This study introduces the concept of pandemic safety climate (SC), employees' perceptions of an organization's policies, procedures and practices aiming to deal with the COVID-19 pandemic. Based on the theory of planned behaviour, we expected that the pandemic SC would be the base of employees' subjective norms, attitudes and perceived control in dealing with the COVID-19 pandemic-related challenges. We hypothesized that pandemic SC would be associated with a series of attitudinal and behavioural criteria that aim to prevent the spread of COVID-19 as well as promote other health and well-being outcomes. Using both prospective and cross-sectional datasets, we developed and validated a measure of pandemic SC which consisted of two scales. Organization pandemic SC refers to the employees' perceptions of the strategies and efforts at the organization level and consists of four dimensions: management commitment and proactivity, workplace flexibility and capacity, equipment and sanitization for COVID-19 prevention and COVID-19-related communication and training programmes. Group pandemic SC refers to the employees' perceptions of the intermediate support and care from supervisors and consists of three dimensions: supervisor commitment and proactivity, safety monitoring and COVID-19-related supervisory communication. Construct validity and criterion-related validity were supported. Theoretical and practical implications of the newly developed pandemic SC scales are discussed.

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.

The purpose of the current study was to use a mixed-methods approach to understanding safety climate and the strategies to improve safety climate among truck drivers. Using both survey (N = 7246) and interview (N = 18) responses provided by truck drivers regarding key safety climate items, the current study identified a number of positive and negative policies, procedures and practices that truck drivers perceived as the determinants of whether their organizations are committed to the promotion of safety at work. Item response theory (IRT) analyses were conducted to identify discrimination parameters indicating which safety climate items were most sensitive to the safety climate level. Discriminative items were identified at both the organization and group levels which can be used to evaluate safety climate and differentiate a high versus low safety climate across groups and organizations in the trucking industry. Based on our results, we also offer safety researchers and practitioners some recommendations on what and/or how to intervene with and promote organizational safety climate in the trucking industry.

Past studies on trucking accidents did not evaluate the importance of different safety climate factors in influencing trucking accident occurrence. In addition, despite its potential, limited studies have used machine learning (ML) to test the relationship between safety climate factors and trucking accidents. Thus, this paper aims to address these questions: using safety climate factors as independent variables, (1) which machine learning algorithm is more accurate in predicting self-reported trucking accidents? (2) which safety climate factor better predicts self-reported trucking accidents? First, factor analysis was conducted on a questionnaire survey dataset to identify the safety climate factors and test the model’s goodness-of-fit. Next, using the safety climate factors as inputs, the study evaluated the performance of different ML algorithms in predicting self-reported accidents. Last, we used the best-performing ML model to evaluate the relative importance of the different safety climate factors in predicting self-reported accidents. The factor analyses generated four organizational factors, three group factors, and two trust and satisfaction factors. Random Forest was the most accurate ML algorithm with an average area under the receiver operating characteristics curve value of 0.861. It suggested that the organizational factor “Proactiveness in safety management” is the most important factor for predicting self-reported accidents. This study makes three main contributions: (1) demonstration of how factor analysis and ML methods can be integrated; (2) identifying Random Forest as a promising ML algorithm for analyzing large safety climate datasets, and (3) deriving the relative importance of trucking safety climate factors in influencing self-reported accidents.