Case Studies in Cross-Asset, Multi-Objective Resource Allocation (2019)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

99 The research described in this report focused on conducting four case studies for implementing cross-asset resource allocation approaches at transportation agencies. The case study agencies represent a mix of motivations, approaches, and uses of decision support tools. • Motivation. All the case study agencies were motivated by a desire to make data-driven and performance-based decisions and to make the allocation of funds more open and transparent. MDOT also had the added motivation of responding to state legislation requiring major expansion projects to be prioritized against given criteria. • Approach. Most of the case study agencies opted to prioritize projects. However, ADOT notably prioritized investment areas. The majority of the case study agencies utilized a mix of quantitative and qualitative measures to determine the scores for each project or invest- ment. Caltrans, however, developed measures that are analogous to monetized benefits. This method avoids many of the pitfalls associated with scaling the benefits and assigning weights to the goal areas in the methodology. • Tools. Finally, the case studies provided a diverse look at the various decision support tools available to aid in the allocation process. Two case studies featured COTS systems in various stages of development. In addition, the cross-asset resource allocation tool developed in NCHRP Report 806 was utilized and offered additional insights for two of the case study agencies. The case studies help illustrate the progress different transportation agencies have made in using cross-asset allocation approaches since the publication of NCHRP Report 806. They demonstrate that is it feasible to implement structured approaches for cross-asset resource allocation, though there are still many challenges with application of such approaches—most notably challenges in obtaining the data needed to support a structured data-driven approach. The lessons learned from these case studies informed the development of guidance for state DOTs seeking to implement a cross-asset resource allocation approach. This guidance updates and expands the guidance developed in NCHRP Report 806. The guidance incorporates practical examples from the case study agencies, as well as other agencies that have implemented allocation approaches of their own. The lessons learned from the case studies also informed the improvement of the spreadsheet tool initially developed in NCHRP Report 806. While the basic structure of the tool remains the same, various updates were made to make the tool more accessible, user friendly, and robust. The tool also now allows users to input their own custom weights on the objectives, rather than going through the pairwise comparison exercise to set the weights. An additional web-based tool also was developed through this project. This tool utilizes DEA to calculate the relative efficiency of projects, rank and select projects based on their relative C H A P T E R 6 Conclusions and Suggested Research

100 Case Studies in Implementing Cross-Asset, Multi-Objective Resource Allocation efficiency and a specified budget, and provides four different visualizations that users can use to communicate the results of the prioritization and selection outputs. The results of this research are intended to be immediately applicable for individuals at state DOTs and other transportation agencies. However, several areas have been identified that merit additional research in the future. These areas include: • Improved methods for monetizing benefits. Utilizing performance measures that are analogous to monetized benefits avoids the pitfalls associated with scaling the scores and assigning weights to objectives. In the Caltrans and MDOT cases studies, methods for mone- tizing the benefits of transportation projects were borrowed from benefit–cost techniques developed previously. However, existing data and approaches are scarce for monetizing certain types of benefits, in particular health and sustainability benefits, from asset invest- ments. Further research is needed to help quantify the benefits of asset investments to help support application of cross-asset prioritization and benefit–cost analysis to assessment of asset investments. • Incorporation of risk. Many projects prioritized in a cross-asset resource allocation approach help to mitigate risk. While this reduction in risk has great value to agencies, it is difficult to quantify and incorporate this value in a resource allocation approach. Also, while many approaches assume that an agency is attempting to value a given set of investments, in incor- porating risk it may be more appropriate to prioritize differently, such as through minimizing the maximum risk exposure or regret from failure to avoid risks. • Improved asset models. The case studies describe an approach to prioritizing cross-asset investment that relies on defining candidate projects through other systems and approaches, such as current pavement and bridge management system. However, existing COTS systems lack the ability to model cross-asset needs comprehensively. Currently, systems are not capable of incorporating the details of asset-level modeling performed in existing management systems and development of projects performed external to these systems. A new generation of asset management systems is needed with a comprehensive modeling approach, incorpo- rating the cross-asset framework described here. • Improved prioritization approaches. The research described in this report explores the use of DEA to optimize a portfolio of projects given a set of candidate projects. In contrast to AHP described in NCHRP Report 806, DEA offers an approach that yields results without an explicit step to set weights on different objectives. However, this approach is more difficult to explain to stakeholders and potentially less transparent. In future research, the use of other prioritization approaches should be explored. This would increase the options available to agencies seeking to implement resource allocation and make better data-driven decisions. • Quantification of level of effort. Establishing and implementing a cross-asset prioritization approach can be resource intensive. It requires staff time to develop the evaluation criteria and resources to collect and process the data required for scoring projects. Several of the efforts described in this report were underway at the time of the research. Follow-up is needed on the research to document the resources expended for the cases described in the research, and/or required for other similar efforts. • Evaluation of implementation outcomes. Now that cross-asset resource allocation approaches are increasingly being implemented at state DOTs and other transportation agencies, future research can include evaluating the implementation outcomes. This effort could investi- gate whether or not there are quantifiable benefits and improvements in an agency’s perfor- mance with a resource allocation approach in place. Research can also look at any changes in the public’s perception and understanding of an agency’s decision-making process as cross- asset resource allocation is intended to increase openness and transparency.