Cross-line train schedule optimization of urban rail transit based on time-dependent passengers demand

Following the rapid development of urban rail transit, networking operation will be an inevitable trend for rail transit. How to formulate reasonable train dispatching program and achieve total optimal efficiency has become the key issue to be addressed. The cross-line operation of the subway can effectively reduce the transfer station pressure and improve the service quality. This paper studies the train schedule optimization problem under the cross-line operation mode for a subway network, where the time-dependent passengers demand is considered. A mixed integer linear optimization (MILP) model is formulated with the objective of minimizing the passengers travel times and train operating costs. A genetic algorithm (GA) is designed to obtain high-quality solutions. The computational results illustrate that adopting cross-line operation can reduce the total travel time of passengers by 49.3% and train operating costs by 62.5%.