Online Passenger Flow Control in Metro Lines

Crowds management during peak commuting hours is a key challenge facing metro systems worldwide, which results in serious safety concerns and unfair public transit service for commuters on different origin-destination (o-d) pairs. In “Online Passenger Flow Control Metro Lines,” the authors investigate impact of online decision making value passenger flow control solution methodologies. The formulate problem as stochastic dynamic program with fairness (fill rate) constraint exploit Blackwell's approachability theorem Fenchel duality to characterize attainable level each o-d pair. They use these insights develop policies that can enable more passengers boarding train (efficiency) well ensure equitable (fairness) provided Numerical experiments set data from Beijing show this approach performs compared existing benchmarks literature.