Life-Cycle Cost Analysis for Management of Highway Assets (2016)

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.

12 CHAPTER THREE AGENCY PERSPECTIVES ON LIFE-CYCLE COST ANALYSIS AGENCY SURVEY LCCA data requirements and available models were dis- cussed in chapter two of this report. To learn more about LCCA use within state highway agencies, a survey was performed in spring 2015 using an online survey tool. The survey was developed to better understand the challenges of applying LCCA with a series of questions related to the software, data, and model needs of state agencies to support LCCA. The survey was designed to achieve an 80% rate (or 40 states) and to collect basic information on LCCA use, as well as to provide a mechanism to identify potential state highway agencies to showcase in the case examples pre- sented in the follow on chapters of this report. The survey also offered insight as to which states are utilizing LCCA in their decision making and management of highway assets and to which assets they have applied the analysis technique. Additional questions were developed to learn at what level state highway agencies are applying LCCA. For example, the questionnaire asked agencies to specify their LCCA use at the asset level, network/program level, and project level as defined in the survey materials: • Asset level—individual items. For example, individual bridges, individual culverts, and 1/10th mi pavement sections as defined by the state transportation agency. • Project level—A proposed project with logical begin- ning and end termini, often related to a milepost or intersection that consists of multiple assets. • Network/program level—A holistic view of the state- wide asset class that addresses current conditions, performance goals, condition prediction, and avail- able treatments within a defined budget. Example asset classes include pavements, bridges, signs, signals, and culverts. Survey questions were developed and refined with input from the panel in fall 2014. The survey focused on three pri- mary areas: • LCCA use by asset and application area • Identification of software and tools used • Data and model needs. The first set of questions focused on identifying if sur- vey respondents are using LCCA for decision making for comparing design alternatives for capital investments and/ or for maintenance treatment selection. Respondents were asked to provide specific information by asset as well as by application level. The information gathered through this set of questions was intended to identify highway agency LCCA use as well as identify unique LCCA applications that may be documented in a follow-on case study. Next, respondents were asked to identify by asset type and application level the specific tools and software used to conduct LCCA. Space was provided within the survey to allow respondents to add specific information about the tools and software packages used. Again, this information helped to identify interesting aspects of LCCA that may be documented in a case study. Finally, detailed information was requested about factors and data used in respondents’ LCCA applications by asset type and application level. Questions were asked about the confidence that agencies had in specific data and areas that they believed needed additional information to confidently apply specific LCCA factors and data. A draft survey was sent to panel members in November 2014. Comments were received and incorporated into the final survey in December 2014 and programmed into the online sur- vey tool SurveyGizmo. The survey was then shared with the project panel for a second review in January 2015. By February, the survey was finalized and sent by e-mail to members of the AASHTO Subcommittee on Asset Management as provided by TRB staff. The survey instrument sent to subcommittee members is included in Appendix A of this report. SURVEY PARTICIPATION The survey was first sent to subcommittee members on Feb- ruary 4, 2015, and within the first week, four surveys had been completed. Following the initial distribution of the survey, reminder e-mails prompting the recipients to com- plete the survey were sent weekly throughout the month of February. The reminder e-mail list was updated each week as completed surveys were submitted to remove from the list agencies that completed the survey. In March, efforts were made to identify additional respondents in states where the project team had been unable to contact subcom-

13 mittee members directly through phone or e-mail. In early April, AASHTO’s program director for planning and policy was contacted to ask for assistance in encouraging state highway agency representatives to complete the survey. With AASHTO’s assistance, participation increased and an additional 14 surveys were completed. In addition, phone calls and e-mails were sent throughout April to encourage survey completion. In total, as of May 20, 38 state high- way agencies had responded to the survey. With assistance from the project panel, three additional states completed the survey in June and as of July 2015 three more surveys were added to achieve an 82% response rate. Reflected in the tables throughout this chapter, one state agency survey response is equal to about 2.5% of all responses. Figure 3 provides a graphical representation of the surveys com- pleted and analyzed. FIGURE 3 LCCA use survey responses. SURVEY RESULTS This section of the chapter describes the survey findings and identifies any trends that were found in the data. The initial set of survey questions focused on identifying agencies that are applying LCCA as part of their decision-making process, identifying to which asset classes LCCA is being applied, and identifying at what level the process is being applied (asset, project, and network/program level). As shown in Figure 4, 30 of the 41 survey participants indicated that their agencies are using LCCA at the asset level to assist with the decision-making process for analyzing design alternatives. Of those respondents who are using LCCA at the asset level, the majority of applications are to pavements and bridges. Some highway agencies indicated applying LCCA to assets other than pavements and bridges. For example, Virginia DOT currently uses LCCA at the asset level for culverts, tunnels, ITS, and traffic signals and signs. The Vermont Agency of Transportation uses LCCA at the asset level for signs, rock fall hazards, and maintenance equipment (dump trucks). Both the North Dakota and Wyoming DOTs indi- cated applying LCCA to decision making for traffic signs. Of those respondents who indicated “Other,” LCCA was reported to be used in fleet, rock fall hazards, high mast lighting, and equipment decision making. FIGURE 4 LCCA use (asset level). The survey results also revealed that 27 of the 41 agencies that completed the survey currently use LCCA for selecting preservation or maintenance treatments for assets. Figure 5 includes a breakdown of which states are using LCCA for maintenance treatment activities by asset type. Eleven states reported using LCCA for selecting preservation or mainte- nance treatments for pavements and bridge management. Twenty-five of the 41 states that responded to the survey reported using LCCA for selecting maintenance treatment for pavements only and 12 states are applying LCCA to select maintenance treatments for bridges. Nearly half of the survey respondents who use LCCA for selecting preserva- tion maintenance treatments use this process for bridges. Of the 41 states that responded to the survey, 14 reported not using LCCA for maintenance treatment. Virginia DOT was the only state to report using LCCA for selecting preserva- tion or maintenance treatments for both culverts and tunnels as well as bridge and tunnel treatments. FIGURE 5 LCCA use (maintenance level). Next, survey participants were asked questions to determine at what level they are applying LCCA. Of the

14 survey responses, 26 of 41 state agencies surveyed are cur- rently using LCCA at the project level. For example, the agency uses LCCA to select between project alternatives that result in the same benefits with the lowest expected life-cycle costs. Figure 6 details which states are cur- rently using LCCA at the project level. Twenty of 41 sur- vey respondents reported using LCCA at the network or program level to identify treatment efficiencies. Figure 7 details which states are currently using LCCA at the net- work or program level. FIGURE 6 LCCA use (project level). FIGURE 7 LCCA use (network level). Next, survey participants were asked if they are using specialized software or tools to perform LCCA. Of the 41 survey respondents, 16 reported using specialized software for LCCA. Survey participants reported utilizing special- ized LCCA software tools most often for pavements, with 11 states applying these tools at the network/program level, nine states at the project level, and nine states at the asset level. Nine states reported currently using specialized soft- ware for bridge LCCA, with seven of those states applying these tools at the network/program level. Currently, none of the state agencies surveyed reported using specialized soft- ware at any level for the LCCA of tunnels, traffic signals, end treatments, or striping. The survey participants reported using several differ- ent types of specialized software; a brief overview of the software and tools are described here. Deighton’s Total Infrastructure Management System (dTIMS) allows users to incorporate all of their agency’s infrastructure assets and corresponding data into one platform. Five states reported using dTIMS for pavement and bridge LCCA across asset, project, and network/program levels. Minnesota Department of Transportation’s own HydInfra, hydraulic infrastruc- ture information application, is used to manage inventory, inspection, and maintenance activities for storm drainage features (culverts) at the asset level. This tool is highlighted in a case example in the following chapters of this report. The FHWA’s RealCost uses Microsoft Excel spreadsheets to perform LCCA for pavement selection in accordance with FHWA best practice methods. RealCost was reported to be used by Washington State DOT, California DOT, and South Carolina DOT for the LCCA of pavements and bridges. Florida DOT uses in-house software PLAT (Project-Level Analysis Tool), which draws information from the Bridge Management Database, and generates cost–benefit ratios for various design and treatment options. Colorado DOT uses specialized software at the asset and network/program levels for ITS for LCCA. Texas DOT developed its own spread- sheets for LCCA of traffic barriers, bridges, pavements, and culverts. Finally, Wyoming DOT reported using an in-house software system for sign LCCA at the asset and network/ program levels. Next, agencies were asked if they have specific analysis periods for various asset classes. An analysis period refers to the time over which costs are evaluated. Of the 41 sur- vey respondents, 23 reported that their analysis periods do vary by asset class. Once again, most of the states reported having specific analysis periods when applying LCCA to bridges and pavements. However, specific analysis periods were also reported to be used for ITS and culvert LCCA. For example, Colorado DOT reported that pavement, bridges, and ITS analyzed at the asset level have an analysis period of 10 years, although that number varies by asset type at the network/program level. Minnesota DOT analysis periods were longer, with pavement analysis periods at 70 years, bridge analysis at 200 years, and other assets at 100 years all at the network/program level. Currently, none of the states that responded to the survey reported specific analysis peri- ods for tunnels, traffic signals and signs, end treatments, lighting, or striping. Figure 8 includes an overview of the information gathered from the survey for analysis periods by asset class and application level. The next portion of the survey focused on specific data and factors used by agencies in their LCCA. Survey respondents were asked to provide information by application level and asset class on asset-specific discount rates used in their LCCA. Of the respondents, 19 of the 41 noted that they do vary their discount rate by application level or asset class. Most notably,

15 LCCA performed for pavements at the project level was most frequently reported to have a specific discount rate. Of the 41 respondents, 11 have asset-specific discounting rates for pave- ment at the project level, seven at the network/program level, and four at the asset level. Currently, five states have discount rates for LCCA for bridges at the project and network/pro- gram levels. Most of the survey respondents reported using a discount rate between 2% and 5% for all assets at all levels. Two states reported using the federally published Office of Management and Budget circular to calculate their agency’s discount rates. At this time, no specific discount rates were reported for applying LCCA to decision making for tunnels, signs, traffic barriers, end treatments, striping, or lighting. Figure 9 shows a breakdown of this information. FIGURE 8 Asset-specific analysis periods. FIGURE 9 Asset-specific discount rates. Next, survey participants were asked to provide informa- tion on factors used in their LCCA by application level. The goal of this set of questions was to learn more about which factors agencies were using and which they were not, to shed light on areas that agencies could benefit from additional information and guidance. State agencies were most likely to use capital and maintenance cost factors and data in LCCA at the project level. Resilience goals are the least likely factor to be taken into consideration by state agencies in their LCCA analysis. In addition, some costs still appear to be a chal- lenge for state agencies, but this information may be lacking in availability; for example, inspection and operations costs. Deterioration curves were also noted to be a challenge at the asset level as compared with their use at the network/program level. Areas that appear to fall behind in their inclusion in LCCA include safety and resilience goals as well as uncer- tainty or risk in anticipated costs. Table 1 contains the find- ings of this question related to which factors and data agencies are incorporating LCCA by application level. TABLE 1 DATA/FACTORS INCLUDED IN LCCA BY STATE HIGHWAY AGENCIES Data/Factors Included in LCCA Asset Level Project Level Network/ Program Level Capital Costs 14 (34%) 25 (61%) 21 (51%) Maintenance Costs 13 (32%) 20 (49%) 15 (37%) Operations Costs 5 (12%) 9 (22%) 4 (10%) Inspection/Support Costs 6 (15%) 4 (10%) 5 (12%) User Costs 5 (12%) 13 (32%) 4 (10%) Discount Rates 7 (17%) 19 (46%) 13 (32%) Deterioration Curves/Models 11 (27%) 13 (32%) 19 (46%) Uncertainty/Risk 3 (7%) 6 (15%) 5 (13%) Resilience Goals 1 (2%) 0 (0%) 0 (0%) Current Safety Performance 6 (15%) 6 (15%) 4 (10%) Expected Safety Performance 3 (7%) 4 (10%) 2 (5%) Desired Performance Levels 10 (24%) 11 (27%) 17 (42%) Geospatial Location of Assets 6 (15%) 8 (20%) 9 (22%) Salvage Value 4 (10%) 10 (24%) 4 (10%) Remaining Service Value 6 (15%) 11 (27%) 8 (20%) The second part of this question focused on asking partici- pants to specifically name factors and data that they felt their agency lacked solid information needed to perform LCCA at the three application levels. According to the survey results, data to perform LCCA for capital, maintenance, and inspec- tion/support costs is most readily available; however, data and information is needed to better incorporate deteriora- tion curves/models, uncertainty/risk, and resilience goals in LCCA, as was also noted in the previous question. Many of the respondents surveyed reported a lack of data for sal- vage value and remaining service value; however, very few states reported a lack of data for desired performance levels or geospatial location of assets. Table 2 includes the findings from this survey question. Of note is the concentration of responses focused on the need for data related to deterioration curves, uncertainty/risk, resilience goals, expected safety performance, and remain- ing service life. Most notably, nearly half of states feel a need for more data about uncertainty/risk at the project level for LCCA. Another common need among state DOTs is the lack of data and information needed to establish resilience goals and how to incorporate these into LCCA. This finding sup- ports the lack of information needed by states to connect LCCA to overall risk management initiatives, which support the delivery of resilience highway systems. These findings are echoed in the case examples contained in chapter five.

16 TABLE 2 DATA/FACTORS NEEDED FOR LCCA BY STATE HIGHWAY AGENCIES Data/Factors Needed for LCCA Asset Level Project Level Network/ Program Level Capital Costs 1 (2%) 1 (2%) 4 (10%) Maintenance Costs 5 (12%) 6 (15%) 9 (22%) Operations Costs 7 (17%) 9 (22%) 11 (27%) Inspection/Support Costs 3 (7%) 7 (17%) 7 (17%) User Costs 10 (24%) 11 (27%) 12 (29%) Discount Rates 4 (10%) 4 (10%) 6 (15%) Deterioration Curves/Models 12 (29%) 12 (29%) 7 (17%) Uncertainty/Risk 12 (29%) 16 (39%) 13 (32%) Resilience Goals 10 (24%) 14 (34%) 14 (34%) Current Safety Performance 7 (17%) 8 (20%) 5 (12%) Expected Safety Performance 9 (22%) 11 (27%) 7 (17%) Desired Performance Levels 4 (10%) 6 (15%) 3 (7%) Geospatial Location of Assets 4 (10%) 3 (7%) 3 (7%) Salvage Value 8 (20%) 8 (20%) 10 (24%) Remaining Service Value 11 (27%) 10 (24%) 12 (29%) SUMMARY AND FINDINGS The purpose of the survey was to better understand the chal- lenges of applying LCCA with a series of questions related to LCCA application, software and tools used, and data and models needed to support LCCA. The survey results pro- vided basic knowledge on how states are currently using LCCA in their decision making and management of highway assets. The survey results also specified which state agencies are applying LCCA at the asset level, project level, and net- work/program level. The findings also identify which state agencies are using LCCA for decision making for analyzing design alternatives for capital investments and for mainte- nance treatment selection. The results detailed the tools and software respondents are currently using to conduct LCCA for various assets. Finally, the survey respondents also speci- fied the factors and data used in LCCA applications by asset type and application level, and which factors and data they feel they lack information to fully utilize. According to the survey results, most state highway agen- cies are currently using LCCA for pavement and bridge man- agement at all application levels. Most often LCCA is being used as part of the decision-making process for analyzing asset-level design alternatives. Currently, 16 state agencies are using specialized software to assist with these LCCA appli- cations. Most notably, the survey results showed that capital costs, maintenance costs, inspection/support costs, and user costs are the most common factors considered in LCCA anal- ysis by state highway agencies. On the other hand, very little consideration is being given to incorporating resilience goals and uncertainty/risk factors into LCCA applications. Since the purpose of this survey was to identify chal- lenges in applying LCCA, it is important to focus on what factors and data state agencies reported to be lacking to prop- erly perform LCCA or improve existing LCCA. As reflected earlier, deterioration curves/models, uncertainty/risk, and resilience goals lack available data to perform LCCA appli- cations. Knowledge gaps also exist regarding salvage value and remaining service. According to the state agencies sur- veyed, there is a significant lack of data available to properly perform LCCA applications for assets beyond pavements and bridges. Also, the lack of understanding of the bene- fits of applying LCCA to assets other than pavements and bridges appears to be significant. In the next chapter, several state highway agencies’ application of LCCA are highlighted to bring attention to unique and innovative approaches to LCCA. In particular, two states’ approaches to pavement and bridge LCCA and two states’ approaches to data gathering to support LCCA are presented.