THE BOLU VIADUCT: A COMPARISON BETWEEN HAND CALCULATIONS-AND SOFTWARE-BASED SEISMIC

The Bolu viaduct, part of the Great Anatolian Highway, was almost completed when the Duzce earthquake occurred on November 12, 1999 adjacent to the bridge site. The structure had suffered extensive damage during the ground shaking, whose intensity was significantly larger than the design level, and a displacement of approximately 1.5 m had occurred be-tween its piers due to fault slip. The viaduct consists of two bridges, having 58 and 59 spans. Each of them is made of 7 precast V-girders 39.4 m long with an in-situ topping, and sup-ported by pot-bearings. The rectangular reinforced concrete hollow piers with different heights were founded on massive reinforced concrete pile caps, standing on twelve 1.8 m diameter Cast-In-Drilled-Hole piles. During this seismic event, piers and foundations had experienced limited damages; on the opposite, the bearing system had been completely incapable of adsorbing the seismic demand, leading to severe damage to the upper-structure of the bridge. The assessment, repair and strengthening strategy was completely based on first principles, equilibrium and compatibility and understanding of the structural behaviour, per-forming hand calculations confirmed by FEM-based model analyses. The aim of this paper is to apply today’s state of the art in the evaluation of the bridge’s performance, taking ad-vantages of more refined modelling techniques and higher computational capacity of computers, in order to compare results obtained from analyses carried during the strengthening process and those obtained performing more advanced analyses. Appropriate spectra and ground motions are selected according to current practices and guidelines. This study pre-sents how different choices could have been taken today, considering cost-benefit analysis based on EAL, relatively to joints, bearing and isolation system. Lastly, an overall comparison between hand calculations-and software-based methodologies is shown.