Please use this identifier to cite or link to this item:
|Optimal bus service operation considering bus bunching elimination, bus contracting and collaborative freight transport
|Nanyang Technological University
|Zhou, C. (2022). Optimal bus service operation considering bus bunching elimination, bus contracting and collaborative freight transport. Doctoral thesis, Nanyang Technological University, Singapore. https://hdl.handle.net/10356/165471
|Public transport services are important assets to any major city. They play a vital role in improving urban mobility, reducing car dependency, and mitigating traffic congestion. It is ideal if public transport services are financially sustainable with expedient service quality. To achieve this goal, the key lies in how the full capacity of the existing public transport system can be exploited through optimal decisions on planning and operation. This thesis uses public bus services as an example to elaborate on how to achieve a more sustainable public transit system via determining the best decisions in three dimensions: self-operation, i.e., how to design the optimal bus service operation strategies at the operational level; outsourcing, i.e., how to determine the best bus contracting policy in the strategic planning level, and sharing, i.e., how to introduce collaborative freight transport service into existing bus transit system in the tactic level. Most specifically, we select one topic for each dimension, and these topics include bus bunching mitigation, bus routes packaging and allocation to private bus service operators, and collaborative freight transport. Various control mechanisms have been recommended to alleviate the issue of bus bunching. We notice that existing measures focus on controlling the bus operation directly from the operator’s perspective. Thus, we present a novel control technique by providing passengers with real-time wait time information and degrees of in-vehicle congestion. We describe the passenger boarding behavior under the action of crowding information provision, providing a bus traffic propagation model to simulate the bus movements with several system performance metrics. The indirect effects of information on passengers’ boarding choice, rather than controlling the bus vehicle directly, work as effectively as the conventional bus bunching measures, indicating a supplement strategy to the existing control measures in mitigating bus bunching problems. When bus service central planners need to contract out the existing bus routes to private service operators, one intrinsic problem to be addressed is how to best bundle bus routes into packages. In this problem, the government agency acts as a leader, packaging and allocating the bus routes to service operators to ensure the best system performance. Accordingly, the private service operators should consider the competing service operator’s strategy and travelers' choice behavior to maximize their profits. A tri-level programming model is constructed to describe the problem. An Augmented Lagrangian method for the intermediate level and an enumeration method, binary differential evolution, and Monte Carlo Tree Search for the upper-level model under different problem scales are developed to solve the model. The results provide insights and instructions to government agencies in policy making in bus route packaging and allocation. In the presence of a rapidly growing demand for urban delivery, existing bus services are recommended to offer collaborative freight transport services, especially during off-peak hours when the bus service capacity is excessive for passenger transportation. We investigate the impacts on bus service quality considering freight transport, which fills in the gap in the existing literature that generally assume the collaborative freight service has no effects on passenger service quality. An exact dynamic programming method is proposed for a closed circular bus route with a transshipment station at the middle position, aiming to obtain the least overall time costs involving delivery delays. We further determine the optimal fleet sizes, solved by Benders Decomposition with convergence proof. In general, bus service quality is determined by synergistic factors. Moreover, the overall time costs roughly increase by 10% at a moderate demand level.
|School of Civil and Environmental Engineering
|This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0).
|Appears in Collections:
Files in This Item:
|Under embargo until Mar 28, 2025
Updated on Feb 24, 2024
Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.