakage, while also considering the control algorithm’s effects in designing electronically controlled dampers.

High-speed rail passenger flow prediction based on crossformer and quantile regression: A deep learning approach assisted by internet search indices

Understanding potential fluctuations in passenger volume is crucial for the operation and management of high-speed rail, especially during peak times. The uncertainty caused by multiple factors is the main obstacle to accurate prediction. To quantify and mitigate the impact of these uncertainties, Internet search indices are utilized as insightful resources to grasp dynamic trends in passenger flow. Leveraging minimum redundancy maximum relevance, we identify the top search index features based on their predictive contribution to high-speed rail passenger flow. A two-level decomposition strategy is then established based on variational modal decomposition to extract significant influencing factors hidden in the Internet index and capture the dynamic uncertainty of passenger flow. By integrating Crossformer with quantile regression, we construct the upper and lower bounds of the prediction interval. Furthermore, the obtained upper and lower bounds are corrected by the error of point prediction, which allows for dynamic adjustment of the prediction intervals width based on fluctuations in uncertainty, thereby refining the precision of the prediction interval. Finally, the developed approaches effectiveness is validated through two real-world experiments, and the experimental results indicate that this method can more accurately capture variations in high-speed rail passenger flow, improving both management and service quality.

Local Neuronal Activity and the Hippocampal Functional Network Can Predict the Recovery of Consciousness in Individuals With Acute Disorders of Consciousness Caused by Neurological Injury.

There is limited research on predicting the recovery of consciousness in patients with acute disorders of consciousness (aDOC). The purpose of this study is to investigate the altered characteristics of the local neuronal activity indicated by the amplitude of low-frequency fluctuations (ALFF) and functional connectivity (FC) of the hippocampus network in patients with aDOC caused by neurological injury and to explore whether these characteristics can predict the recovery of consciousness. Thirty-seven patients with aDOC were included, all of whom completed resting-state functional magnetic resonance imaging (rsfMRI) scans. The patients were divided into two groups based on prognosis of consciousness recovery, 24 patients were in prolonged disorders of consciousness (pDOC) and 13 in emergence from minimally conscious state (eMCS) at 3 months after neurological injury. Univariable and multivariate logistic regression analyses were used to investigate the clinical indicators affecting patients' recovery of consciousness. The ALFF values and FC of the hippocampal network were compared between patients with pDOC and those with eMCS. Additionally, we employed the support vector machine (SVM) method to construct a predictive model for prognosis of consciousness based on the ALFF and FC values of the aforementioned differential brain regions. The accuracy (ACC), area under the curve (AUC), sensitivity, and specificity were used to evaluate the efficacy of the model. The FOUR score at onset and the length of mechanical ventilation (MV) were found to be significant influential factors for patients who recovered to eMCS at 3 months after onset. Patients who improved to eMCS showed significantly increased ALFF values in the right calcarine gyrus, left lingual gyrus, right middle temporal gyrus, and right precuneus compared to patients in a state of pDOC. Furthermore, significant increases in FC values of the hippocampal network were observed in the eMCS group, primarily involving the right lingual gyrus and bilateral precuneus, compared to the pDOC group. The predictive model constructed using ALFF alone or ALFF combined with FC values from the aforementioned brain regions demonstrated high accuracies of 83.78% and 81.08%, respectively, with AUCs of 95% and 94%, sensitivities of 0.92 for both models, and specificities of 0.92 for both models in predicting the recovery of consciousness in patients with aDOC. The present findings demonstrate significant differences in the local ALFF and FC values of the hippocampus network between different prognostic groups of patients with aDOC. The constructed predictive model, which incorporates ALFF and FC values, has the potential to provide valuable insights for clinical decision-making and identifying potential targets for early intervention.

Influential factors for medical students’ classroom concentration—evaluation with speech recognition and face recognition technology

Statement of the problemThe concentration of medical students in the classroom is important in promoting their mastery of knowledge. Multiple teaching characteristics, such as speaking speed, voice volume, and question use, are confirmed to be influential factors.PurposeThis research aims to analyze how teachers’ linguistic characteristics affect medical students’ classroom concentration based on a speech recognition toolkit and face recognition technology.Materials and methodsA speech recognition toolkit, WeNet, is used to recognize sentences during lectures in this study. Face recognition technology (FRT) is used to detect students’ concentration in class. The study involved 80 undergraduate students majoring in stomatology. The classroom videos of 5 class hours in the dental anatomy course were collected in October 2022. A quantitative research methodology is used in this study. Pearson correlation, Spearman correlation and multiple linear regression analyses were used to analyze the impact of time and teachers’ linguistic characteristics on students’ concentration.Results As a result of regression analysis, the explanatory power of the effect of the linguistic characteristics was 7.09% (F = 83.82, P < 0.001), with time, volume and question being significant influencing factors (P < 0.01). The local polynomial smooth of the scatter between the concentration degree and the use of questions with time appears to fluctuate cyclically and suggests a potential inverse relationship between the use of questions and the concentration degree.ConclusionsThe results of this study support the significant positive influence of volume and questioning technique, the negative influence of time, and the insignificant influence of speaking speed and the interval between sentences on students’ concentration. This study also suggested that teachers may adjust their questioning frequency based on their observation of students’ concentration.

Diverse Attitudes Toward AI-Produced Music: A Case Study between Instrument Learners and Non-Instrument Learners in Yangtze River Delta Region

Recently, with the gradual popularization of AI technology in music, AI music generation technology, as a tool, has triggered thoughts about the transformation of traditional music composition while enhancing the innovation and efficiency of music creation. This study aims to explore the different attitudes toward AI-generated music among instrumental and non-instrumental learners in the Yangtze River Delta region of China, so as to increase the knowledge of AI music composition technology. This study uses a questionnaire method to collect quantitative data, with semi-structured interviews as a qualitative supplement. It is found that most people are aware of AI music technology and express acceptance. The style, uniqueness, and quality of AI-generated works, as well as the convergence of the technology on traditional platforms, and copyright issues, are topics of concern. The differential impact of whether or not one has learned a musical instrument on attitudes towards AI music centers on personal preference. The low price of the service was a significant influencing factor for those who embraced the improved AI technology. In addition, instrumental learners placed more emphasis on the professionalism, artistry, and emotionality of AI music, while non-instrumental learners placed more emphasis on the technological convenience and affordability of AI music. This study helps to demonstrate the influence of instrumental learning experiences on people’s attitudes towards AI music generation, and can provide inspiration for further exploration of music education research in cross-cultural contexts. Meanwhile, the emotional relationship and cognitive responses between humans and AI music can be revealed more.

The correlated factors of anxiety and depression among Chinese hospital staff during the COVID-19 local outbreak.

Hospital staff in the COVID-19 local outbreak were facing different situations, their mental status and influencing factors were also different. The aim of this study is to investigate the anxiety and depression of hospital staff and its potential influence factors during the COVID-19 local outbreak. This was a cross-sectional survey based on a hospital with a local outbreak of COVID-19. We collected the demographic characteristics, COVID-19-related issues, self-rating anxiety scale (SAS), and self-rating depression scale (SDS) of participants through an anonymous questionnaire. Factors associated with anxiety and depression were explored through univariate and multivariate analyses. We also constructed nomograms and calibration curves to predict the probability of anxiety and depression. A total of 800 people completed the questionnaire. 239 (29.9%) of them were doctors, 249 (31.1%) of them were nurses and 312 (39.0%) of them were others. There were 173 (21.6%) cases had anxiety, including 36 (20.8%) doctors, 76 (43.9%) nurses and 61 (35.3%) from other occupations and 281 (35.1%) cases had depression, including 64 (22.8%) doctors, 101 (35.9%) nurses, and 116 (41.3%) from other occupations. Nurses had higher SAS and SDS score than doctors and others (F = 17.856, P < .001 and F = 14.376, P < .001). In addition, multivariate analysis found that occupation, education level, health condition, and reduced sleep were significant influences on anxiety and depression. At the same time, reduced income was also significantly associated with anxiety. During the local outbreak of COVID-19, hospital staff still had varying degrees anxiety and depression. Occupation, education level, health condition and reduced sleep were both significant influencing factors for anxiety and depression. The mental state of hospital staff, including nonmedical-related staff should still be taken seriously.

The effects of diet and semiochemical exposure on male Bactrocera dorsalis (Hendel) (Diptera: Tephritidae) metabolic rate at a range of temperatures

The oriental fruit fly, Bactrocera dorsalis (Hendel) (Diptera: Tephritidae), is an invasive species that has rapidly spread across the African continent, endangering the security of agricultural industries. The sterile insect technique (SIT) is being investigated as a viable additional pest management tool to suppress B. dorsalis populations after its successful implementation in other parts of the world. There is evidence to suggest that pre-release nutritional and semiochemical treatments for sterilised males can enhance their competitive performance against wild type males in SIT programs. This study examined how sterilisation, a diet rich in protein (addition of yeast hydrolysate) or containing semiochemicals (methyl eugenol or eugenol) affected the resting metabolic (RMR) of male B. dorsalis at different temperatures (15 – 30 °C), measured using flow-through respirometry. Our results indicated that the negative effect of sterilisation on RMR decreased as temperature increased and that duration of exposure to semiochemicals for 1 to 4 days was not a significant influencing factor on male B. dorsalis RMR. Protein-rich diet increased average RMR, but the difference in RMR between dietary groups decreased as temperature increased. Semiochemical feeding reduced the average RMR in male B. dorsalis. The difference in RMR between males that consumed semiochemical and those that did not increased with as temperature increased.

Utilizing Insects as Bioindicators: An Approximation for Conservation in Urban Lentic Ecosystems in Central Chile.

This study considered using insect families as bioindicators to establish the health status of an ecosystem of lentic bodies. The water quality in urban lentic bodies in the Metropolitan Region, Chile, was evaluated from aquatic insect family assemblages and physicochemical variables for conserving aquatic life. Evaluations were carried out in parallel at four sampling stations of three water bodies (Batuco Wetland, Carén Lagoon, and Chada Reservoir) in a 2-3-year series, spring (2015, 2017, and 2018) and fall (2016 and 2018), with three replicates. Families were randomly sampled and aquatic insects were identified; abundance and richness differences were compared with non-parametric tests. Physicochemical variables were measured using portable multiparametric and laboratory chemical analyses to determine the water quality. A Canonical Correspondence Analysis was applied for insect families and the physical-chemical variable. In order to categorize the health of these water bodies based on aquatic insect composition and abundance, the Family Biotic Index (FBI), British Biological Monitoring Work Party (BMWP), and Stream Invertebrate Grade Number-Average Level (SIGNAL), adapted for Chile by Figueroa et al. (2007), were calculated. The eudominant insect families were Corixidae and Chironomidae in Batuco, Chironomidae and Corixidae in Carén, and Corixidae in Chada. Baetidae was dominant in Carén and Chada. The water bodies were classified in descending order of water quality by Chilean physicochemical standards: Chada > Carén > Batuco. The TSS (total suspended solids), phosphorus, and electrical conductivity were strongly positively correlated and negatively associated with dissolved oxygen. The TSS level was the most significant influential factor. The BMWP value and the SIGNAL differed from the FBI, but the first was more restrictive, contributing to the conservation of these ecosystems. Based on the nitrogen and phosphorus levels in the water bodies, all of them were eutrophic. Given the ecosystem diversity and complexity, studies should delve deeper into wetlands to establish methods that contribute to determining water quality, using insect families as bioindicators and physicochemical variables.

Test of Mechanical Properties of Composite Recycled Concrete After Medium and High Temperature

Compressive strength is one of the important indexes to test the mechanical properties of concrete. The compressive strength of recycled composite concrete after high temperature is studied by single factor experimental analysis and double factor experimental analysis. Combined with variance analysis, the influence of replacement rate of recycled aggregate, volume incorporation amount of polypropylene fiber and volume incorporation amount of nani CaCO3 on compressive strength of recycled concrete after high temperature was systematically studied. The results show that when the replacement rate of recycled aggregate is 50%, the volume incorporation amount of polypropylene fiber is 0.08%, and nani CaCO3 is 1.2%, the compressive strength of composite recycled concrete reaches the maximum value. Nano CaCO3 is the most significant factor affecting the compressive strength of recycled concrete after high temperature. The significant influencing factors are the incorporation amount of polypropylene fiber and the replacement rate of recycled aggregate.

The status and changing trend in the burden of refractive errors in China from 1990 to 2019

Objective: To comprehend the status and changing trend in the burden of refractive errors in China from 1990 to 2019, and to furnish data-driven support for the formulation of rational strategies in refractive error prevention and control. Methods: Using the Global Burden of Disease Database 2019, data on refractive error prevalence and years lived with disability (YLD) in China from 1990 to 2019 were extracted and their standardized rates were calculated. The trend of refractive error prevalence, age, period, and gender was analyzed using joinpoint regression models. Results: From 1990 to 2019, the number of cases, prevalence rate, and YLD rate of refractive errors in China showed an upward trend, which was consistent with global trends. However, the magnitude of the increase in these parameters was higher than the global average. Conversely, the age-standardized prevalence rate and age-standardized YLD rate demonstrated a declining trend, which was also consistent with global trends, but the reduction rate was less pronounced compared to global levels. Notably, all indicators for females, including prevalence and YLD, exceeded those of males. The prevalence and YLD rates of refractive errors in China increased with age. The older individuals had higher rates compared to younger individuals. In 1990, the highest prevalence and YLD rates were found in the ≥70 years group, while the lowest rates were in the <5 years group. The age group with the most substantial increase in the burden of refractive errors was the 15 to 49 years population, which remained consistent during the three decades. The joinpoint regression analysis results revealed a stage-wise fluctuation in the age-standardized prevalence and YLD rates of refractive errors in China from 1990 to 2019. Specifically, the age-standardized prevalence rate increased by 0.765% from 1990 to 2001, decreased by 0.963% from 2001 to 2014, increased by 4.214% from 2014 to 2017, and decreased by 3.029% from 2017 to 2019. In contrast, the age-standardized YLD rate decreased by 0.113% from 1990 to 1995, increased by 2.252% from 1995 to 2000, decreased by 1.102% from 2000 to 2014, increased by 4.326% from 2014 to 2017, and decreased by 5.090% from 2017 to 2019. Conclusions: The standardized prevalence of refractive errors and the standardized YLD rate in China showed a downward trend, which was less than the global average level, but the disease burden was still heavy. Gender and age emerged as significant influencing factors for the prevalence of refractive errors and YLD. It is imperative to enhance efforts in the prevention and treatment of refractive errors among the middle-aged and elderly population. Attention should also be directed towards the increasing burden in the<50 years population.

Estimation of compressive strength of concrete with manufactured sand and natural sand using interpretable artificial intelligence

To satisfy the design strength of manufactured sand concrete (MSC) in practical engineering applications, a plethora of geotechnical tests are frequently conducted. An effective approach is imperative to reduce the consumption of labor and resources during these tests. The objective of this paper is to introduce an interpretable machine learning (ML) method to evaluate the compressive strength (CS) of MSC. Firstly, a dataset was established by compiling experimental results from 208 literatures. 3382 data points were selected from the dataset for algorithm training. Recursive Feature Elimination with Cross-Validation (RFECV) was employed to select input parameters. Four algorithms with 12 selected input variables and 1 output variable were constructed to predict CS of MSC using Random Forest (RF), Gradient Boosting Decision Trees (GBDT), eXtreme Gradient Boosting algorithm (XGBoost), and Categorical Boosting (CatBoost). The results show that XGBoost has the highest accuracy and generalization ability (R2=0.934, MAE=3.44, RMSE=5.16, MAPE=0.07). To enhance model transparency, SHapley Additive exPlanations (SHAP) was adopted to explain the underlying predictive mechanisms of ML models. Analyses show that, 1) the cement content, curing time, and water content were the main feature parameters influencing the CS of MSC, 2) the cement content, curing time, and water content have a linear increase, logarithmic increase, and exponential decrease with the CS of MSC, respectively. Partial Dependence Plots (PDP) and Individual Conditional Expectation (ICE) plots were used to further analyze the impacts of these significant influencing factors on the CS of MSC. Additionally, the Local Interpretable Model-Agnostic Explanations (LIME) method was employed to investigate thresholds for various material dosages in MSC containing 5–10 % stone powder. Two typical scenarios were selected for analysis, yielding recommended dosage ranges for concrete of two distinct strengths. Finally, a graphical user interface (GUI) for the CS of MSC has been designed, which might be of great use to material engineers. This provides reference and guidance for concrete engineering practice.

Experimental Study on Proportion Optimization of Rock-like Materials Based on Genetic Algorithm Inversion.

It is very important to clarify the optimization method of the rock-like material ratio for accurately characterizing mechanical properties similar to the original rock. In order to explore the optimal ratio of rock-like materials in gneissic granite, the water-paste ratio, iron powder content and coarse sand content were selected as the influencing factors of the ratio. An orthogonal test design and sensitivity analysis of variance were used to obtain the significant influencing factors of the ratio factors on seven macroscopic mechanical parameters, including compressive strength σc, tensile strength σt, shear strength τf, elastic modulus E, Poisson's ratio ν, internal friction angle φ and cohesion c. A multivariate linear regression equation was constructed to obtain the quantitative relationship between the significant ratio factors and the macroscopic mechanical parameters. Finally, a rock-like material ratio optimization program based on genetic algorithm inversion was written. The results show that the water-paste ratio had extremely significant effects on σc, σt, τf, E, ν and c. The iron powder content had a highly significant effect on σc, σt, τf and c, and it had a significant effect on ν and φ. Coarse sand content had a significant effect on σc, E and c. The multiple linear regression model has good reliability after testing, which can provide theoretical support for predicting the macroscopic mechanical parameters of rock-like materials to a certain extent. After testing, the ratio optimization program works well. When the water-paste ratio is 0.5325, the iron powder content is 3.975% and the coarse sand content is 15.967%, it is the optimal ratio of rock-like materials.