Empirical Testing for Use of Safety Related Measures at the Organizational Level to Assess and Control the On-Site Risk Level

Abstract

In construction, safety related measures have been proposed to evaluate ongoing safety performance and reduce the identified risk proactively. However, the relationships between these measures and on-site incident records merit further investigation from a practical perspective (e.g., data collection and sharing at an organizational level). This paper collects and analyzes the safety related measures available in an organization to review and assess the effect of these measures for assessment and control of the on-site risk level (ORL) in practice. Through a case study conducted at four construction projects, cross-sectional data of seven safety related measures were identified and collected from three departments (payroll, human resources, and safety). Then, regression models were built to investigate the influence of these measures on incident records. Overall, the results demonstrate the influence of safety related measures on ORL (adjusted R-2 = 0.488), reasonable estimation of the ORL trend for a different project (mean absolute percentage error = 36.69%), and potential use in safety policies. These findings also imply that data sharing in an organization may facilitate a proactive and integral approach to control on-site safety performance.