Integration of building information modeling (BIM) and artificial intelligence (AI) to detect combined defects of infrastructure in the railway system

Due to the high demand for the railway system nowadays, the speed and load of rolling stocks tend to increase. At the same time, the effect of extreme climate is also more severe. These result in the deterioration of the railway infrastructure which cause defects to the railway infrastructure. Defects can affect passenger comfort and operating safety of the railway system. Detecting defects of the railway infrastructure in the early stage of defect development can reduce the risk to the railway operation, cost of maintenance and make the asset management more efficient. This study aims to apply building information modeling (BIM) integrated with artificial intelligence (AI) to develop the detection system of defects in railway infrastructure. In this study, dipped joint and settlement are used as examples of combined defects in the railway infrastructure. To detect defects, AI techniques are applied. Deep neural network and convolutional neural network are used to develop predictive models to detect defects in the railway infrastructure and rolling stock. The results of the study show that the developed models have the potential to detect defects with accuracies up to 99% and are beneficial for the asset management of the railway system in terms of risk management, passenger comfort, and cost-efficiency.