Skip to main content
To KTH's start page To KTH's start page

Simulation-based Evaluation of Fixed to Flexible Transit

Time: Fri 2022-10-07 10.00

Location: D37, Lindstedtsvägen 5, Stockholm

Video link:

Language: English

Subject area: Transport Science, Transport Systems

Doctoral student: David Leffler , Transportplanering, Centrum för trafikforskning, CTR

Opponent: Associate Professor Chris Tampère, KU Leuven, Belgium

Supervisor: Docent Erik Jenelius, Transportplanering, Centrum för trafikforskning, CTR; PhD Wilco Burghout, Transportplanering, Centrum för trafikforskning, CTR; PhD Oded Cats, Transportplanering, Centrum för trafikforskning, CTR

Export to calendar

QC 20220907


Emerging technologies have inspired a wide array of flexible public transit system designs characterized by varying degrees of demand-responsive routing and scheduling. The availability and synthesis of new data sources with higher degrees of spatial and temporal richness brought on by advancements in Intelligent Transport Systems allow for monitoring and responding to evolving supply-demand imbalances in real-time. The emergence of smartphone-enabled ride-pooling services and the development of automated vehicles have shown promise in offering flexible transit systems at a higher level-of-service and a lower per-vehicle operational cost.

Preliminary studies have indicated that reallocating resources from fixed transit services with low utilization rates to flexible transit services can improve public transit accessibility in low demand-density areas. Studies have also shown that shared automated vehicle services can drastically reduce the number of vehicles required to serve urban transport demand, reducing congestion and pollution. New technologies and flexible transit designs could foster city infrastructure planning oriented around people instead of cars. However, if high enough levels of ride-pooling and integration with existing high-capacity transit are not achieved, there are also indications that such services can poach passengers from more sustainable modes of transportation, increase total vehicle-kilometers traveled, and amplify trends of urban congestion.

A question that arises is how to evaluate the effects of novel flexible transport solutions as a competing or complementary alternative to traditional fixed public transit under alternative demand settings, technological settings, road network topologies, and objectives. Flexible transit systems are difficult to trial in parallel with the technologies that inspire their design, due to their cost of implementation and the time frame required for stable use patterns to emerge. 

Agent-based simulation frameworks have been utilized to systematically understand and develop theories around the dynamics of transport systems and traveler behavior using diverse data sources, while ideally also forecasting the effect of alternative transit designs and operational policies. Interest in applying agent-based simulation models to evaluate flexible transit systems has grown significantly over the past decade, however, are still limited in their ability to represent the vast flexible transit service design space. In this thesis, flexible public transit systems ranging from services with partially fixed routes and timetables to services with demand-responsive routes and timetables determined in real-time are appraised through extensions to the public transit simulation framework BusMezzo.

In each of the included papers, a flexible transit service design inspired by Intelligent Transport Systems and automated vehicle use cases is developed. This system is formalized with simplifying assumptions to make the problem tractable in terms of modeling, and then implemented in BusMezzo. The system, model, and implementation are evaluated in several case studies based on recurring fixed public transit supply-demand scenarios. Through the work of this thesis, key level-of-service trade-offs between fixed and flexible transit operations are explored. The resulting simulation framework includes essential components for modeling supply-demand dynamics of mixed fixed and flexible transit systems and enables systematic evaluation of a wider range of emerging public transit designs and scenarios.