Learning from failure propagation in steel truss bridges

Abstract

Although truss-type bridge collapses usually have catastrophic consequences, their analysis present opportunities for improving different aspects in the field of bridge engineering, such as structural assessment, structural health monitoring, maintenance and conservation or even design strategies. As the world experiences more extreme events, efforts have been made to design more resilient bridges that can withstand local failures. Forensic techniques have contributed to understanding the causes and risk factors of bridge failures, and the creation of collapse databases has provided valuable insights for preventing such failures. However, these databases often focus on the hazards and do not provide information on initial damage and how it propagates, which is essential for improving the progressive collapse resistance of truss-type bridges. The main novelty of this paper is to present a methodology to identify triggering events leading to progressive collapse on truss-type bridges. It is the first time that a methodology includes a novel database which collects detailed information on initial damages and its propagation, as well as the consequences of the collapse. The methodology was implemented by collecting information from 25 case studies present in the literature. Results have allowed to identify most frequent initial constituted damages states or failures (ICDS) leading progressive collapse. In terms of consequences, results were thoroughly analysed and compared with predictions from different casualty models. The findings showed that the proposed methodology serves as an effective tool for identifying the triggering events of progressive collapse in truss-type bridges.