Scholarlink

<p>The construction industry remains one of the most hazardous industries, where vulnerable workers often face severe accidents. The study aims to bridge the gap of research specific to the country by pursuing two key. Objectives (1) To validate the relationships among previously identified factors and assess their collective impact on Occupational Health and Safety (OHS), and (2) To rank these factors and evaluate their relative significance. For achieving the research objectives, data were gathered using purposive/judgmental sampling, which involved groups of academics and industry professionals, as well as snowball sampling. Data were collected from 570 respondents across five zones of India, with 561 valid responses used for analysis. The methodological novelty lies in combining a two-stage analytical framework that integrates Partial Least Squares Structural Equation Modelling (PLS-SEM), and Artificial Neural Network (ANN) techniques. First, PLS-SEM are employed to validate the measurement and structural relationships among key determinants and to test the model’s hypotheses. Second, ANN is applied as a nonlinear predictive tool to rank the relative importance of significant predictors. This hybrid PLS-SEM–ANN approach overcomes the limitations of linear-only modelling by capturing both theoretical causality and nonlinear predictive strength, thereby enhancing robustness and decision-making accuracy. The structural model explained 43.63% of the variance (R² = 0.4363) in Overall Impact on Health and Safety (OIHS). The ANN model achieved high prediction accuracy with R² = 0.85 and low RMSE values (Training RMSE = 0.109; Testing RMSE = 0.107). The model was validated as a good fit based on both the PLS-SEM and ANN approaches. Among the Governmental Regulatory Factor (GRF) was identified as the most influential in improving H&amp;S in the Indian construction industry, followed by the Organizational Regulatory Factor (ORF). The Worker’s Self-Protective Factor (WSPF) and Job Factor (JF). This research provides practical guidance for creating a safer and healthier construction environment in India.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Data-driven prioritization of occupational hazards: a neural network model for Indian construction sector

  • 期刊论文

  • 发表日期：

    2026年4月13日

查看全文
  • Deep Shaileshkumar Upadhyaya,
  • MohammedShakil S. Malek,
  • Piyushkumar Jayantilal Patel

摘要

The construction industry remains one of the most hazardous industries, where vulnerable workers often face severe accidents. The study aims to bridge the gap of research specific to the country by pursuing two key. Objectives (1) To validate the relationships among previously identified factors and assess their collective impact on Occupational Health and Safety (OHS), and (2) To rank these factors and evaluate their relative significance. For achieving the research objectives, data were gathered using purposive/judgmental sampling, which involved groups of academics and industry professionals, as well as snowball sampling. Data were collected from 570 respondents across five zones of India, with 561 valid responses used for analysis. The methodological novelty lies in combining a two-stage analytical framework that integrates Partial Least Squares Structural Equation Modelling (PLS-SEM), and Artificial Neural Network (ANN) techniques. First, PLS-SEM are employed to validate the measurement and structural relationships among key determinants and to test the model’s hypotheses. Second, ANN is applied as a nonlinear predictive tool to rank the relative importance of significant predictors. This hybrid PLS-SEM–ANN approach overcomes the limitations of linear-only modelling by capturing both theoretical causality and nonlinear predictive strength, thereby enhancing robustness and decision-making accuracy. The structural model explained 43.63% of the variance (R² = 0.4363) in Overall Impact on Health and Safety (OIHS). The ANN model achieved high prediction accuracy with R² = 0.85 and low RMSE values (Training RMSE = 0.109; Testing RMSE = 0.107). The model was validated as a good fit based on both the PLS-SEM and ANN approaches. Among the Governmental Regulatory Factor (GRF) was identified as the most influential in improving H&S in the Indian construction industry, followed by the Organizational Regulatory Factor (ORF). The Worker’s Self-Protective Factor (WSPF) and Job Factor (JF). This research provides practical guidance for creating a safer and healthier construction environment in India.