| Allocating resources to competing engineering infrastructure improvements is a multiobjective optimization problem. Most traditional decision-aiding methodologies focus on the tradeoffs among performance, risk and resource objectives. The issue of diversification in the distribution of resources is given much less attention. The traditional methodologies may fail to account for unknown and emergent requirements that are typical of large-scale infrastructure investment allocation problems. In modern portfolio theory, it is well known that a diversified portfolio can be very effective to reduce unsystematic risks, in particular the risk of extreme events. We believe that the concept of diversification is equally important in choosing robust portfolios of engineering infrastructure investments that may be subject to unknown and emergent risks including human and demographic shifts, land-use changes, new technologies, variability of service and material costs, and others. Moreover, diversification can address known and tangible issues such as the fair and equitable distribution of resources, and quantification of organizational goals. In this research, we develop diversification-based methodologies to analyze multimodal portfolios of transportation investments. We classify and explore several metrics of diversification and integrate them with other performance objectives in a multiobjective combinatorial optimization. We test the new metrics and the methodology in a case study of hundreds of millions of dollars of transportation infrastructure investments. We demonstrate that integrating the metrics of diversification can be useful to refine among the Pareto-optimal solutions generated using a traditional multiobjective optimization that has not considered diversification. The results suggest that the solutions that consider diversification may be more robust to emergent programmatic and technical risks, thus, identifying an opportunity to incorporate diversification-based methodology to support a variety of problems involving infrastructure investments. The dissertation will be of interest to decision-makers and practicing risk analysts involved in engineering, infrastructure, security, environmental protection, economics, and related topics.;Keywords: diversification metrics, multiobjective combinatorial optimization, risk management, resource allocation, systems engineering. |
