SIMBA MOBi: Microscopic Mobility Simulation for Corporate Planning

Travel demand models are used to predict traffic volumes on the infrastructure and user benefits from new service concepts. Most travel models applied in regional or national planning practice today simulate aggregated travel flows, i.e. they are of macroscopic nature. In contrast, the authors present a microscopic model, which simulates each traveller as an individual entity. The model is called SIMBA MOBi. It has been developed over the past two years by SBB (Swiss Federal Railways). As far as the authors know, this is one of the first microscopic models that is developed inside a transport operating company in Europe and applied to support real world business decision making. It is person-based, multimodal, simulating 24 hours of the average weekday (Monday through Friday) and covers the entire country of Switzerland, including urban, rural and intercity travel. The paper starts out by recalling the objectives of this modeling effort and how SBB’s management decided to go the innovative way of microscopic modeling despite higher risk and increased cost of model development. Among the reasons for this decision is the need to represent the full complexity and dynamics of travel, especially all legs of a trip from door to door, and the abilities to model new technologies such as autonomous cars, demographic evolution and disruptive shifts in cost and pricing. To develop SIMBA MOBi, concepts, methods and sometimes tools were drawn from several academic schools of travel modelling but also from methods that are already used in practice. The sources of this related work are given in the paper. The architecture and methodology which were finally chosen for SIMBA MOBi and implemented are explained on a high level: Starting from a synthetic resident population of Switzerland, activity plans and travel plans are constructed for each person, balancing for each traveler both the individual preferences as well as constraints. Preferences are represented by discrete-choice steps such for long-term mobility choices (car ownership, PT subscription, work and school location) as well as for daily choices (tour and activity frequencies, destinations for secondary activities, modes). Constraints dominate the scheduling step, using an approach based on rules and time budgets. Finally, 24-hour dynamic network flows for cars and public transport are simulated using the agent-based software MATSim. In addition to the traffic flows, the MATSim in the authors model computes also an adjustment of mode and time choices in the original plans according to the travel conditions experiences by each individual agent. Hence, the model is microscopic through all model steps. At the time of the conference, the model is finalized in the existing state, and pilot applications on real world business cases have been run. The paper shows the calibration approach and examples of model validation. The authors believe that microscopic and activity-based models must - despite the increased complexity – fulfill the same calibration and validation requirements as conventional macroscopic models do. The integration of simulation results in corporate planning processes and the road forward towards simulation of future mobilities will be discussed.

• Record URL:
  https://aetransport.org/past-etc-papers/conference-papers-2019
• Supplemental Notes:
  • Abstract used by permission of Association for European Transport.
• Corporate Authors:

  Association for European Transport (AET)

  1 Vernon Mews, Vernon Street, West Kensington
  London W14 0RL,    
• Authors:
  • Scherr, Wolfgang
  • Manser, Patrick
  • Bützberger, Patrick
• Conference:
  • European Transport Conference 2019
  • Location: Dublin , Ireland
  • Date: 2019-10-9 to 2019-10-11
• Publication Date: 2019

Language

• English

Media Info

• Media Type: Digital/other
• Pagination: 15p
• Monograph Title: European Transport Conference 2019

Subject/Index Terms

• TRT Terms: Calibration; Corporations; Decision making; Microscopic traffic flow; Mobility; Mode choice; Planning methods; Simulation; Traffic volume; Travel demand; Validation
• Non-Preferred Terms: Swiss Federal Railways
• Geographic Terms: Switzerland
• Subject Areas: Data and Information Technology; Planning and Forecasting; Transportation (General);

Filing Info

• Accession Number: 01753731
• Record Type: Publication
• Files: TRIS
• Created Date: Sep 29 2020 11:19AM