Large numbers of 1950's vintage conventionally reinforced concrete (CRC) bridges remain in-service in the national bridge inventory. Many of these bridges are lightly reinforced for shear. Evaluation of these bridges to prevent unnecessary and costly repairs requires refined analytical techniques. This dissertation presents finite element (FE) modeling and comparisons of various analytical methods for predicting capacity of CRC girders typical of reinforced concrete deck girder (RCDG) bridges. Analyses included bridge-system load distribution, member capacity prediction, and consideration of corrosion damage for strength deterioration. Two in-service RCDG bridges were inspected and instrumented to measure response under known load configurations. Load distribution was developed for the bridges based on the field data. Comparisons with AASHTO factors indicated the design factors for load distribution are conservative. Load distribution of the tested bridges was numerically obtained using FE analysis. The comparisons between predicted results and field-test data indicated the elastic FE analysis can be used for modeling of cracked RCDG bridges to predict load distribution factors for more accurate bridge evaluation. Analyses were performed for a large set of full-size RCDG, designed to reflect 1950's vintage details, and tested using various loading configurations. Four different analysis methods were used to predict the capacity of the specimens considering details of various stirrup spacing, debonded stirrups, flexural-bar cutoff, anchorage of flexural reinforcing, and moving supports. Nonlinear FE analyses were performed to predict behavior of two groups of experimental reinforced concrete (RC) specimens. Two different span-to-depth ratios were included: 2.0 and approximately 3.0. Concrete confinement effects were included in the material modeling. A quasi-displacement control technique was developed to reduce solution times. The FE predicted results correlated well with the experimental data. FE modeling techniques were developed to isolate different contributions of corrosion damage to structural response of experimental RC beams designed to produce diagonal-tension failures. Corrosion-damage parameters included concrete cover spalling; uniform stirrup cross-sectional loss; local stirrup cross-sectional loss due to pitting; and debonding of corrosion-damaged stirrups from the concrete. FE analyses were performed including both individual and combined damages. The FE results matched experimental results well and quantitatively estimated capacity reduction of the experimental specimens. / Graduation date: 2005 / Best scan available.