A Robust Real Time Train Delay Prediction System to Support Passengers by using Fuzzy based Learning Scheme

The poor performance of public transportation delays is one of the most important problems confronting railway networks globally. In the railway industry, train delays are serious issue ads well as the Passengers' trust might be damaged if their transportation is often late, and they may start looking for alternative options as a result. Poor signaling and a lack of rails mean that trains rarely arrive or depart at their designated times. Multiple attribute information in fixed, time-series and geospatial forms are produced by dynamic structures that include objects that move. In this paper, a novel deep learning assisted fuzzy based prediction scheme is introduced called Intelligent Fuzzy Learning to Delay Prediction (IFLDP). The proposed scheme is cross validated with the conventional learning scheme called RNN to prove the efficiency of the proposed scheme ILFDP. The analysis of principal components is employed to extract characteristics and it is used as a statistical technique to accomplish what you want. When the trip could be organized to accommodate passengers' schedules, everyone would profit more. Train delay operations in the context of the duration and the train will be predicted, and the itinerary for transportation will be adjusted accordingly. The resulting section shows the efficiency of the work by means of the following metrics such as: training accuracy, testing accuracy, training loss, testing loss, specificity ratio and the sensitivity ratio measurements, in which all these metrics are cross-validated with the conventional learning scheme RNN to prove the performance of the proposed approach to predict the train delay in fine manner.