Predictive models of safety based on audit findings: Part 1: Model development and reliability

Hsiao, Yu-Lin; Drury, Colin G.; Wu, Changxu; Paquet, Victor

doi:10.1016/j.apergo.2012.07.010

Cited by 20 publications

(5 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Shappell and Wiegmann established the HFACS model based on the Swiss cheese model ( 18 ). This model identifies the vulnerabilities in cheese and has been widely used in various fields ( 13 ).…”

Section: Methodsmentioning

confidence: 99%

Risk Factors Identification of Unsafe Acts in Deep Coal Mine Workers Based on Grounded Theory and HFACS

Yang

Wang

Zhu

et al. 2022

Front. Public Health

View full text Add to dashboard Cite

The risk factors affecting workers' unsafe acts were comprehensively identified by Human Factors Analysis and Classification System (HFACS) and grounded theory based on interview data and accident reports from deep coal mines. Firstly, we collected accident case and field interview data from deep coal mines issued by authoritative institutions. Then, the data were coded according to grounded theory to obtain relevant concepts and types. The HFACS model was used to classify the concepts and categories. Finally, the relationship between core and secondary categories was sorted out by applying a story plot. The results show that risk factors of unsafe acts of deep coal mine workers include environmental factors, organizational influence, unsafe supervision and unsafe state of miners, and the main manifestations of unsafe acts are errors and violations. Among them, the unsafe state of miners is the intermediate variable, and other factors indirectly affect risky actions of coal miners through unsafe sates. Resource management, organizational processes and failure to correct problems are the top three risk factors that occur more frequently in unsafe acts. The three most common types of unsafe act are unreasonable labor organization, failure to enforce rules, and inadequate technical specifications. By combining grounded theory and the HFACS framework to analyze data, risk factors for deep coal miners can be quickly identified, and more precise and comprehensive conceptual models of risk factors in unsafe acts of deep coal miners can be obtained.

show abstract

Section: Methodsmentioning

confidence: 99%

Risk Factors Identification of Unsafe Acts in Deep Coal Mine Workers Based on Grounded Theory and HFACS

Yang

Wang

Zhu

et al. 2022

Front. Public Health

View full text Add to dashboard Cite

show abstract

“…An increase in a common effort within the research community has been noted to develop data-based Human Reliability Assessment (HRA) processes that can produce accessible predictive indicators, while leveraging the already acquired data. However, some of these prominent processes, especially those that rely on the contents of text reports, often require manual categorization of human factor categories, an expensive and errorprone task [17,18].…”

Section: Introductionmentioning

confidence: 99%

Machine Learning and Natural Language Processing for Prediction of Human Factors in Aviation Incident Reports

et al. 2021

View full text Add to dashboard Cite

In the aviation sector, human factors are the primary cause of safety incidents. Intelligent prediction systems, which are capable of evaluating human state and managing risk, have been developed over the years to identify and prevent human factors. However, the lack of large useful labelled data has often been a drawback to the development of these systems. This study presents a methodology to identify and classify human factor categories from aviation incident reports. For feature extraction, a text pre-processing and Natural Language Processing (NLP) pipeline is developed. For data modelling, semi-supervised Label Spreading (LS) and supervised Support Vector Machine (SVM) techniques are considered. Random search and Bayesian optimization methods are applied for hyper-parameter analysis and the improvement of model performance, as measured by the Micro F1 score. The best predictive models achieved a Micro F1 score of 0.900, 0.779, and 0.875, for each level of the taxonomic framework, respectively. The results of the proposed method indicate that favourable predicting performances can be achieved for the classification of human factors based on text data. Notwithstanding, a larger data set would be recommended in future research.

show abstract

“…Another study is the design of the human factors/ergonomics maintenance audit classification framework used to detect aviation-related risks according to text patterns of safety audit findings (Hsiao et al, 2013). While their proposed text classification models are found to be effective, the authors recommended to use more input data to establish deep learning and to conduct further testing to evaluate the robustness of the said models.…”

Section: Introductionmentioning

confidence: 99%

ISO 9001:2015 Quality Management System Requirements and Audit Findings Classification Using Support Vector Machine and Long Short-Term Memory Neural Network: An Optimization Method

Corpuz

2021

MJST

View full text Add to dashboard Cite

Generating accurate and timely internal and external audit reports may seem difficult for some auditors due to limited time or expertise in matching the correct clauses of the standard with the textual statement of findings. To overcome this gap, this paper presents the design of text classification models using support vector machine (SVM) and long short-term memory (LSTM) neural network in order to automatically classify audit findings and standard requirements according to text patterns. Specifically, the study explored the optimization of datasets, holdout percentage and vocabulary of learned words called NumWords, then analyzed their capability to predict training accuracy and timeliness performance of the proposed text classification models. The study found that SVM (96.74%) and LSTM (97.54%) were at par with each other in terms of the best training accuracy, although SVM (67.96±17.93 seconds [s]) was found to be significantly faster than LSTM (136.67±96.42 s) in any dataset size. The study proposed optimization formulas that highlight dataset and holdout as predictors of accuracy, while dataset and NumWords as predictors of timeliness. In terms of actual implementation, both classification models were able to accurately classify 20 out of 20 sample audit findings at 1 and 3 s, respectively. Hence, the extent of choosing between the two algorithms depend on the datasets size, learned words, holdout percentage, and workstation speed. This paper is part of a series, which explores the use of Artificial Intelligence (AI) techniques in optimizing the performance of QMS in the context of a state university.

show abstract

Predictive models of safety based on audit findings: Part 1: Model development and reliability

Cited by 20 publications

References 31 publications

Risk Factors Identification of Unsafe Acts in Deep Coal Mine Workers Based on Grounded Theory and HFACS

Risk Factors Identification of Unsafe Acts in Deep Coal Mine Workers Based on Grounded Theory and HFACS

Machine Learning and Natural Language Processing for Prediction of Human Factors in Aviation Incident Reports

ISO 9001:2015 Quality Management System Requirements and Audit Findings Classification Using Support Vector Machine and Long Short-Term Memory Neural Network: An Optimization Method

Contact Info

Product

Resources

About