Analysis of Human Factors and Engineering Design Contributions to Aerospace Incidents

This study investigates the factors influencing aviation incident outcomes, with a focus on distinguishing between crash events and emergency landings. A curated dataset of 53 aviation incidents was analyzed using logistic regression and machine learning classifiers to identify the most influential human, technical, and operational variables. The analysis adopts a sociotechnical framework, examining how interactions between human performance and engineering design contribute to incident severity. Statistical results indicate that human factor variables and issues are the predictors exhibiting significant differences between crash and noncrash outcomes, with human factors demonstrating the strongest association. Chi-square analysis further confirms the significance of human factors ([Formula: see text]), while multicollinearity diagnostics (variance inflation factor [VIF] ≈ 2) support model stability. Among machine learning models, the support vector machine achieved the highest predictive accuracy (91%), indicating that aviation incident outcomes exhibit structured and separable patterns within the feature space. The findings emphasize that adverse outcomes are more strongly associated with deficiencies in training, communication, and situational awareness than with isolated technical faults. The study concludes that integrating human factors analysis and predictive modeling into the system development lifecycle can enable proactive risk identification and enhance aviation safety.