Investigating the evolution of construction safety in Chinese provincial-level administrative regions: a thirteen-year analysis

The analysis of construction safety is essential for identifying weaknesses in management practices and offering strategic insights to enhance overall safety performance. The evolution of construction safety across Chinese provincial-level administrative regions from 2009 to 2021 was examined using the Malmquist productivity index (MPI) and the data envelopment analysis (DEA) window model, with a focus on input-output efficiency and variable returns to scale. Over the past thirteen years, the traditional DEA model has consistently identified five provinces as benchmarks, with the eastern region particularly excelling in construction safety. In the investigation of temporal evolution using DEA-MPI, most regions exhibited higher scores in technical change compared to efficiency change, indicating that improvements in construction safety across Chinese provincial-level administrative regions have primarily been driven by frontier technical innovations. Notably, the central region demonstrated significant improvements in construction safety development between 2009 and 2021. The DEA-window model not only delivered similar performance rankings for provincial-level administrative regions that aligned with those of the traditional DEA model but also provided more precise performance scores, indirectly corroborating the temporal analysis results from the DEA-MPI evaluation. This study employs input-output methodology for identifying provincial construction safety benchmarks and examining provincial construction safety evolution, providing construction policymakers with data-driven and actionable insights to develop targeted macro-level safety improvement strategies.