Abstract
The seismic design for bridge-substructure has become more complicated in comparison with former. Recentry, there are so many variables, for desig ning substructure, applied to the seismic designstandards in JAPAN. Since there is a non-linear response between input and outcome of seismic design for bridge-substructure, it's a difficult to determine the initial design variables and to estimate the quality of the results. In this study, we tried to make a supporting system for the bridge engineer on the seismic design for substructure with the Self-Organizing Map, a kind of Neural-Network.
