Date of Award


Level of Access

Open-Access Thesis

Degree Name

Master of Science in Civil Engineering (MSCE)


Civil Engineering


Thomas C. Sandford

Second Committee Member

William G. Davids

Third Committee Member

Dana N. Humphrey


The advantages of constructing bridges with integral abutments are recognized by transportation agencies worldwide. However, pile supported integral abutments are limited to locations where the depth of overburden can provide fixed support conditions. In Maine, there are often cases where the depth to bedrock prohibits integral abutment bridges from being used. The goal of this research is to determine the feasibility of constructing integral abutments in conditions that cannot provide the fixed support conditions that are traditionally assumed. A finite element model was created that incorporates realistic constitutive and surface interaction models. These models allow for a good prediction of the soil/structure interaction and the structural response. Three critical model responses were identified: pile stresses, pile kinematics, and pilelbedrock interaction. These responses were examined in later parametric studies, which investigated how changes in girder length, pile length, loading, geometry, member properties, and subsurface conditions influenced the pile responses. It was shown that for piles less than 4 m in length on bedrock, the tip of the pile rotates but does not translate horizontally or vertically. This is similar, in principle, to a column with a pinned support. Dead and live loading of the girder induces a rotation of the abutments, which causes pile head displacement. Typically, displacements due to thermal loading are the only lateral pile displacements considered in integral abutment design. Under cyclic live and thermal loading, plastic deformation of the pile did not accumulate if the strains in the head were kept below 125% of the yield strain (1.25 E,). Observations of behavior from the parametric study were used as a basis for a set of design guidelines for piles that did not meet the length criteria of the current Maine Department of Transportation procedure. Using the criteria that pile head strains are kept below 1.25 E,, pile head moments based on data from the parametric studies are calculated from a relationship with the axial load. These relationships were created for various soil conditions and loadings, as well as pile sections. Forces at the pile tip are estimated from the moments at the head of the pile in order to determine if the pinned idealization is valid for the proposed pile/soil/load combinations. The ratio of shear forces and normal forces are compared to an equivalent coefficient of friction between the pile tip and bedrock, along with a factor of safety. The proposed design procedure results in values of moments and shear forces that are higher than those obtained from the finite element model. This is due to the inherent conservatism built into the methods used to calculate pile forces, which presents a worstcase design scenario. The proposed method expands the application of integral abutments to instances where an integral abutment supported by short piles is currently considered impractical. However, even with the expanded design criteria, finite element modeling indicates that there are cases where the combination of geometry, loading, and subsurface conditions may prohibit the use of integral abutments.