Optimization of Metro Trains Operation Plans Based on Passenger Flow Data Analysis

Metro intelligent system produces massive passenger flow and traffic data every day, among which route, station, operation are important for optimizing the train scheme. We collect information of Shenzhen metro, analyze pattern its distribution characteristics based on warehouse Hadoop platform, optimize scheme in this paper. Using dynamic data, an optimization model with departure dwell time as decision variables waiting time, ride full load ratio, balance objectives is developed. An improved parallel genetic algorithm (GA) incorporating a simulated annealing (SAA) optimal individual retention strategy used to find result. To verify usefulness method, simulation experiments conducted method using real results compared original plan.