Metode Neural Network Dalam Prediksi Jumlah Penumpang Kereta Api Berbasis Web

The demand for railway transport in Java has increased along with the increasing population growth and increasingly complex mobility needs. Trains have become one of the main modes of transport due to their efficiency and reliability in travelling medium and long distances. However, there is a significant imbalance between ticket demand and train capacity availability, especially during holiday seasons, holidays, and weekends. To solve this problem, a simulation is needed to predict the number of passengers in the future using the Neural Network Backpropagation method. The implementation of this system resulted in a prediction accuracy rate of 80.33%, which provides PT Kereta Api Indonesia with an important tool to better manage schedules and capacity. It is hoped that this research can make a significant contribution to improving the company's operational efficiency, while providing a better experience for passengers. In addition, this research is also expected to provide stakeholders with greater insight into the dynamics of transport demand in Java, and help formulate more effective policies to support the growth of the public transport sector. With the results of this study, PT Kereta Api Indonesia is expected to develop optimal strategies in adjusting train capacity to the unstable passenger demand, while local governments can utilise this information to design policies that support the sustainability of transport services in Java.