The objective of transportation decision-making is to decide which projects to invest in and implement, such that total utility of the... Show moreThe objective of transportation decision-making is to decide which projects to invest in and implement, such that total utility of the transportation network after project implementation viewed from economic (total cost to the transportation agency and the user), social (traffic mobility and safety), and environmental (energy consumption and vehicle emissions) dimensions is maximized. In current state-of-practices, decisions are made in accordance with project benefits that are estimated by independently capturing localized impacts of individual projects. The ignorance of network-level impacts and interdependency relationships of individual projects does not ensure accurately estimating project benefits and making truly optimal investment decisions based on estimated project benefits. This thesis first discusses the need for conducting a comprehensive study to estimate network-wide benefits of interdependent transportation projects and identify the most cost-effective sub-collection of projects under budget and other constraints for implementation. Using data on travel demand, roadway network designs, and traffic operations, as well as six major projects proposed for possible investments to the Chicago Loop area, the proposed methodology is applied to the Frontline Solver optimization software in a computational study. In general, the hypergraph Knapsack model as part of the proposed methodology is found to be able to identify the best sub-collection of investment projects for implementation under various budget levels. Specifically, the computational study has revealed that the overall benefits in terms of travel-time savings after considering project network-wide impacts and their interdependency relationships are much lower (by 38-64 percent) and the rate of benefit increase follows a diminishing trend with a higher level of budget, suggesting that an optimal level of investments can be established for mobility improvements. Future research directions are pointed out on developing efficient solution algorithms to identify a reasonably small number of possible sub-collection of projects that contains the optimal solution. It is also proposed to consider additional transportation costs items that extend from travel times to include costs of facility construction, rehabilitation, and maintenance and vehicle operations, crashes, and emissions in the estimation of network-wide benefits of single and multiple projects. M.S. in Civil Engineering, December 2011 Show less
