New, i.e., shared and digitized, mobility services have been entering urban mobility markets around the globe. Among these new offerings is ridepooling, a mobility solution that bundles requests from passengers with similar routes in real-time and matches them with a vehicle. Ridepooling is quite novel in Germany and knowledge about users, changes in travel behavior, and impacts on the urban traffic system is scarce. To address this gap, we conducted an online survey among users and non-users of MOIA, a German ridepooling provider. Over 12,000 respondents completed the survey. The article presents results on ridepooling users’ characteristics and usage patterns. We found that MOIA users cover all age groups and are multimodal travelers—which leads us to assume that ridepooling enriches mobility portfolios and also serves as an alternative to the private car. MOIA is mostly used occasionally and, in particular, during the evening or the night. A specific focus of the article lies on users with mobility impairments as well as how and by whom ridepooling is used on work-related trips. Both topics are particularly relevant in light of changing travel patterns and transforming urban transport systems towards more sustainability.

Individual travel behavior, such as mode choice, is determined to a distinct degree by the respective portfolio of available mobility tools, such as the number of cars, public transit pass ownership, or a carsharing membership. However, the choice of different mobility tools is interdependent, and individuals weigh alternatives against each other. This process of parallel trade-offs is currently not reflected in typically used sequential logit models of agent-based travel demand models. This study fills this research gap by applying discrete choice and neural network models on a synthetic population to model multiple mobility tool ownership simultaneously. Using data from a national household travel survey, both model types approximated the given target distributions of mobility tools more accurately than the sequence of three corresponding logit models. Owing to its greater flexibility, the tested shallow and deep neural network exhibited higher predictive accuracy than simultaneous discrete choice models. The results indicated that neural networks with only one hidden layer were more robust and easier to formulate and interpret than deep networks with three hidden layers. Finally, the flat neural network was applied to a different synthetic population resulting in equally accurate results.

Autonomous busses and on-demand (OD) services have the potential to improve the public transport system. However, research on potential traffic impacts is still ongoing, mainly because of a lack of existing use cases of autonomous driving as part of public transport. The availability of revealed preference data for mode choice decisions is thus very limited. Therefore, we conducted a stated choice experiment to assess mode choice preferences with regard to use cases as the main mode of transport and as the solution for the first and last mile. We also distinguished between OD and schedule-based (sched.) services. The target population of the survey is the population of Baden-Württemberg, a state in southwestern Germany. The responses of 1,434 people were analyzed using a nested logit approach. On this basis, we established exemplary utility functions and descriptively derived recommendations for efficient forms of deploying autonomous busses in addition to already existing well-developed public transport systems. It was found that, under the given conditions, public transport pass owners without a car in their household would be the most interested in using autonomous busses. Car owners without a smartphone see less benefit. It was also shown that the recruiting method of the respondents is crucial. Those reached via social media were significantly more positive than those contacted via an online panel. Further evaluations show that autonomous busses are rated similarly to existing public transport and consequently have particularly high potential on medium distances, especially if their deployment leads to shorter access routes.

This paper considers which work-related trip patterns are included in household travel surveys and which in commercial travel surveys and if there are certain patterns that are distinctly underrepresented in either one. The study is structured as a comparison between data from a household travel survey and data from a commercial travel survey. Both surveys were conducted in Germany and within close temporal proximity. We applied cluster analysis to identify differences in the data and identify work-related travel patterns. The results show that work-related travel patterns are quite complex. Although some patterns are covered in both surveys, mobile workers’ travel patterns in particular are not represented well in the household travel survey. Furthermore, our analysis shows that not all commercial trips are generated by motorized vehicles and a considerable share of work-related trips are undertaken using public transport or active modes of transport that are not covered by the commercial travel survey. The results indicate that researchers and transport planners creating travel demand models need to pay more attention to work-related travel behavior and acknowledge that depending on the area of study, traditional household travel surveys may not provide a complete sample of the population; however, simply adding data on commercial trips from commercial travel demand models to data from household travel surveys does not provide a complete picture of work-related travel either.