Gast ::
| Hauptseite > Publikationsdatenbank > ETHOS.MODE.regional: A framework for activity-based modelling of long-termm mobility demand |
| Hauptseite > Publikationsdatenbank > ETHOS.MODE.regional: A framework for activity-based modelling of long-termm mobility demand |
| Abstract | FZJ-2025-02855 |
; ; ; ;
2025
Abstract: Long-term development of mobility demand cannot be evaluated without also accounting for the simultaneous evolution of the social, political, and economic framework. This is of special importance in regions that will undergo far-reaching structural change processes on their way to carbon neutrality. Therefore, transport models integrating detailed activity simulation with flexible assignment methods are needed, that can be used to evaluate the mobility demand in these scenarios. With ETHOS.MO|DE.regional we present a framework to build models for such use cases that can quickly be applied to regions worldwide by using open data like OpensStreetMap and publicly available transit schedules.Our framework utilizes the Load Profile Generator by Pflugradt et al. [1] to generate realistic schedules of different activities and their respective geographic locations for individual agents. The activities are planned over the course of a full year with a temporal resolution of one minute. The schedules of the agents are determined by multiple, agent-specific desires, which increase until fulfilled. During generation of the schedules, interactions between different household members are also considered. This allows for a highly detailed modeling of the mobility demand regarding social and economic influences on individual people.Trips from the activity profiles are then distributed on to networks for the different transportation modes, considering aggregated flows for each street segment rather than individual vehicles. Networks for automobile traffic, cycling and walking are extracted from OpenStreetMap considering attributes like road type, number of lanes or the speed limit among others. In addition, a network for public transport is constructed by using schedules following the General Transit Feed Specification. To account for park-and-ride behavior, switching between the car- or cycle- and the public transport network can be considered as well.During traffic assignment, for every trip, the fastest paths on each network are found. To account for congestion effects on the road network, when traffic flow on a link is close to or exceeds the link capacity, the car network is equilibrated with an iterative process. Mode choice for each trip is then performed by a nested logit choice model with travel time and costs, calculated by the assignment to networks as variables while also accounting for the mutual influence between the different transportation modes. The parameters for the choice model are estimated for user groups with different attributes, distinguished by factors like car ownership or transit subscriptions, using data form the German mobility survey Mobility in Germany 2017 [2]. The ETHOS.MO|DE.regional framework then allows for the agent to adapt their behavior to the prevailing state of network utilization by feeding back travel times and mode choice probabilities to the activity schedule generator within an iterative process. Travel times and mode choice probabilities from the network assignment will then be fed back into the activity generator within an iterative process to allow the agents to adapt their activity schedules to the prevailing state of network utilization.The framework will further be used to deduct the temporal and spatial energy demand to allow for an integrated evaluation of the transport and energy system. Joint modeling of transportation and energy demand like this is going to become more important in the future due to the ongoing electrification of the transportation sector. We will show the application of our framework on the Rhenish lignite region in West Germany as a case study, which will undergo a significant structural change due to the transition away from lignite mining in the coming years. The results of our model will then be used as valuable input for simulating energy infrastructure as well as research into the social and political acceptance of different pathways towards sustainable transportation.[1]: N. Pflugradt, P. Stenzel, L. Kotzur, and D. Stolten, ‘LoadProfileGenerator: An Agent-Based Behavior Simulation for Generating Residential Load Profiles’, Journal of Open Source Software, vol. 7, no. 71, p. 3574, Mar. 2022, doi: 10.21105/joss.03574.[2]: infas, DLR, IVT and infas 360 on behalf of the Federal Ministry of Transport and Digital Infrastructure (BMVI) (FE no. 70.904/15): Mobility in Germany 2017. url: www.mobilitaet-in-deutschland.de.
|
The record appears in these collections: |