| Author: | Zhen, Li |
| Title: | Joint optimization of trunk-feeder urban transit networks |
| Advisors: | Gu, Weihua (EEE) |
| Degree: | Ph.D. |
| Year: | 2025 |
| Subject: | Urban transportation Local transit Transportation -- Planning Hong Kong Polytechnic University -- Dissertations |
| Department: | Department of Electrical and Electronic Engineering |
| Pages: | ix, 117 pages : color illustrations |
| Language: | English |
| Abstract: | In the context of rapidly sprawling urban areas and expanding intercity express rail networks, efficient public transportation systems are increasingly vital for sustainable urban mobility. This thesis addresses the critical challenges of designing and optimizing transit networks through a comprehensive examination of feeder service design and trunk-feeder system integration. By developing continuum approximation models and optimization approaches, this research contributes to the evolution of transit network planning, offering solutions that can significantly enhance service quality while reducing generalized costs. The research explores the optimization of feeder transit services across two complementary categories: fixed-route services and demand-responsive services, each suitable for different operational contexts and demand patterns, before culminating in an integrated trunk-feeder system design that synthesizes these approaches. For fixed-route feeder services, continuum approximation models are developed to optimize heterogeneous networks in rectangular service regions adjacent to rail terminals. This work extends previous research by: (i) simultaneously optimizing heterogeneous stop spacings along with line spacings and headways to account for various demand patterns; (ii) incorporating the effects of passenger boarding and alighting numbers on bus dwell times and patron transfer delays at rail terminals; and (iii) examining the advantages of asymmetric coordination between trunk and feeder schedules in both service directions. To address the increased modeling complexity, a novel semi-analytical method is introduced that combines analytically derived properties of the optimal solution with an iterative search algorithm. The research reveals several important insights, including that integrating heterogeneous stop spacing optimization reduces system costs by approximately 4% under specific operating conditions, with cost savings increasing with demand heterogeneity but decreasing with demand rate and service region size. Additionally, designing feeder lines where buses pick up and drop off passengers along the service region's shorter side significantly lowers system costs (by 6% when the region's aspect ratio is 0.5), while coordinating trunk and feeder schedules in both service directions yields additional cost savings of up to 20%. Complementing this analysis, the research also examines flex-route feeder (or in some studies termed demand-responsive connector, DRC), which provide seamless connections between travelers' origins/destinations and major transportation hubs under conditions where fixed-route services may be less efficient. This phrase distinctly identifies and analyzes two commonly used DRC operating strategies: (i) the "fully-flexible routing" strategy, where vehicles serve only requests received before dispatch through optimal tours, and (ii) the "semi-flexible routing" strategy, where vehicles follow predefined paths through swaths to serve requests received en route. Unlike previous studies that often adopted oversimplified approaches, this work develops refined analytical models for both strategies that accurately incorporate the second-order effects of stochastic demand and utilize improved local tour length formulas. Numerical experiments demonstrate that these enhanced models significantly reduce cost estimation errors—to within 2% for fully-flexible routing and 0.25% for semi-flexible routing—compared to previous models with errors of 8–12% and 6.3%, respectively. The research provides precise determination of critical demand densities for selecting between the two DRC strategies and fixed-route feeder service, revealing that as demand density and region size increase, the optimal strategy transitions from fully-flexible to semi-flexible routing before ultimately shifting to fixed-route service. Furthermore, the study identifies zoning as pivotal in DRC service design, with fully-flexible routing favoring square-shaped zones and semi-flexible routing preferring elongated rectangular zones. Culminating this comprehensive investigation, the thesis presents a continuous model for optimizing integrated trunk-feeder transit networks under spatially heterogeneous demand. This model accounts for the integration of trunkline and feeder transit modes, offering significantly greater flexibility and applicability compared to prior studies that focused primarily on uniform networks or single service types. To address the heightened complexity in modelling, a sophisticated semi-analytical approach is developed that merges key analytical characteristics of the optimal solution with an iterative algorithm. Numerical results demonstrate that this integrated approach achieves approximately 21% cost savings compared to previous models, highlighting the substantial benefits of considering heterogeneous demand distributions and multi-modal integration in transit network design. |
| Rights: | All rights reserved |
| Access: | open access |
Copyright Undertaking
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item:
https://theses.lib.polyu.edu.hk/handle/200/14245