J.M.D.L. (Johann) Hartleb MSc

Passengers only use public transport if the public transport system offers a high quality: travel time should be short and the number of transfers small; connections should be robust against delays; and vehicles should not be congested. If only poor connections are offered for an origin-destination-pair, passengers may decide to use alternative means of transport or stay at home. Hence, in order to obtain a good public transport system, the consequences of planning decisions on passengers' behavior must be taken into account.

State of the art travel demand models provide appropriate solutions for describing passengers' behavior concerning mode and route choices. Nevertheless, optimization approaches for public transport problems often rely on strongly simplified models. Often it is even assumed that demand is independent of the provided connections. To take into account passengers' behavior in the optimization, demand models must be integrated in the planning process.

The goal of this project is to investigate the structure of such problems, to develop efficient solution algorithms, and to test these on realistic problem instances. Based on this, the trade-off between detailed modeling and algorithmic tractability as well as the benefit of integration should be investigated.