Transportation Network Design Models And Algorithms For Maximizing Space-time Accessibility

As a comprehensive performance measure for the level of service in transportation systems,space-time accessibility is primarily used to reflect whether or not the travelers'needs to reach desired activities can be satisfied,based on the analysis of individual's temporal and spatial activities.In order to make more travelers reach their particular places more conveniently,strategies,such as constructing transportation infrastructure and promoting transportation services,are usually adopted to significantly improve the system-wide accessibility.Therefore,the accessibility-orientd transportation network design and corresponding service plan optimization problems are greatly important and necessary for real-world applications as well as theoretical research efforts.This dissertation integrates concepts of both space-time path and space-time prism from the field of time geography to characterize individual's accessibility in transportation systems.Specifically,a general transportation network problem based on space-time accessibility and a specific customized bus service network design problem are systematically examined.Further,a unified theoretical framework for optimizing agent-based network design is proposed to improve individuals' accessibility and makes fundamentally theoretical contributions for accessibility-oriented transportation planning and management.The major research contents of this dissertation are listed as follows.(1)The modeling principles of building space-time networks are discussed systematically,and then equivalent multicommodity network flow models are established for three types of networks:(a)physical network,(b)independent space-time network and(c)space-time network with accessibility measures.By adopting the dynamic network-time prism analysis framework,this research considers the system-wide transportation accessibility between major activity locations,subject to a given construction budget.(2)This research specifically introduces space-time prisms to form a time-discretized space-time transportation network so as to evaluate the accessibility of individuals under different transportation modes and time-dependent link travel times.This study develops a space-time accessibility-oriented transportation network design model to minimize the overall inaccessible activities across different groups of travelers.(3)By constructing a time-extended/discretized space-time network,this study formulates the problem as a linear integer programming model to maximize the number of accessible activity locations within given travel time budget for each user.A Lagrangian relaxation solution framework is used to effectively decompose the original complex problem into classical subproblems such as knapsack and time-dependent least cost path problems.Several scenarios using the Chicago test network are provided to consider the effectiveness of the proposed method in modeling accessibility-enhancement strategies such as congestion mitigation and land use policies.(4)This study also focuses on how to optimize customized bus routing and timetabling plans to satisfy a wide range of specific user constraints,such as passengers'pickup and delivery locations with preferred time windows,through flexible decision for matching passengers to bus routes.From the perspective of time geography,this study develops a joint optimization model for addressing a number of practical challenges for providing flexible public transportation.The flexible and differentiated spatial and temporal behavior of passengers are highly respected to satisfy the minimum loading rate requirements and increase the number of customers per bus for the bus operators to reach long-term profitability.(5)This dissertation further studies the design of customized bus service network.A solution algorithm is developed based on the Lagrangian decomposition for the primal problem and a space-time prism based method is proposed to reduce the solution search space.Case studies using both an illustrative example and a real-world large-scale transportation network in Beijing are conducted to demonstrate the effectiveness of the proposed algorithm and its sensitivity under different practical operating conditions.