ABSTRACT
The Sunshine Skyway Bridge, inaugurated in 1987, was born out of necessity following a tragic vessel collision that destroyed the southbound main span, killing 35 people. Conceived by the accomplished French engineer, Jean M. Mueller, this iconic bridge boasts a 366 m cable-stayed main span, complemented by a 1,219 m main unit featuring 11 spans employing a combination of cable-stayed box girder and balanced cantilever post-tensioned concrete box girder components for the back spans. Acknowledging the significant milestone of the bridge reaching its 37th year,, the Florida Department of Transportation is actively engaged in intensive efforts to extend the service life of the bridge. This paper delineates the innovative Data-Driven Preventive Maintenance and Service Life Extension approach currently being applied to the Sunshine Skyway Bridge. Spearheaded by TYLin, the approach integrates structural monitoring employing nearly 100 sensors, encompassing conventional structural health monitoring (SHM) sensors combined with custom-built sensors designed to cater specifically to the distinctive monitoring requirements of this structure. Concurrently, a finite element model is under development to precisely represent the as-built structure. This model will be fine-tuned with sensor data, culminating in the creation of a comprehensive digital twin of the bridge, which is poised to enhance comprehension of the bridge’s behavior and furnish crucial data for gauging the remaining service life of critical components such as stay cables and stay cable dampers. In tandem with monitoring and modeling, the data procured from sensors undergoes advanced post-processing techniques using “Big Data” analytics. Currently, this data is used to train a machine learning model, enabling real-time prediction of structural parameters. This capability will facilitate the instantaneous identification of any parameter deviating from the normal range, thereby enhancing the bridge’s overall safety and performance.
