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.

4 This chapter provides an overview of cross-asset resource allocation concepts. It describes the characteristics of a cross-asset resource allocation problem, provides an overview of the basic approach for making cross-asset investment decisions, and summarizes the NCHRP Report 806 framework. Characteristics of the Cross-Asset Resource Allocation Problem A decision maker must balance a number of competing objectives in determining what invest- ments to make in a set of transportation assets, including, but not limited to, improving infra- structure condition, making the transportation system more safe, enhancing mobility, improving the environment and communities, and supporting economic development. Ultimately, the funds for achieving these objectives are tightly constrained, and a decision maker must work with various stakeholders to determine what investments will best achieve the various competing objectives. Figure 2-1 illustrates the competing factors the decision maker must consider in determining how to prioritize transportation investments across multiple asset types, includ- ing the level of investment that can be supported, what candidate projects can be performed, and the resulting performance of a given set of investments. Transportation agencies use a number of different formal and informal systems to support the decision-making process. In particular, pavement and bridge management systems store data on the existing inventory of pavements and bridges, respectively, predict future conditions, and recommend potential asset investments to improve conditions and maximize agency and user benefits. Additional systems are used to analyze investments in improving safety, establish resources to achieve a specified level of service for traffic and safety assets, evaluate potential mobility improvements, and analyze other types of investments. However, no one system performs an integrated analysis of all of a transportation agency’s existing assets, let alone all of its different types of investments. Further, even if an agency had such a comprehensive system, agencies typically lack the level of detail in their asset inventories required to support the sort of analysis performed by a pavement or bridge management system. Actual transportation investments often address multiple needs and multiple asset classes in the same physical location. For instance, a project to improve an existing highway corridor might involve adding another lane; repaving the existing highway; improving the condition of existing bridges and culverts along the corridor; replacing traffic and safety assets such as signs, signals, and guardrails; adding a bike lane providing environmental and community benefits; and making improvements to existing intersections to enhance safety and mobility. Thus, when developing projects and deciding which projects to fund, transportation agencies C H A P T E R 2 Cross-Asset Resource Allocation Overview

Cross-Asset Resource Allocation Overview 5 must typically perform supplemental analyses beyond those supported by any one existing management system to develop realistic projects and determine how to prioritize these projects considering competing objectives. Approaches for Supporting Cross-Asset Resource Allocation Although there are many issues inherent in making cross-asset investment decisions, trans- portation agencies do in fact make these decisions day in and day out using a mix of approaches. Different approaches are used for allocation of different types of investments and/or within different geographical areas. For instance, an agency might follow a highly structured approach involving many stakeholders for prioritizing major mobility investments, while relying on a more informal approach implemented at a district level for determining how to prioritize asset preservation or maintenance investments. One basic approach to prioritizing asset investments is to “divide and conquer” the cross- asset resource allocation problem into multiple sub-problems. For instance, an agency can use pavement, bridge, and other management systems to analyze needs for a selected asset class or investment type and then prioritize work within each subset (e.g., separate subsets for pavement, bridges, safety, and maintenance). This simplifies the problem and allows for use of existing management systems, which tend to have asset-specific models. However, for each subset of investments an agency defines, it is necessary to establish an overall budget, and as more subsets are defined, there may be more cases in which a particular investment addresses assets or investment types in multiple subsets (e.g., a pavement project that includes safety improvements). Figure 2-1. Factors in cross-asset resource allocation decisions.

6 Case Studies in Implementing Cross-Asset, Multi-Objective Resource Allocation A second basic approach is to define a set of candidate projects, each of which may address multiple asset classes and/or investment objectives and then prioritize the individual candidates. An advantage of this approach is that it allows for making final decisions about what work to perform by combining multiple asset classes and investment types. However, the prioritization approach affects only the final decision of how to prioritize a set of defined investments. Thus, with this approach there is still a need for management systems and other structured approaches to identify candidate work, which can then be prioritized. When candidate projects are prioritized across asset and investment categories, the process followed may be either formal or informal. Often the prioritization process is performed infor- mally, with a small group of decision makers deciding what projects to implement based on expert judgment and consideration of factors such as what projects are “shovel ready.” Increasingly many transportation agencies are implementing more formalized, structured approaches to prioritizing cross-asset investments, and this area is the focus of the remainder of this report. As noted in Chapter 1, using a structured approach has many potential benefits, including better outcomes (decisions that better support an agency’s objectives) and increased agency accountability through following a documented, transparent process. A significant body of research exists in the area of how to formally structure making decisions through the consideration of multiple objectives. This general area is often described as multi- objective decision analysis (MODA) or multi-criteria decision making (MCDM). A starting point for this discussion can begin with Keeney and Raiffa’s Decisions with Multiple Objectives (1). This text describes how a decision maker can structure the set of objectives he or she is trying to achieve through his or her decisions, and how to measure progress toward accomplishing these objectives through defining a utility function. Once the utility function is defined there are various techniques for prioritizing investments and/or optimizing which investments to make given a budget and other constraints, and considering uncertainty. For instance, one may wish to simply observe which decisions yield the greatest utility, prioritizing based on the ratio of utility to costs, or use an alternative approach to find the right decisions to minimize regret, reduce risk, and/or maximize progress toward multiple objectives. Operations research texts such as Winston (2) describe the various approaches for solving a multi-objective resource allocation problem once it has been formulated. NCHRP Report 806 summarizes the literature in this area in the context of transportation investment decision making (3). Since publication of NCHRP Report 806, Danesh, et al. performed a comparison of different MCDM methods for use in project portfolio management (4). Their review considers over 100 different MCDM methods. Of these, the authors performed a detailed analysis of the eight methods deemed to be most promising. They concluded that while no method is ideally suited for project portfolio management, the two that are most suitable are the Analytical Hierarchy Process (AHP) (5) and DEA (6). NCHRP Report 806 describes AHP and the tool accompanying the report utilizes this approach. The Appendix provides additional information on DEA, and the web tool developed through the current research demonstrates application of this method. Cross-Asset Resource Allocation Framework Figure 2-2, reproduced from NCHRP Report 806, presents an overall framework for cross- asset investment decisions. The figure shows the basic steps in applying a structured approach to cross-asset investment decisions and poses the major questions a decision maker must answer in implementing a cross-asset prioritization approach.

Cross-Asset Resource Allocation Overview 7 The framework follows the basic approach to decision making with multiple objectives outlined by Keeney and Raiffa (1), but it is specific to transportation investment decisions and introduces the use of performance measures for characterizing investment impacts. NCHRP Report 806 (3) details each of the steps in this framework, and includes a prototype tool structured around the above steps. The remainder of this report focuses on a set of case studies in which the above framework has been implemented either implicitly or explicitly, resulting in the definition of structured process for prioritizing cross-asset, multi-objective decisions. The case studies are followed by updated guidance incorporating lessons learned from the case studies, as well as documentation of an updated set of prototype tools transportation agencies can use in their efforts to prioritize cross-asset investments. References 1. Keeney, R.L. and Raiffa, H. Decisions with Multiple Objectives. Cambridge, UK, Cambridge University Press, 1993. 2. Winston, W. Operations Research: Applications and Algorithms (Fourth Edition). Belmont, California, Duxbury Press, 2003. 3. Maggiore, M., Ford, K., High Street Consulting Group, and Burns & McDonnell. NCHRP Report 806: Guide to Cross-Asset Resource Allocation and the Impact on Transportation System Performance. Trans- portation Research Board of the National Academies, Washington, D.C., 2015. 4. Danesh, D., Ryan, M.J. and Abbasi, A. “A Systematic Comparison of Multi-Criteria Decision Making Methods for the Improvement of Project Portfolio Management in Complex Organizations.” International Journal of Management and Decision Making, Vol. 16, No. 3, pp. 280–320, 2017. 5. Saaty, T. The Analytical Hierarchy Process. New York City, New York, McGraw Hill, 1980. 6. Charnes, A., Cooper, W.W. and Rhodes, E. “Measuring the Efficiency of Decision Making Units.” European Journal of Operational Research, Vol. 2, No. 6, pp. 429–444, 1978. Figure 2-2. Cross-asset resource allocation framework (3).