Urban rail transportation (URT) has long become the preferred public transportation choice for major metropolitan areas such as New York, London, Paris, Moscow, Tokyo, and Beijing. The highest daily record for Beijing's URT reached 5.71 million passenger trips in 2010, which makes the network extremely crowded in rush hours. To accommodate the increasing demand for URT, the service frequencies have been increased tremendously. To address these safety, efficiency, and reliability issues, the paper presents a novel parallel system for URT operations that uses the concept of parallel system and computational experiments based on artificial systems (ACP). The parallel URT system can analyze and facilitate passenger-flow management, vehicle scheduling, and other operational issues while considering human-related, environmental, and other social and economical factors.