In this paper, we have developed a prediction model for bridge condition rating. To achieve this goal, feature selection and hybridization of machine learning algorithm is done. The feature selection is done using infinite feature selection algorithm to select most appropriate features of the bridge. Further, decision tree and KNN machine learning algorithm is taken under consideration for bridge condition rating purposes. The simulation evaluation is done on standard NBI database and three performance metrics such as accuracy factor, mean absolute error, and mean square error are determined. The result shows that the proposed model achieves lower value of these parameters over the existing models such as “artificial neural network (ANN)”, “Markov Model”, “Hidden Markov Model”, and “Semi-Markov Model”