Bridge Life-Cycle Cost Analysis Using Artificial Neural Networks

Life-Cycle Cost analysis can significantly assist in making investment decisions. Several recent studies have recognized the potential benefits of Life-Cycle Cost analysis and call for use of such analyses when making infrastructure investments, including investments in bridges. The Life-Cycle Cost of a bridge consists of the total investment throughout the life of the bridge. This includes the initial construction cost, repair and rehabilitation costs, and all maintenance costs. The ability to accurately determine the Life-Cycle Cost of a bridge will help agencies evaluate the asset value of existing bridges, make better decisions on the design and construction of new ones, and choose improved methods and approaches for rehabilitating existing structures. Research has shown that timely maintenance, repair, and rehabilitation can lower the Life-Cycle Cost of a bridge. However, this is a complex and nonlinear problem, and previous studies have failed to develop a satisfactory model. One effective technique for solving nonlinear problems with complicated functions is an Artificial Neural Network. A neural network is a powerful data-modeling tool that captures and represents complex input/output relationships. Using a set of input and output data belonging to a particular problem, a neural system can be trained to predict outcomes for new versions of the same problem. Accordingly, an extensive set of data (bridge dimensions, age, initial cost, and Life-Cycle Cost) for 14 Chicago bridges was used to quantify the degree of success that could be achieved with this model. Sixty percent of the data was used as input to train the model and the remaining forty percent was used to assess the success of the model for predicting the Life-Cycle Cost. The results achieved were encouraging and suggest that the neural network model is a promising tool for predicting the Life- Cycle Cost of a bridge.