Längsschnittmodellierung der Verkehrsnachfrage zur Abbildung multimodalen Verhaltens. Dissertation Universität Fridericiana zu Karlsruhe.

Longitudinal multiday travel surveys such as the German Mobility Panel have contributed fundamentally to a better understanding of individual travel behaviour in the recent years. Longitudinal travel demand models promise new possibilities of appraisal and improvements in modelling the interrelationship of external factors and individual mobility behaviour. However, findings from longitudinal analyses have up to date hardly been incorporated in travel demand modelling. This thesis presents a microscopic model which simulates travel demand over the course of one week. This enables the integration of new interrelationships in travel demand modelling: On the one hand the presented approach models sensitivity of travellers to budget constraints using travel time budget as an example. This is not possible when the model period is only one day like in most demand models up to now. The reason for this is that individual travel and time use vary too significantly from day to day. On the other hand the presented approach reproduces the mode use habits of travellers in the longitudinal section: The model predicts the clientele of the different modes and their multimodal mode use. In order to do so new elements reflecting the longitudinal approach were introduced in the utility function of the mode choice model. Moreover, the longitudinal perspective of the model allows for new possibilities of appraisal: Firstly the microscopic longitudinal approach enables a high temporal resolution of the modelled demand, Secondly the model allows to predict how the composition of the user collective changes with time for specific parts of the infrastructure. The presented model was applied for forecasting travel demand and mode use in the city of Erlangen in the year 2020. The results indicate that total travel demand will stagnate while the mode share of public transport decreases even though there will be more multimodal travellers. This is due to the ageing of the population and growth in car availability. These results conform to expectation and thus confirm the overall approach and applicability of the presented model. (Author/publisher)