Towards an autonomous real-time tracking system of near-miss accidents on construction sites

It has been widely accepted that reported accidents on construction sites are just the tip of the iceberg representing the safety challenge because a very large number of near-miss accidents exist and constitute the portion of under-water surface. Undoubtedly, near-miss accidents provide insights into possible accidents and provide a great opportunity to further improve safety margins. It is possible to significantly improve the safety performance by learning from previous near-miss accidents and tracking near-miss accidents in real time leading to taking appropriate action before a potential up-coming accident. However, little research has been done in both autonomous data requirement analysis of near-miss accidents and technological solutions to track near-miss accidents based on real-time information on construction sites. The objective of this paper is to analyze and verify the autonomous information requirement of near-miss accidents and investigate the feasibility and performance of using a Zigbee RFID sensor network to fulfil these requirements. First, the autonomous data requirement of near-miss accidents is systematically analyzed based on typical historical accident cases. Then, considering the features of construction sites and characteristics of near-miss accidents, an autonomous real-time tracking system of near-miss accidents (ARTTS-NMA) on construction sites is proposed, which employs ultrasonic for outdoor and indoor real-time location tracking, adopts sensors for environment surveillance, RFID for access control as well as storage of safety information about workers, equipment and materials, and wireless sensor networks for data transmission. All system components are integrated into a Zigbee RFID sensor network architecture that features a relatively low cost and fast implementation with a pure wireless network backbone. The demonstration system based on the hybrid RFID sensor network architecture is fully working in a laboratory environment and for safety reasons we have chosen to evaluate it in a warehouse, which is considered as the main scenario of a construction site containing both multi-storey structure and open area, to evaluate the complexity of implementation.