Generating Multimodal Transfer Hubs and Level-of-Service (LOS) Matrices to Integrate MaaS into Activity-based Models

Abstract

The integration of Mobility as a Service (MaaS) into activity-based travel demand models relies on the role of intermodal transfer hubs. Personal mobility depends on seamless combinations of transport modes to reach their destinations. To represent these real travel behaviors, we introduce the concept of compound modes (CM), which combines unimodal segments into an integrated travel chain consisting of access, main, and egress modes. These CM reflect the comprehensive range of common multimodal mode combinations , such as using a bike to reach a public transport stop (access), traveling by train (main), and then walking to the final destination (egress). However, the effectiveness of these multimodal trip chains depends on the availability and functionality of transfer hubs. Transfer hubs serve as critical interchange nodes , enabling seamless transitions between transport modes by connecting different vehicle states (e.g., parking locations and transit stops) Liao et al. (2010). By generating and evaluating these hubs in the context of CM, we ensure that the generated Level of Service (LOS) matrices reflect realistic travel chains, allowing for more reliable integration with activity-based frameworks such as ALBATROSS Arentze and Timmermans (2004) and FEATHERS Bellemans et al. (2010). The traditional unimodal LOS method ignores multimodal and micromobil-ity options, especially during the first and last mile of a trip, resulting in a limited representation of real travel behavior. A challenge lies in gener