An Application of Congested Transit Network Loading with the Markov Chain Approach

Uncertainty and unreliability is a topic for all modes of transport. Car drivers as well as public transport users might not arrive at their destination in time, or in severe cases, not arrive at all. Du and Nicholson (1997) distinguished two sources of uncertainty. Firstly, supply variations such as road or track blockage leading to network degradations. Secondly, demand variations engendered by peak periods or sporting events, which can lead to capacity problems for the road network as well as for the public transport systems. Both sources of variation can lead to unpredictable delays and uncertainty in travel times. The nature of the delays is, however, different for public and private transport. Car-users will spend more time in their vehicles waiting in queues or driving with reduced speed, but the in-vehicle time for public transport users keeps relatively constant (except maybe for longer dwelling times). In congested situations transit network users might, however, not be able to board the first vehicle arriving at their bus stop or platform because of insufficient space. This uncertainty in waiting time, because of a non-zero probability that the transit vehicle is overcrowded, seems to be underresearched in the context of transit assignment.