Capacity-oriented passenger flow control under uncertain demand: Algorithm development and real-world case study

• Introduce a problem how to passenger flow control under uncertain demand.
• Propose an analysis framework of subway station service capacity for passenger flow control.
• Develop a simulation-based algorithm embedded genetic algorithm and data envelopment analysis to solve these models.
• Provide a real-world large-scale case study with detailed analysis results.

This paper proposes a problem of passenger flow organization in subway stations under uncertain demand. The existing concepts of station service capacity are extended and further classified into three in different demand scenarios. Mathematical models are put forward to measure the three capacities and a unified simulation-based algorithm is developed to solve them. To increase computing speed, data envelopment analysis (DEA) and genetic algorithms (GA) are embedded in this algorithm. A case study will demonstrate the performance of the proposed algorithm and give a detailed procedure of passenger flow control based on station service capacity in various demand scenarios.