Evaluating crowding in individual train cars using a dynamic transit assignment model

As travel demand grows in many cities around the world, overcrowding public transport systems has become a major issue and negative effects for both users operators. Measures to address on-board congestion span from large-scale strategic investments (e.g. increasing infrastructure capacity), through tactical planning stopping pattern) real-time operational measures information provision, gate escalator control). Thus there is need evaluate impact of these prior their implementation. To this end, study aims at capturing effective capacity utilization train, by considering passengers' distribution among individual train cars into an agent-based simulation model. The developed model validated applied case Stockholm metro network. findings suggest that increase peak hour leads more even passenger cars, which partially counteracts increased disutility caused higher volumes. Interestingly, closure most popular entrance point one stations lower crowding unevenness downstream stops consequently reduces experienced discomfort. We find user cost significantly underestimated when not accounted for.