Flexure–Shear Response of a RC Double-Column Bridge Pier: Individual Column Cyclic Tests and System-Level Numerical Analysis

Double-column reinforced-concrete (RC) hollow bridge piers are often assessed using flexure-dominated models, although their thin walls can develop shear-related deterioration. This study reanalyzes seven previously reported cyclic tests on scaled square hollow-pier columns and extends the assessment to a full-scale double-column system in OpenSees. The tests varied the shear-span ratio and transverse and longitudinal reinforcement. A flexure-only fiber model and an axial–flexure–shear interaction membrane–beam–truss element model (AFSI-MBTEM) were evaluated using six response indicators before cyclic and nonlinear time-history analyses of the prototype system. Lower shear-span ratios increased resistance but intensified inclined cracking, stiffness loss, and post-peak deterioration; increasing L/D from 3.9 to 7.9 reduced peak strength from 320.5 to 145.9 kN. AFSI-MBTEM reduced the mean absolute errors in peak strength, yield displacement, and effective stiffness to 3.08%, 11.83%, and 12.86%, respectively, but did not improve residual-displacement or mean hysteretic-loop-energy predictions. At the system level, shear-span ratio most strongly affected cyclic stiffness and peak base shear; wall-thickness ratio, width-to-depth ratio, and longitudinal reinforcement were also influential. Under three near-fault records, AFSI-MBTEM predicted 5.31–10.32% lower peak base shear, while displacement changes remained record-dependent. Shear-sensitive modeling improves force and deformation assessment, but the system-level trends remain conditional on the adopted prototype, parameter ranges, and records.