A Distributionally Robust Optimization Method for Passenger Flow Control Strategy and Train Scheduling on an Urban...

Area of Science:

• Operations Research
• Transportation Science
• Public Health

Background:

• COVID-19 necessitates enhanced safety and efficiency in urban transit operations.
• Congestion and cross-infection risks are significant challenges for urban rail systems.
• Existing models may not adequately address uncertainties in operational scenarios.

Purpose of the Study:

• To develop an innovative operation strategy for urban rail transit.
• To mitigate congestion and reduce infection risks during the COVID-19 pandemic.
• To introduce a two-stage distributionally robust optimization (DRO) model for transit management.

Main Methods:

• Construction of a two-stage DRO model with partially known probability distributions.
• Utilization of the mean-conditional value-at-risk (CVaR) criterion for risk-reward balancing.
• Transformation of the robust counterpart into a tractable form using a discrepancy-based ambiguity set.
• Development of a hybrid algorithm combining local search and mixed-integer linear programming (MILP) for efficiency.

Main Results:

• The proposed DRO model effectively balances expected passenger waiting times with congestion risks.
• The hybrid algorithm demonstrates computational efficiency for large-scale urban rail transit problems.
• Numerical examples using real-world data validate the model's effectiveness.

Conclusions:

• The novel DRO model offers a robust solution for urban rail transit operation strategy innovation.
• The approach successfully addresses challenges posed by uncertain operational scenarios and public health concerns.
• The developed hybrid algorithm enhances the practical applicability of the model for real-world transit systems.