Utilizing and optimizing non-disrupted lines for evacuating passengers in Urban rail transit networks during disruptions

In urban rail transit operations, conventional disruption management measures, such as train rescheduling and bus bridging services, play a crucial role in alleviating passenger evacuation pressures. Despite their utility, these measures often fall short during peak hours or in densely populated downtown areas due to delayed responses and capacity limitations. Addressing this gap, this study introduces an approach to efficiently manage large-volume evacuations by guiding passengers to alternative paths comprised of the non-disrupted lines within the urban rail network to complete their trips, alongside adjusting train schedules of these non-disrupted lines to enhance capacity for the influx of rerouted passengers. Essentially, this approach utilizes and optimizes non-disrupted lines to evacuate passengers. To tackle this issue, this study develops four mathematical optimization models aimed at optimizing passenger re-routing and adjusting train schedules. These models cater to different scenarios: whether passengers independently choose their paths or adhere to path guidance, and whether train schedules are adjusted. The inclusion of a Path-Sized Logit model within the optimization framework accurately reflects passenger path-choice behaviours, while an iterative algorithm is introduced to tackle the nonlinear models. Applied to a case study of the Zhengzhou Metro, the implementation of the disruption management schemes obtained from these models and algorithm significantly increases the number of affected passengers completing their trips and minimizes passenger delays during disruptions, thereby enhancing the urban rail transit network's resilience. Moreover, the findings from this study offer valuable insights into line redundancy analysis and enable a targeted measure to manage diverse passenger needs during disruptions. These insights provide a foundation for urban rail transit operators to manage disruptions more reliably and efficiently, ensuring a higher level of preparedness for future disruptions.