A G/G(n)/C/C state-dependent simulation model for metro station corridor width design

Summary Metro station corridor and passengers are described as a G/G(n)/C/C state-dependent queuing system with a general random arrival interval G and a general random and state-dependent service time G(n) to offset the shortcomings in existing design methods. The corresponding G/G(n)/C/C state-dependent discrete event simulation model is developed, and its high-fidelity is tested. Then the optimization algorithm based on the simulation model is designed to determine corridor width. The proposed simulation optimization method and the existing analytical optimization methods, based on M/G(n)/C/C and D/D/1/C queuing models, are applied to design corridor width in a numerical example of 48 combinations of passenger flow rates and level of service (LOS). The designed corridor widths are tested in a micro-simulation model, and the performance measure is compared. The result shows that the corridor widths obtained by the new method are 0.357 m (7.4%) larger than that of the other two methods on average; the area per passenger of the new method increases 10.53% and 11.63%, respectively, compared with that of the other two methods; the widths designed by the new method satisfy the requirement of LOS under various passenger flows, whereas 93% of the corridor widths obtained by the other two methods fail to meet the requirement of LOS, and the corridor widths designed by the new method have high elasticity coefficients of LOS-width. Copyright © 2015 John Wiley & Sons, Ltd.