Factors influencing unsafe behaviors: A supervised learning approach

Experimental Study on Emergency Psychophysiological and Behavioral Reactions to Coal Mining Accidents

Effective emergency responses are crucial for preventing coal mine accidents and mitigating injuries. This paper aims to investigate the characteristics of emergency psychophysiological reactions to coal mine accidents and to explore the potential of key indicators for identifying emergency behavioral patterns. Initially, virtual reality technology facilitated a simulation experiment for emergency escape during coal mine accidents. Subsequently, the characteristics of emergency reactions were analyzed through correlation analysis, hypothesis testing, and analysis of variance. The significant changes in physiological indicators were then taken as input features and fed into the three classifiers of machine learning algorithms. These classifications ultimately led to the identification of behavioral patterns, including agility, defensiveness, panic, and rigidity, that individuals may exhibit during a coal mine accident emergency. The study results revealed an intricate relationship between the mental activities induced by accident stimuli and the resulting physiological changes and behavioral performances. During the virtual reality simulation of a coal mine accident, subjects were observed to experience significant physiological changes in electrodermal activity, heart rate variability, electromyogram, respiration, and skin temperature. The random forest classification model, based on SCR + RANGE + IBI + SDNN + LF/HF, outperformed all other models, achieving accuracies of up to 92%. These findings hold promising implications for early warning systems targeting abnormal psychophysiological and behavioral reactions to emergency accidents, potentially serving as a life-saving measure in perilous situations and fostering the sustainable growth of the coal mining industry.

Predicting types of human-related maritime accidents with explanations using selective ensemble learning and SHAP method

Maritime accidents frequently lead to severe property damage and casualties, and an accurate and reliable risk prediction model is necessary to help maritime stakeholders assess the current risk situation. Therefore, the present study proposes a hybrid methodology to develop an explainable prediction model for maritime accident types. Based on the advantages of selective ensemble learning method, this study pioneers to introduce a two-stage model selection method, aiming to enhance the predictive accuracy and stability of the model. Then, SHAP (Shapley Additive Explanations) method is integrated to identify effective mapping associations of seafarers' unsafe acts and their risk factors with the prediction results. The results demonstrate that the model developed achieves good prediction performance with an accuracy of 87.50 % and an F1-score of 84.98 %, which benefits stakeholders in assessing the type of maritime accident in advance, so as to make proactive intervention measures.

An interpretable clustering approach to safety climate analysis: Examining driver group distinctions

The transportation industry, particularly the trucking sector, is prone to workplace accidents and fatalities. Accidents involving large trucks accounted for a considerable percentage of overall traffic fatalities. Recognizing the crucial role of safety climate in accident prevention, researchers have sought to understand its factors and measure its impact within organizations. While existing data-driven safety climate studies have made remarkable progress, clustering employees based on their safety climate perception is innovative and has not been extensively utilized in research. Identifying clusters of drivers based on their safety climate perception allows the organization to profile its workforce and devise more impactful interventions. The lack of utilizing the clustering approach could be due to difficulties interpreting or explaining the factors influencing employees' cluster membership. Moreover, existing safety-related studies did not compare multiple clustering algorithms, resulting in potential bias. To address these problems, this study introduces an interpretable clustering approach for safety climate analysis. This study compares five algorithms for clustering truck drivers based on their safety climate perceptions. It also proposes a novel method for quantitatively evaluating partial dependence plots (QPDP). Then, to better interpret the clustering results, this study introduces different interpretable machine learning measures (Shapley additive explanations, permutation feature importance, and QPDP). The Python code used in this study is available at https://github.com/NUS-DBE/truck-driver-safety-climate. This study explains the clusters based on the importance of different safety climate factors. Drawing on data collected from more than 7,000 American truck drivers, this study significantly contributes to the scientific literature. It highlights the critical role of supervisory care promotion in distinguishing various driver groups. Moreover, it showcases the advantages of employing machine learning techniques, such as cluster analysis, to enrich the scientific knowledge in this field. Future studies could involve experimental methods to assess strategies for enhancing supervisory care promotion, as well as integrating deep learning clustering techniques with safety climate evaluation.

Mechanism Models of the Conventional and Advanced Methods of Construction Safety Training. Is the Traditional Method of Safety Training Sufficient?

Cognitive failures at the information acquiring (safety training), comprehension, or application stages led to near-miss or accidents on-site. The previous studies rarely considered the cognitive processes of two different kinds of construction safety training. Cognitive processes are a series of chemical and electrical brain impulses that allow you to perceive your surroundings and acquire knowledge. Additionally, their attention was more inclined toward the worker's behavior during hazard identification on-site while on duty. A study is proposed to fill the knowledge gap by developing the mechanism models of the two safety training approaches. The mechanism models were developed based on cognitive psychology and Bloom's taxonomy and six steps of cognitive learning theory. A worker's safety training is vital in acquiring, storing, retrieving, and utilizing the appropriate information for hazard identification on-site. It is assumed that those trained by advanced techniques may quickly identify and avoid hazards on construction sites because of the fundamental nature of the training, and when they come across threats, they may promptly use their working memory and prevent them, especially for more complex projects. The main benefit of making such a model, from a cognitive point of view, is that it can help us learn more about the mental processes of two different types of construction safety training, and it can also help us come up with specific management suggestions to make up for the approaches' flaws. Future research will concentrate on the organizational aspects and other cognitive failures that could lead to accidents.

The Cost Burden of Safety Risk Incidents on Construction: A Probabilistic Quantification Method

The construction sector is vulnerable to safety risk incidents due to its dynamic nature. Although numerous research efforts and technological advancements have focused on addressing workplace injuries, most of the studies perform empirical and deterministic postimpact evaluations on construction project performance. The effective modeling of the safety risk impacts on project performance provides decisionmakers with a valuable tool toward incidents prevention and proper safety risk management. Therefore, this study collected Australian incident records from the construction industry from 2016 onwards and conducted discrete event simulation to quantitatively measure the impact of safety risk incidents on project cost performance. Moreover, this study investigated the correlation between safety risk incidents and the age of injured workers. The findings show a strong correlation between the middle-aged workforce and the severity of incidents on project cost overruns. The ex-ante, nondeterministic analysis of safety risk impacts on project performance provides insightful results that will advance safety management theory in the direction of achieving zero harm workplace environments.

Analysis of Factors Influencing Miners' Unsafe Behaviors in Intelligent Mines using a Novel Hybrid MCDM Model

Coal mine accidents seriously affect people’s safety and social development, and intelligent mines have improved the production safety environment. However, safety management and miners’ work in intelligent mines face new changes and higher requirements, and the safety situation remains challenging. Therefore, exploring the key influencing factors of miners’ unsafe behaviors in intelligent mines is important. Our work focuses on (1) investigating the relationship and hierarchy of 20 factors, (2) using fuzzy theory to improve the decision-making trial and evaluation laboratory (DEMATEL) method and introducing the maximum mean de-entropy (MMDE) method to determine the unique threshold scientifically, and (3) developing a novel multi-criteria decision-making (MCDM) model to provide theoretical basis and methods for managers. The main conclusions are as follows: (1) the influence degree of government regulation, leadership attention, safety input level, safety system standardization, and dynamic supervision intensity exert the most significant influence on the others; (2) the causality of government regulation, which is the deep factor, is the highest, and self-efficacy displays the smallest causality, and it is the most sensitive compared to various other factors; (3) knowledge accumulation ability, man–machine compatibility, emergency management capability, and organizational safety culture has the highest centrality among the individual factors, device factors, management factors, and environmental factors, respectively. Thus, corresponding management measures are proposed to improve coal mine safety and miners’ occupational health.

Advancing safety analytics: A diagnostic framework for assessing system readiness within occupational safety and health

Big data and analytics have shown promise in predicting safety incidents and identifying preventative measures directed towards specific risk variables. However, the safety industry is lagging in big data utilization due to various obstacles, which may include lack of data readiness (e.g., disparate databases, missing data, low validity) and personnel competencies. This paper provides a primer on the application of big data to safety. We then describe a safety analytics readiness assessment framework that highlights system requirements and the challenges that safety professionals may encounter in meeting these requirements. The proposed framework suggests that safety analytics readiness depends on (a) the quality of the data available, (b) organizational norms around data collection, scaling, and nomenclature, (c) foundational infrastructure, including technological platforms and skills required for data collection, storage, and analysis of health and safety metrics, and (d) measurement culture, or the emergent social patterns between employees, data acquisition, and analytic processes. A safety-analytics readiness assessment can assist organizations with understanding current capabilities so measurement systems can be matured to accommodate more advanced analytics for the ultimate purpose of improving decisions that mitigate injury and incidents.

Is There a Difference in Brain Functional Connectivity between Chinese Coal Mine Workers Who Have Engaged in Unsafe Behavior and Those Who Have Not?

(1) Background: As a world-recognized high-risk occupation, coal mine workers need various cognitive functions to process the surrounding information to cope with a large number of perceived hazards or risks. Therefore, it is necessary to explore the connection between coal mine workers’ neural activity and unsafe behavior from the perspective of cognitive neuroscience. This study explored the functional brain connectivity of coal mine workers who have engaged in unsafe behaviors (EUB) and those who have not (NUB). (2) Methods: Based on functional near-infrared spectroscopy (fNIRS), a total of 106 workers from the Hongliulin coal mine of Shaanxi North Mining Group, one of the largest modern coal mines in China, completed the test. Pearson’s Correlation Coefficient (COR) analysis, brain network analysis, and two-sample t-test were used to investigate the difference in brain functional connectivity between the two groups. (3) Results: The results showed that there were significant differences in functional brain connectivity between EUB and NUB among the frontopolar area (p = 0.002325), orbitofrontal area (p = 0.02102), and pars triangularis Broca’s area (p = 0.02888). Small-world properties existed in the brain networks of both groups, and the dorsolateral prefrontal cortex had significant differences in clustering coefficient (p = 0.0004), nodal efficiency (p = 0.0384), and nodal local efficiency (p = 0.0004). (4) Conclusions: This study is the first application of fNIRS to the field of coal mine safety. The fNIRS brain functional connectivity analysis is a feasible method to investigate the neuropsychological mechanism of unsafe behavior in coal mine workers in the view of brain science.

The role of machine learning in the primary prevention of work-related musculoskeletal disorders: A scoping review

To determine the applications of machine learning (ML) techniques used for the primary prevention of work-related musculoskeletal disorders (WMSDs), a scoping review was conducted using seven literature databases. Of the 4,639 initial results, 130 primary research studies were deemed relevant for inclusion. Studies were reviewed and classified as a contribution to one of six steps within the primary WMSD prevention research framework by van der Beek et al. (2017). ML techniques provided the greatest contributions to the development of interventions (48 studies), followed by risk factor identification (33 studies), underlying mechanisms (29 studies), incidence of WMSDs (14 studies), evaluation of interventions (6 studies), and implementation of effective interventions (0 studies). Nearly a quarter (23.8%) of all included studies were published in 2020. These findings provide insight into the breadth of ML techniques used for primary WMSD prevention and can help identify areas for future research and development.

Management of safe distancing on construction sites during COVID-19: A smart real-time monitoring system

The outbreak of Coronavirus Disease 2019 (COVID-19) poses a great threat to the world. One mandatory and efficient measure to prevent the spread of COVID-19 on construction sites is to ensure safe distancing during workers' daily activities. However, manual monitoring of safe distancing during construction activities can be toilsome and inconsistent. This study proposes a computer vision-based smart monitoring system to automatically detect worker breaching safe distancing rules. Our proposed system consists of three main modules: (1) worker detection module using CenterNet; (2) proximity determination module using Homography; and (3) warning alert and data collection module. To evaluate the system, it was implemented in a construction site as a case study. This study has two key contributions: (1) it is demonstrated that monitoring of safe distancing can be automated using our approach; and (2) CenterNet, an anchorless detection model, outperforms current state-of-the-art approaches in the real-time detection of workers.

Artificial Intelligence Methodologies for Data Management

This study analyses the main challenges, trends, technological approaches, and artificial intelligence methods developed by new researchers and professionals in the field of machine learning, with an emphasis on the most outstanding and relevant works to date. This literature review evaluates the main methodological contributions of artificial intelligence through machine learning. The methodology used to study the documents was content analysis; the basic terminology of the study corresponds to machine learning, artificial intelligence, and big data between the years 2017 and 2021. For this study, we selected 181 references, of which 120 are part of the literature review. The conceptual framework includes 12 categories, four groups, and eight subgroups. The study of data management using AI methodologies presents symmetry in the four machine learning groups: supervised learning, unsupervised learning, semi-supervised learning, and reinforced learning. Furthermore, the artificial intelligence methods with more symmetry in all groups are artificial neural networks, Support Vector Machines, K-means, and Bayesian Methods. Finally, five research avenues are presented to improve the prediction of machine learning.

Virtual Fence System Based on IoT Paradigm to Prevent Occupational Accidents in the Construction Sector

Many occupational accidents in construction sites are caused by the intrusion of a worker into a hazardous area. Technological solutions based on RFID, BIM, or UWB can reduce accidents, but they still have some limitations.The aim of the current paper is to design and evaluate a new system of “virtual fences” based on Bluetooth Low-Energy (BLE) to avoid intrusions. First of all, the system was designed using a number of beacons, a Bayesian filter, a finite state machine, and an indicator. Secondly, its safety attributes were evaluated based on a scientific questionnaire by an expert panel following the staticized groups’ methodology. Results showed that the proposal is inexpensive and easy to integrate and configure. The selected experts evaluated positively all the attributes of the system, and provided valuable insights for further improvements. From the experts’ discussions, we concluded that successful adoption of this “virtual fence” system based on BLE beacons should consider the influence of factors such as cost savings, top management support, social acceptance, and compatibility and integration with existing systems, procedures, and company culture. In addition, legislation updates according to technical advances would help with successful adoption of any new safety system.

Predicting the consequences of accidents involving dangerous substances using machine learning

A new dimension of learning lessons from the occurrence of hazardous events involving dangerous substances is considered relying on the availability of representative data and the significant evolution of a wide range of machine learning tools. The importance of such a dimension lies in the possibility of predicting the associated nature of damages without imposing any unrealistic simplifications or restrictions. To provide the best possible modeling framework, several implementations are tested using logistic regression, decision trees, neural networks, support vector machine, naive Bayes classifier and random forests to forecast the occurrence of the human, environmental and material consequences of industrial accidents based on the EU Major Accident Reporting System's records. Many performance metrics are estimated to select the most suitable model in each treated case. The obtained results show the distinctive ability of random forests and neural networks to predict the occurrence of specific consequences of accidents in the industrial installations, with an obvious exception concerning the performance of this latter algorithm when the involved datasets are highly unbalanced.

Antecedents of safety behavior in construction: A literature review and an integrated conceptual framework

There has been no scarcity in the literature of suggested antecedents of employee safety behavior, and this paper brings together the disaggregated antecedents of safety behavior in the construction field. In total, 101 eligible empirical articles are obtained. Bibliometric and context analyses are combined to identify the influential journals, scholars, keywords, use of theory, research methods, and countries or regions of the empirical samples. The 83 factors that are identified are divided into five groups, namely (a) individual characteristics, (b) workgroup interactions, (c) work and workplace design, (d) project management and organization, and (e) family, industry, and society. This indicates that the causes of safety behavior are manifold. Various factors from different systems likely work in concert to create situations in which an individual chooses to comply with safety rules and participate voluntarily in safety activities. Given this, we propose that safety behavior is only an ostensible symptom of more complex "The Self-Work-Home-Industry/Society" systems and establish a safety behavior antecedent analysis and classification model. Based on this model, we develop a resource flow model, illustrating why, how, and when the flow of resources between the five systems-namely the self system, work system, home system, work-home interface system, and industry/society system-either promotes or inhibits safety behavior. The safety behavior antecedent analysis and classification model and resource flow model are based mainly on bioecological system theory and resources theories. Avenues for future theoretical development and method designs are suggested based on the reviewed findings and the two conceptual models. The intention with this systematic review together with the two integrated conceptual models is to advance theoretical thinking on how safety behavior can be promoted, or instead, inhibited.

Knowledge transfer for occupational health and safety: Cultivating health and safety learning culture in construction firms

Within the last decades the incidence of workspace injuries and fatalities in the UK construction industry has declined markedly following the developments in occupational health and safety (OHS) management systems. However, safety statistics have reached a plateau and actions for further improvement of OHS management systems are called for. OHS is a form of organizational expertise that has both tacit and explicit dimensions and is situated in the ongoing practices. There is a need for institutionalization and for the transfer of knowledge across and along construction supply chains to reduce OHS risks and facilitate cultural change. The focus of this article is the factors that facilitate OHS knowledge transfer in and between organizations involved in construction projects. An interpretative methodology is used in this research to embrace tacit aspects of knowledge transfer and application. Thematic analysis is supported by a cognitive mapping technique that allows understanding of interrelationships among the concepts expressed by the respondents. This paper demonstrates inconsistency in OHS practices in construction organizations and highlights the importance of cultivating a positive safety culture to encourage transfer of lessons learnt from good practices, incidents, near misses and failures between projects, from projects to programmes and across supply chains. Governmental health and safety regulations, norms and guidelines do not include all possible safety issues specific to different working environments and tied to work contexts. The OHS system should encourage employees to report near misses, incidents and failures in a 'no-blame' context and to take appropriate actions. This research provides foundation for construction project practitioners to adopt more socially oriented approaches towards promoting learning-rich organizational contexts to overcome variation in the OHS and move beyond the current plateau reached in safety statistics.

Understanding the Sociocognitive Process of Construction Workers' Unsafe Behaviors: An Agent-Based Modeling Approach

Previous literature has recognized that workers’ unsafe behavior is the combined result of both isolated individual cognitive processes and their interaction with others. Based on the consideration of both individual cognitive factors and social organizational factors, this paper aims to develop an Agent-Based Modeling (ABM) approach to explore construction workers’ sociocognitive processes under the interaction with managers, coworkers, and foremen. The developed model is applied to explore the causes of cognitive failure of construction workers and the influence of social groups and social organizational factors on the workers’ unsafe behavior. The results indicate that (1) workers’ unsafe behaviors are gradually reduced with the interaction with managers, foremen, and workers; (2) the foreman is most influential in reducing workers’ unsafe behaviors, and their demonstration role can hardly be ignored; (3) the failure of sociocognitive process of construction workers is affected by many factors, and cognitive process errors could be corrected under social norms; and (4) among various social organizational factors, social identity has the most obvious effect on reducing workers’ unsafe behaviors, and preventive measures are more effective than reactive measures in reducing workers’ unsafe behaviors.

Exploring the Impact of Unsafe Behaviors on Building Construction Accidents Using a Bayesian Network

Unsafe behavior is a critical factor leading to construction accidents. Despite numerous studies supporting this viewpoint, the process by which accidents are influenced by construction workers’ unsafe behaviors and the extent to which unsafe behaviors are involved in this process remain poorly discussed. Therefore, this paper selects cases from Chinese building construction accidents to explore the probabilistic transmission paths from unsafe behaviors to accidents using a Bayesian network. First, a list of unsafe behaviors is constructed based on safety standards and operating procedures. Second, several chains of unsafe behaviors are extracted from 287 accident cases within four types (fall, collapse, struck-by and lifting) to form a Bayesian network model. Finally, two accidents are specifically analyzed to verify the rationality of the proposed model through forward reasoning. Additionally, critical groups of unsafe behaviors leading to the four types of accidents are identified through backward reasoning. The results show the following: (i) The time sequence of unsafe behaviors in a chain does not affect the final posterior probability of an accident, but the accident attribute strength of an unsafe behavior, affects the growth rate of the posterior probability of an accident. (ii) The four critical groups of unsafe behaviors leading to fall, collapse, struck-by, and lifting are identified. This study is of theoretical and practical significance for on-site behavioral management and accident prevention.