A Dominance based Rough Set analysis for investigating employee perception of safety at workplace and safety compliance. (July 2020)

• Record Type: Journal Article
• Authors: Singh, Arpit; Misra, Subhas Chandra
• Is Part Of: Safety science. Volume 127(2020)
• Publication Date: 2020-07
• Subjects: Rough sets -- Dominance based rough sets -- Occupational safety -- Data mining -- Safety compliance; Industrial accidents -- Periodicals; Accident Prevention -- Periodicals; Safety -- Periodicals; Travail -- Accidents -- Périodiques; 363.11
• DOI: 10.1016/j.ssci.2020.104702
• Languages: English
• ISSNs: 0925-7535

Abstract

Highlights: Data collection from an Indian construction site. Dominance based Rough set approach (DRSA) to predict safety compliance behavior. Management's direct involvement on-site encouraged compliance towards safety. Organized documentation on safety protocols important to impart safety education. DRSA performed better classification as compared to other machine learning tools. Abstract: Managing safety in the workplace is a pressing need for organizations and firms especially in the highly competitive world where the success of firms hinges on the overall productivity of organizations. Ensuring high productivity clearly relies on the efficiency of its workforce. Safety of employees is, therefore, an immediate concern that needs to be addressed to avoid negative consequences in terms of cost to the organizations and the subsequent reduction in productivity. Past literature is indicative of the fact that the major reason responsible for accidents occurring at workplace is unsafe behavior. To assess the behavior of the employees in safety context, perception analysis was conducted through a survey-based study of 98 employees of a construction industry where the Workplace Safety Scale (WSS) was administered and the analysis was performed using Dominance based Rough Set Analysis (DRSA). DRSA proved instrumental in dealing with the ambiguities in the preference ordered scenario and decision rules that were derived were a better and more intuitive display of the real-world situation with exceptionally high accuracy. Analysis revealed that the safety of the Supervisor and the co-workers amounted to be the better predictors of safety compliance of the employees. This primarily signifies the role that management has to play in ensuring the safety at workplace and increasing awareness amongst the employees about the importance of following safety rules while at work. Interestingly, the results also showed that all the constructs in the WSS contributed positively in the prediction of the safety compliance of the employees which was found to be consistent with the past studies done in the field of occupational hazards and safety.