Application of Support Vector Machine (SVM) in Human Factors Analysis of Aviation Accidents: A Study Based on Chinese Accident Databases
Bibliographic record
Abstract
Aviation safety remains critical for mainland China, in which human factors continued as a dominant factor in causal relationship to accidents. The objective of this study is to investigate the causal pathways of aviation accident in China using the Localized Human Factor Analysis and Classification System(L-HFACS) with 65 accident reports over the years between 2008-2017 in order to identify human factor pathways from the organizational level to unsafe acts and estimate the higher-order impacts of human factors on the occurrence of unsafe behavior. To achieve this objective, we integrated the conventional HFACS model and machine learning technique. Specifically, we applied a Support Vector Machine (SVM) algorithm to simulate and predict accident causalities with encoded information of human factors. Our SVM model employing a radial basis function kernel and optimization by grid search and 5-fold cross-validation were used. Evaluation of the model indicated very high levels of predictive accuracy of 95.8%, precision 95.5%, recall 94.8% and F1-score 95.1%. It showed that SVM model was superior to traditional statistical approaches such as the Pearson chisquare test to identify the non-linear and multi-level interactions between human factors. It also shows that significant contributing categories were inadequate supervision, crew resource management and organizational climate. The SVM model provided better insights of inter-level causal relationships between these factors that were hard to identify by classical methodologies. This study shows the implications of integrating machine learning with structured human factors models in aviation safety analysis. This study's results suggest the effectiveness of the complementary role of SVM as an augmentation of existing HFACS-based risk analysis to yield a more effective and complete assessment of causes of aviation accidents in the mainland China. This in turns improves further the development of targeted and practical safety management practices and safety risk controls.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.005 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.003 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".