Trade-off analysis in multiobjective optimization for transportation asset management

Qiang Bai, Purdue University

Abstract

Trade-off analysis, one of the principles of transportation asset management, can help decision-makers understand how different resource allocations can affect system's performance and the relationships between performance measures in the decision-making process. However, the trade-off analysis in transportation asset management has not been studied in depth. In this dissertation, therefore, a general methodology is proposed for conducting trade-off analyses for the project selection problem of transportation asset management. A general trade-off-based multiobjective optimization framework for transportation asset management is first established. Next, a Hybrid NSGA II method is developed to generate Pareto frontiers to conduct trade-off analyses between the cost and performance measures and also between different performance measures. The chance-constrained programming is then applied to incorporate performance measure uncertainties into the multiobjective optimization. Using field data, the proposed methodologies are applied to conduct various trade-off analyses under deterministic and uncertainty situations. The proposed Hybrid NSGA II method ultimately is found to converge faster and generate better-distributed Pareto frontiers compared to the traditional NSGA II method. The trade-off relationships between performance measures are different at different budget levels. Further, at the same budget level, the trade-off relationships between two performance measures can be influenced by the specified levels of other performance measures. Under the uncertainty situation, trade-off relationships at different confidence levels are different; typically, a high confidence requirement results in lower performance at the same cost level. This dissertation also quantified the extent to which high costs are needed to achieve high confidence levels at a given level of pre-specified performance.

Degree

Ph.D.

Advisors

Sinha, Purdue University.

Subject Area

Civil engineering|Transportation planning

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server
.

Share

COinS