Analysis of Bridge Condition Rating Data Using Neural Networks

摘要: Currently, bridges are evaluated using either a visual inspection process or detailed structural analysis. When bridge evaluation is conducted by inspection, subjective rating assigned to component. With analytical evaluation, the computed based on load applied and resistance of There have been several attempts correlate rating. The conventional statistical analyses, as well methods fuzzy logic, not very successful in providing clear relationship between 2 systems. This paper describes application neural network systems developing relation ratings parameters that ratings. It shown networks can be trained used successfully estimating parameters. specific problem for railroad commuter rail system Chicago metro area presented.