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.

17 CHAPTER FOUR CASE EXAMPLES ON THE USE OF LIFE-CYCLE COST ANALYSIS UDOT staff acknowledged that the lack of funding for these facilities will result in degradation of performance measures for these facilities. Utilizing information generated by the agency’s annual pavement condition data-gathering exer- cise and projections of performance provided by its pave- ment management program, the agency was able to provide information to the state’s legislature to support the decision to increase the gas tax in January 2016. The five cent sales tax increase per gallon of gasoline is expected to generate an additional $17.14 million in revenue in FY 2016 and $55 million in FY 2017. The agency intends to target Level 2 facilities during these fiscal years to slow and reverse the anticipated degradation of these facilities. UDOT representatives noted that their LCCA for pave- ment includes maintenance, preservation, repair, rehabili- tation, and replacement actions that achieve and sustain a desired state of good repair over the life cycle of the asset at a minimum practicable cost. Having generated substantial performance and condition information for 1,446 identified sections of pavement across the state, UDOT pavement man- agement staff reported being very confident with the deterio- ration curves produced by its pavement management system. These deterioration curves feed the LCCA process for its statewide pavements program. Information was gathered as to specific data used within the agency’s LCCA. The agency included in its analyses capital costs, maintenance costs, inspection/support costs (at the asset level), deterioration curves and models, cur- rent safety performance, desired performance levels, and geospatial location of assets. The agency noted it lacked sufficient information and models for the establishment of discount rates, deterioration curves/models (for bridges), and remaining service value (for bridges). UDOT has begun to investigate remaining service life for its pavement program. Substantial efforts have been made over the past several decades to establish detailed information on the performance of specific sections of pave- ment throughout the state. Often referred to as “A Plan for Every Section of Road,” UDOT has identified and tracked the condition of 1,446 sections of roadway allowing for opti- mization of surface treatments. The identified sections of pavement typically are between 5 and 10 mi in length and in many cases were defined at the time of construction. In This chapter documents five case examples related to the collection of necessary data and models to fully implement LCCA by state highway agencies. The case examples were developed based on information gathered through the sur- veys, as well as through in-depth interviews with staff and for the following agencies and application areas: • Utah Department of Transportation—Pavement • Florida Department of Transportation—Bridges • Washington State Department of Transportation— Ancillary Asset Management • Minnesota Department of Transportation—Culvert Cost and Life-Cycle Management • Initiative • Public–Private Partnerships—A Concessionaire’s Take on LCCA. LIFE-CYCLE COST ANALYSIS FOR PAVEMENTS— UTAH DEPARTMENT OF TRANSPORTATION Utah Department of Transportation (UDOT) maintains approximately 16,000 lane miles and has established a vision statement for the pavement management program: Good Roads Cost Less. In 1977, a study titled Good Roads Cost Less was conducted on behalf of UDOT (26). This study sought to demonstrate that with timely, cost-effective treatments, the cost of roadways can be minimized while maintaining a desired level of performance (27). This philosophy of good roads cost less is still present within UDOT and has shaped its vision of asset manage- ment. Since 1993, UDOT has used dTIMS CT, a manage- ment system developed by Deighton Associates Limited, to manage its pavement investments (26). The agency estimates that $250 million dollars annu- ally is needed to preserve the estimated $25 billion dollar pavement asset it maintains; however, for the past six years, only $200 million has been available to maintain pavements. To address the gap in funding, UDOT developed a tiered approach to pavement management that seeks to optimize investments. The tiers include Interstates; Level 1 facilities, which service more than 1,000 AADT (average annual daily traffic) and truck volume greater than 200; and Level 2 facil- ities, which service less than 1,000 AADT. Level 2 facili- ties are actively maintained by regional maintenance staff;

18 2014, UDOT conducted a pavement distress and detection survey using LiDAR (light detecting and ranging) to further enhance its understanding of pavement conditions across the state. The agency reported that it has contracted to collect two additional data sets over the next 5 years using the same type of technology, to further develop deterioration curves and remaining service life. Additional efforts by the agency include incorporating its maintenance management system to the dTIMS model and developing a data warehouse to create a mechanism to incor- porate and exchange of information between the two systems. LIFE-CYCLE COST ANALYSIS FOR BRIDGES— FLORIDA DEPARTMENT OF TRANSPORTATION This case example focuses on Florida Depart- ment of Transportation’s (FDOT’s) Bridge Asset Management practices and LCCA use. FDOT maintains more than 12,000 lane miles and 6,500 bridges across eight semiautonomous districts. Since 1997, FDOT has implemented AASHTO’s Pontis Bridge Manage- ment System to support decision making. The agency has con- ducted a series of research studies to fine-tune Pontis to better suit the agency’s needs including the following: 1. Development of a localized user cost model 2. Establishment of unit costs for maintenance, repair, and rehabilitation actions 3. Development of deterioration curves based on expert elicitation and field inspection reports 4. Development of truck and weight histograms to esti- mate user costs from detours and closures 5. Development of a project-level decision support tool. Recent research efforts entailed the development of a project-level decision support tool to interpret Pontis results in a form more applicable to bridge-level deci- sion making (28). In addition, a network-level tool was developed to provide estimates of expected performance and required funding at a systemwide level. In 2011, fur- ther research was conducted on the two custom tools to improve recommendations (17). As part of this recent research, recommendations were made to update deterio- ration models that were found to be overestimating bridge deterioration and underestimating repair costs. Utilizing two years of bridge inspection data, researchers were able to develop an improved standalone computer program to create a single condition rating for each bridge component, similar to FHWA’s National Bridge Inventory Translator. Using case studies and field inspection reports, research- ers were able to calibrate the single condition ratings for Florida conditions. Additional research has been conducted on improv- ing deterioration curves and action effectiveness models, to update previous models that were primarily based on expert-elicitation data. Data drawn from Florida’s mainte- nance management system and AASHTO’s Trns•Port Esti- mator database to support LCCA use for bridges were used to refine previously developed models that were found to underestimate the effectiveness of repair and rehabilitation actions. Researchers believe the improvements to these mod- els and curves will greatly improve the condition predictions in Pontis and PLAT/NAT (Project-Level Analysis Tool and Network Analysis Tool), which they believe will improve funding decisions. Research was also conducted to validate existing cost mod- els within Florida’s Pontis and PLAT/NAT utilizing informa- tion from a statewide construction bids database (AASHTO’s Trns•Port Estimator), FDOT District Bridge Construction Bids Records, and the FDOT Work Library-Maintenance Management System Cost data for bridge-related mainte- nance work. Further research was conducted to estimate user costs at bridge sites where no detour is considered or pos- sible. Researchers sought to include accident-related user costs within the overall user cost model that reflects travel time costs and vehicle operating costs. Utilizing Florida crash data at bridge locations between 2003 and 2006, researchers developed a negative binomial model that was found to be more accurate than the previously utilized linear regression model when compared with observed crashes in 2007. The department noted that having rich element-level bridge inspection data allowed for in-depth analysis of bridge deterioration and the ability to better forecast life- cycle costs for the planning of maintenance, repair, reha- bilitation, and replacement work. The investment made to gather the necessary data facilitated improvements in the applicability of recommendations and estimates of bridge conditions from Florida’s Pontis Bridge Management Sys- tem and PLAT/NAT. Confidence gained in the decision support tools is believed to have had a significant impact on capital and maintenance programs, potentially better utiliz- ing constrained maintenance and replacement dollars. Utilizing the improved decision support tools, FDOT has implemented LCCA for major bridge projects in the plan- ning stage; however, LCCA is not yet used in routine bridge projects. LCCA is used to decide if it is more cost-effective to repair an existing bridge or to replace it by calculating both anticipated life-cycle costs as well as cost–benefit ratios for various design options. One of the lessons learned by FDOT through the develop- ment and refinement of its bridge management tools is the need

19 to standardize definitions and measurements used by various offices within the agency. For example, bridge management reporting systems were found to differ from maintenance and construction reporting systems, which made it challenging to compile information to support the calibration of the bridge management analysis tools. But, in the end, the products cre- ated appear to be generating results and recommendations that are much more reasonable and acceptable to FDOT staff. WASHINGTON STATE DEPARTMENT OF TRANSPORTATION ANCILLARY ASSET MANAGEMENT One of the LCCA appli- cation challenges noted by survey participants and through the literature review is the lack of maintenance cost information. Washington State DOT’s (WSDOT’s) Maintenance Division has been crafting an evidence-based approach to maintenance priority-setting, bud- geting, and legislative requests for many years, in addition to maintaining an ongoing government quest for efficiency. Short of having a comprehensive cradle-to-grave LCCA system in place at WSDOT, the Maintenance Division is doing what it can within its purview. Maintenance is working partially in coordi- nation with other programs such as Design, Construction, and Preservation, to create and implement the building blocks of LCCA-based management. WSDOT is anticipating that these and other building blocks will someday come together into a comprehensive, LCCA-based highway asset management plan. WSDOT communicates to designers the cost and safety savings (because of quicker and easier maintenance) of design considerations such as the following: • Adequate width of shoulders (so maintenance does not have to close a lane of traffic) • Access to stormwater ponds and culverts (an area that continues to be overlooked to a surprising degree, according to staff) • Good-quality roadside soils with low-maintenance vegetation. WSDOT Maintenance has captured these design con- siderations in a laundry list of good practices that it has communicated with its design staff to help minimize the maintenance costs associated with asset management. Within the Maintenance Division, WSDOT is using the owner’s manual/maintenance schedule concept as a basis for the maintenance portion of LCCA. The easy-to-com- municate and easy-to-understand analogy of this concept is the car owner’s manual/maintenance schedule. The vast majority of people will not understand complex predictive computer models or mathematical equations, but they would agree that the owner’s manual is a summary of what the owner needs to know. Although many of the maintenance tasks that are completed today will not make an immediate observable difference, they will make a difference in asset performance over time. At these longer intervals, proactive maintenance will extend the cycles of rehabilitation and preservation over what can be a long and even indefinite life span of many of WSDOT’s highway assets. Ongoing Implementation As part of its ongoing push to implement sound asset man- agement and maintenance LCCA, starting July 1, 2015, WSDOT is equipping maintenance staff with 800 tab- lets. The Highway Activity Tracking System (HATS) is WSDOT’s in-house maintenance management system with which it manages its highway asset inventory and records the details of completed maintenance work. The asset-specific information that will be collected henceforth (versus hours of a certain maintenance activity and units accomplished) will provide the agency with a rich data set for LCCA. With the rollout of the tablets and the continued detail going into the creation of a record of maintenance, issues, conditions, and actions for each asset, WSDOT will have asset-specific maintenance histories available at the touch of a finger. Pavement maintenance information collected in HATS has already been incorporated into the agency’s Pavement Management System. Highway asset inventory, condition rat- ings, and completed maintenance work data for many other highway assets will be accumulated in HATS. When asset management systems, similar to those currently in place for pavements and bridges, are developed for the design, con- struction, rehabilitation, and preservation details of these assets, Maintenance will be ready to plug its portion of LCCA into the broader management system by means of HATS. In the past few years, WSDOT has set a new direction in evaluating maintenance and overall cost accrual by measur- ing preventive maintenance and work accomplishments. Its approach has yielded noteworthy results, with lower analyti- cal requirements than conventional LCCA approaches. By using an approach that is easier to communicate and under- stand, WSDOT found that the agency stood a much better chance of bringing partners on board with LCCA. WSDOT was helped along in its new direction by a state legislative audit that found that “WSDOT’s maintenance management system does not measure the backlog of essen- tial maintenance, limiting the ability to determine effective- ness of effort” (29). At the conclusion of their 2007 audit, the auditors made the following recommendations to WSDOT: • Determine needs from the agency’s respective main- tenance management systems (MMS) and the current backlogs of essential maintenance and repair; • Prepare a comprehensive listing of the backlogs of essential maintenance and repair, assess the risk that

20 the backlogs may pose, if any, and include those in M&O budget justifications; • Prioritize the development of a centralized MMS; • Annually calculate an estimate of the current replace- ment cost of the infrastructure; • Establish minimum maintenance and operations level of service (LOS) priorities and targets; • Include each measurement of maintenance performance in WSDOT’s performance measures program and Statewide Accountability Service Level Reports; and • Increase the detail of the Maintenance Accountability Program (MAP) organizational review-level achieve- ments to provide additional indication of accomplish- ments (e.g., not just condition assessments). MAP measures and communicates the outcomes of the maintenance activities, providing tools to link stra- tegic planning, the budget, and maintenance service delivery. Once a year, field inspections are made of randomly selected sections of highway. The results of WSDOT’s work are measured, recorded, and compared with the MAP criteria to determine the LOS delivered. The audit called for Maintenance to increase its ability to determine actual maintenance needed and impact on work- loads. The audit also called for the estimate of the extent of essential maintenance and repair backlog and cost thereof, as well as the budget impacts of compliance with new require- ments. Overall, WSDOT needed to increase its ability to iden- tify and communicate the cumulative effects of maintenance requirements. With the help of this type of documentation, WSDOT Maintenance won additional funding to help catch up on the maintenance backlog for eight activities, includ- ing signal maintenance. WSDOT also found considerable accountable benefits (reduced time spent on “call-outs” cov- ered the time needed for more preventive maintenance) and important unaccounted ones (the congestion, safety impacts, and customer unhappiness associated with signal outages that were avoided since outages were minimized). Although Maintenance is one of many contributors to asset condition, task completion is Maintenance’s responsibility. Task completion provides a sense of ownership to Maintenance staff; it also communicates well—people understand the tie between task completion and the “car owner’s manual” anal- ogy. In the December 2010 Gray Notebook, WSDOT explains: Task completion will increasingly be the primary tool used to measure maintenance performance. Asset condition (MAP surveys) will serve as a quality assurance tool used to verify or support changes in the maintenance task completion measure. Task completion quantifies the number of tasks needed for a specific activity each year, and how many of those tasks were completed. Completion of higher percentages of needed maintenance work contributes to good asset condition. Using these two performance measures together, overall program delivery can be more accurately explained. The owner’s manual approach and work accomplished ultimately comes down to work-unit planning. Most Main- tenance staff are used to being “more reactive, having more control over their daily lives, and acting indepen- dently.” More preventive maintenance will allow greater control and lower stress in some ways, but work processes are more set and prescribed. When WSDOT develops an owner’s manual/maintenance schedule for a type of high- way asset, it uses whatever resources are available for the contents. Manufacturer recommendations are available for a limited number of assets, such as some automated traffic management systems and cable barriers. High-tension cable barrier manufacturers recommend that systems be visually inspected and tension checked annually. Legal requirements sometimes constitute the LCCA maintenance schedule. For example, WSDOT’s National Pollutant Discharge Elimina- tion System (NPDES) permit lays out a detailed and strict schedule of inspection frequencies and maintenance stan- dards. It has made WSDOT’s job of identifying the owner’s manual/maintenance schedule very easy for this particular asset. This required maintenance and inspection schedule was easy to communicate to WSDOT partners and paved the way for full support and funding. As a result, WSDOT has the resources and is implementing a complete LCCA program (at least for maintenance LCCA) for catch basins and stormwater ponds. As described in the December 2010 Gray Notebook, because WSDOT could quantify the backlog for eight main- tenance activities, the legislature provided $16.8 million for the 2009–11 biennium to begin the process of catching up on the identified $85 million backlog of maintenance work. Task completion is to be captured as a percentage of identi- fied tasks and reflected as an LOS performance measure. With the establishment of funding, WSDOT was able to sta- bilize falling LOS scores and begin to meet or exceed its plan in all eight areas. In the next (2011–13) biennium the legislature provided an additional $6.4 million toward the maintenance backlog, not tied to specific activities, but this was a relatively small portion of what Maintenance calcu- lated was needed. The Washington DOT Maintenance Division has uti- lized an owner’s manual/maintenance schedule approach to begin to convey to its staff tasks that need to be accom- plished to proactively manage assets. In addition, the information gathered in its efforts to better understand the demands required for asset maintenance has been shared with its design teams to help reflect maintenance costs within overall life-cycle costs of highway assets. Ulti- mately, WSDOT believes that understanding maintenance costs in a manner that is easily conveyed to all staff and stakeholders will support the agency’s objectives of LCCA for highway assets.

21 MINNESOTA DEPARTMENT OF TRANSPORTATION CULVERT COST AND LIFE-CYCLE MANAGEMENT INITIATIVE Minnesota DOT’s ( M n D O T ’ s ) HydInfra system was first profiled in detail on the national level as part of AASHTO’s Compen- dium of Environmental Stewardship Practices, Policies, and Procedures in 2004. Since that time it has been documented in several publications, including the following: • FHWA’s Transportation Asset Management Case Studies: “Culvert Management Systems Alabama, Maryland, Minnesota, and Shelby County” in 2005; • MnDOT MAP-21 Transportation Asset Management (TAMP) federal study with Minnesota, Louisiana, and New York in 2014; and • Federal Lands Highways’ Chapter 2-Culvert Assessment Tool of the FHWA Federal Lands Highways Culvert Assessment Guide, which borrowed elements from HydInfra condition rating criteria in 2014. HydInfra stands for “Hydraulic Infrastructure” and is the culvert and storm drainage system inventory and inspection program MnDOT has developed for pipes with spans less than 10 ft. In 2014, MnDOT’s TAMP committee ranked culverts as the agency’s number one priority and area of risk. To sup- port risk analysis and life-cycle cost assessment for culverts, MnDOT recently completed an extensive culvert repair cost data collection effort. Three specific pieces of information were collected: • Condition Rating Codes—5-point qualitative scale plus Not Able to Rate/Unknown. • Inspection Flags—problems identified in the field including deformation, joint separation, and so forth (Table 3 contains an example of potential inspection flags by material type). • Pipe Material—to help track expected deterioration over time. As presented by MnDOT hydraulic staff at the 2014 national hydraulics meeting, the HydInfra system capa- bilities include a range of management and design products including life-cycle cost analysis: • Performance Measures • Prioritize Repairs • Cost Estimation • Maintenance Tasks • Project Predesign • Respond to Flood Damage • MS4 Water Quality Record Keeping • Utilities Locations • Research • Life-Cycle Cost. TABLE 3 MATERIAL/FLAG COMBINATIONS SEEN AT MNDOT Material Defect or “Flag” Defect or “Flag” Defect or “Flag” Concrete Deformation Cracks Spalling Concrete Joint Separation Road Void Concrete Inslope Cavity Joint Separation Concrete Joint Separation Infiltration Steel Holes Road Distress Steel Holes Deformation Steel Holes Piping Steel Holes Road Void HDPE Cracks HDPE Misalignment (floating) Linear HDPE Deformation Source: MnDOT. Note: HDPE = high-density polyethylene pipe. Currently, HydInfra users can run a web-based report that uses an automated sorting process for pipe inspection data. It then produces a suggested repair method, to give a first-pass idea for the repair of individual pipes in poor condition. Fig- ure 10 provides an overview of the type of information gener- ated through the Suggested Repair Report. In 2010, MnDOT staff used their Suggested Repair Report for culverts to create a statewide cost estimate of prospective repairs of MnDOT culverts. This also positioned MnDOT to request more fund- ing for what they had determined was a high-risk area. FIGURE 10 Suggested repair report. Note : CIPL = cured- in-place pipe liner. To improve the understanding of maintenance costs asso- ciated with culverts, MnDOT Maintenance collected data on individual culvert repairs and cleanings by recording all the labor, equipment, and materials in the agency’s new Cul-

22 vert Cost app (built on ArcGIS Collector software) in 2014. The estimated cost of each repair was then calculated from data collected in the field. This project was recommended by the Drainage Asset Management group to support the prediction of life-cycle asset costs. MnDOT Maintenance is continuing to collect culvert main- tenance costs for repairs, replacements, and cleanings. It is antic- ipated that 2015 data will include all culverts, not just highway culverts, and has been broadened to accommodate drainage features other than culverts (though the others are not required to be recorded). MnDOT anticipates that the data collection method will evolve when HydInfra and Culvert Cost move into an asset management software package, as soon as 2018. Culvert repair cost data have been developed for sev- eral repair methods—repairs that are routinely completed by Maintenance forces—to improve the accuracy of cul- vert repair and replacement cost estimates. Data from 2015 repairs will give the agency more information to support the Suggested Repair Report. In addition, MnDOT plans to apply these costs starting in 2016, to create statewide MnDOT culvert repair cost estimate. Given the anticipated influx of data and information for culvert repair cost data and the current lack of data, at this time MnDOT is using the average cost for each repair category to support decision making. Figure 11 contains an example of such an average cost estimate for a variety of maintenance repairs broken out by materials, equipment, and staff costs. Next year, once additional data are collected, the data can be reanalyzed to quantify the influence of parameters such as pipe material or size on maintenance repair costs. FIGURE 11 Average estimated cost of material, equipment, and labor by repair category (Source: MnDOT). Regular validation and refinement of costs is planned to ensure the consistency and quality of the data collected. MnDOT’s 2014 and 2015 data collection was significant and is improving understanding statewide on what repairs are being completed by Maintenance as well as where, how often, and at what cost. It is critical to have this type of infor- mation when moving toward asset management. MnDOT is planning to migrate HydInfra to the new departmental asset management system in the next 2 to 5 years. Ideally, the maintenance cost application and data will be integrated into the new HydInfra asset management system at that time. PUBLIC–PRIVATE PARTNERSHIPS—A CONCESSIONAIRE’S TAKE ON LIFE-CYCLE COST ANALYSIS The project panel that oversaw the development of this NCHRP synthesis report capitalized on its membership by requesting that one of its panel member’s share her experi- ence with LCCA from a concessionaire’s view. Through this case example, efforts were made to differentiate between the identified methods employed by state highway agencies to model and reflect total asset costs through LCCA and those utilized by organizations engaged in public–private partner- ships (P3s). Panel member Andrea Warfield arranged for an in-person interview with several members of her staff to provide some insight on the similarities and differences in LCCA between these two organizational structures. System Versus Asset-Class Approach to LCCA One of the most noted differences between the documented application of LCCA within state highway agencies and P3s is the view of assets by asset class versus viewing assets as a system. It was noted multiple times that a concessionaire is focused on reducing costs and, therefore, is driven by view- ing the system as a whole as compared with asset-driven deci- sion making. For example, concessionaires view the system from right-of-way line to right-of-way line and optimize per- formance across the entire spectrum of assets present. They also benefit from multiyear funding cycles, which allow them to align maintenance requirements for various asset classes to reduce lane closures and multiple deployments by maintenance crews. The “use it or lose it” philosophy of many highway agen- cies often leads to less-than-optimum decision making. P3s benefit from having a set period of performance that allows for easier coordination of required maintenance activities across multiple years to reduce overall costs. There is also an “owner” mentality to the system as a whole, and efforts are made in the planning, design, and construction stages of projects to include O&M staff, to ensure they can contribute to the process of reducing the overall life-cycle costs of the systems operated and maintained for multiple decades. Experience and knowledge gained from O&M staff who oversee the bulk of the systems’ lives help to recognize costly design and construction choices upstream of project delivery to drive down overall LCC. Holistic View of Costs Another advantage to including O&M staff in the planning, design, and construction stages of projects is the realistic and documented cost information they can bring to the table regarding often overlooked expenses including painting, irrigation, mowing, lane availability for maintenance, and

23 snow and ice removal. As a concessionaire, the driving force is to reduce project costs while meeting performance goals set forth by the public agency. Understanding the full cost of maintenance activities and being engaged in the project planning process allows the private investor to make better- informed decisions that will meet the project’s performance goals, while reducing overall maintenance costs over the project’s life. Also, there are no perceived barriers between maintenance and pavement management programs. For example, just-in-time maintenance to address potholes may help to reduce more major pavement repairs, again resulting in savings to the overall bottom line. As was noted, potholes never go away or get smaller. SUMMARY This chapter documented the experiences of four state highway agencies and a P3 concessionaire utilizing LCCA. Utah and Florida DOTs successfully implemented LCCA (1) to improve the public’s understanding of the funding needed to maintain pavement performance and (2) to fully utilize maintenance records to better allo- cate limited preservation and maintenance dollars. Their efforts support the projects under way at Minnesota and Washington State DOTs to develop the building blocks needed for LCCA. The holistic view of a system of assets by P3 concessionaires may provide insight to state high- way agencies on how to further improve LCCA methods and models to optimize investments. Multiyear mainte- nance budgets were also noted by P3 concessionaires to allow for better engineering decision making and preven- tative maintenance schedules. Also, engaging operations and maintenance staff in the project planning, design, and construction stages was viewed as a way to further reduce overall life-cycle costs. The next chapter outlines the overall study findings and recommendations for further research in the area of LCCA.