Please use this identifier to cite or link to this item:
Title: Optimal design of shared public transport services
Authors: Zhang, Jie
Keywords: DRNTU::Engineering::Civil engineering::Transportation
Issue Date: 24-May-2019
Source: Zhang, J. (2019). Optimal design of shared public transport services. Doctoral thesis, Nanyang Technological University, Singapore.
Abstract: Public transport services are always recognised as one of the most efficient ways to solve traffic problems. While it has been reaching a bottleneck in the most metropolis as the traditional public transport could not meet the increasing requirements from diversified traffic demand. Those demands are hardly switched to public transport services. As a result, shared public transport services have been launched to improve the public transport usages. Among them, dockless bike sharing service and on- demand public bus (ODPB) service have emerged as alternative options to attract more passengers with diverse travel requirements. Dockless bike sharing system is a self-service public bike rental scheme, which enables users to unlock a public bike through scanning a smartphone application and lock it after using it anywhere. ODPB is a demand responsive public transport service, which offers users the public transport services according to their requests. Due to the recent emergence of dockless bike sharing and ODPB service, the research works in the literature are still very few, and most of the existing works are very descriptive using statistical approaches. This research aims to develop a set of methodologies based on analytical approaches, rather than statistical approaches, to determine the optimal strategies for dockless bike sharing and ODPB service’s planning and design. In line with this aim, five research tasks have been defined. These are initial bike allocation problem; bike relocation problem; feasibility analysis of ODPB service; applicability analysis of ODPB service; optimisation stop spacing design. In this regard, this research proposes five sets of models and algorithms to address the planning and design issues. The first set of models and algorithms is to guide the market leader and follower to allocate the bikes. Two market scenarios are designed including both monopoly and competitive situation. The initial bike allocation position, coverage area, and allocation method are determined based on community structure and diffusion theory. A new dynamic pricing strategy with negative price is introduced in dockless bike sharing system in the second set of model and algorithm to achieve bike sharing relocation through travellers’ behaviour. In a normal situation, the user pays a positive price to the operator for using a bike. When bikes are imbalanced distributed in the system compared to the real-time demand, the user who cycles from the oversupply area to undersupply area will get paid from the operator. A user equilibrium dynamic traffic assignment formulation is developed to capture travellers’ mode-path choice in response to the proposed dynamic pricing strategy. The third set of model and algorithm is to provide a systematic account of the impact of ODPB service on a multimodal transport corridor by using a traffic equilibrium formulation. Six alternative travel modes: auto, park-and-ride (P&R), taxi, metro, conventional bus, and ODPB are considered, while the generalised travel utility functions for each mode are developed with incorporating travel time, monetary cost and travel discomfort. To investigate the applicable condition of ODPB service, the fourth set of model and algorithm is to maximise the social welfare after introducing the ODPB service by analysing different scenarios. The length and fare of ODPB service will be considered, together with the size of the residential area and population density along the service. The fifth set of model and algorithm is to optimise the bus stop spacing of ODPB service with the objective to minimise the total travel time of all the passengers. To capture the unique characteristics of ODPB service, the model constraints will include the length of the segment, the configuration of shed line, minimum stop spacing, vehicle capacity limitation, and nonnegative constraints. Preliminary results have found that suitable bike allocation strategies could cover maximum users and service area. The dynamic pricing strategy with negative price could efficiently achieve bike relocation when the resources are limited in the initial promotion stage. The introduction of ODPB service can bring a positive impact to mode shift, especially for the long-distance trips. The mode shift performance depends greatly on the fare, usage of the bus lane and the out-of-vehicle travel time of ODPB service. Users with different VOT have different mode choice preferences. The optimal bus stop spacing for an on-demand public bus is very different from a conventional bus. Passenger demand density, travel time perception weight ratio, access/egress walking speed, and the time loss near/at the stop will affect the on- demand public bus stop spacing significantly. Future work could extend the proposed methodologies to other situations. To the best knowledge of the author, this is the first work on systemically exploring the planning and design issues for dockless bike sharing and ODPB services.
Fulltext Permission: open
Fulltext Availability: With Fulltext
Appears in Collections:CEE Theses

Files in This Item:
File Description SizeFormat 
Optimal Design of Shared Public Transport Services_Final.pdfOptimal Design of Shared Public Transport Services (Thesis)20.61 MBAdobe PDFThumbnail

Google ScholarTM



Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.