Purpose: The aim of this study was to investigate peri-implant bone remodeling as a response to biomechanical factors, including implant size and contour, magnitude of occlusal load, and properties of osteogenic bone grafts through the use of a computational algorithm. Materials and Methods: A bone-remodeling algorithm was incorporated into the finite element method, where bone remodeling takes place as a result of the biomechanical alteration caused by dental implant placement and continues until the difference between the homeostatic state and the altered state is minimized. The site-specific homeostatic state was based on a model consisting of a natural tooth. Three long (11-mm) implants and two short (5-mm) implants were investigated. A three-dimensional segment of the mandible was constructed from a computed tomographic image of the premolar region, and an extraction socket was filled with bone graft. Results: Generally, the extent of bone loss in the cortical region was greater and denser bone developed at both the implant crest and apex with increased occlusal loads. The areas between implant threads were prone to bone resorption. Bone graft materials that were relatively stiff and that had high equilibrium stimulus values appeared to cause increased bone loss. Conclusions: Short implants are better for conserving the mechanotransductive signaling environment of the natural tooth than long implants. Also, short implants are predicted to lead to less interfacial bone loss at high loads over the long term, while long implants are associated with a more consistent level of bone loss for different amounts of loading. It is also predicted that in the long term, bone grafts with relatively low elastic modulus lead to lower levels of interfacial bone loss. Keywords: bone grafts, bone remodeling, dental implants, extraction socket, site-specific stimulus, wide implants