Alternate Design/Alternate Bid Process for Pavement-Type Selection (2017)

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.

22 CHAPTER FOUR DEPARTMENT OF TRANSPORTATION ALTERNATE DESIGN/ALTERNATE BID PROCUREMENT POLICIES, PROCEDURES, AND PROGRAMS INTRODUCTION The information provided in this chapter is primarily drawn from the survey of state DOTs described in chapter one. It is supplemented by information found in DOT pavement design manuals and procurement documents, as well as the general literature. It is important to note that only 16 of 40 respond- ing state DOTs reported using ADAB. Those that don’t were asked to answer questions about their current pavement-type selection procedures and the extent to which LCCA was used to inform that decision. Therefore, the chapter is split between reporting the response of those who have used ADAB and those who have not. As reported in the previous chapter, DOTs turn to ADAB when engineering and economic analysis does not indicate a clear choice between competing pavement types. Contractors are provided two or more pavement-type alternatives and are permitted to bid their preferred design alternative (Hallin et al. 2011). Pavement-type selection involves balancing the risk resulting from the variations in material costs and performance. ADAB allows contractors to bid current market prices for competing equivalent design alternatives before a pavement-type selection decision is made (FHWA 2012). Different design alternatives are typically evaluated with the inclusion of an adjustment factor to account for the different future maintenance and rehabilitation costs (Hallin et al. 2011). Although one would expect that LCC-based analysis would be an integral part of alternate bidding practices, there is no standard on how agencies should manage their alternate bidding practices (Jeong 2012). Clearly, the adoption of an LCC- based pavement-type selection approach considerably facilitates the practice of an ADAB program. Many crucial factors for a successful ADAB program are direct products of an LCCA. In its 2012 advisory on alternate bidding, FHWA summarizes the motivation behind alternate bidding as the “risk for an agency due to potential variations in material costs and performance” in the pavement-type selection process (FHWA 2012). The primary goal of ADAB procedures is to reveal the current market prices for competing alternatives. Most agencies have sound estimates of the actual market prices before opening their projects to bidding. However, under volatile market condi- tions, or when agencies have no previous experience with certain design types, they may find it necessary to receive market prices for alternatives before they make a final pavement-type decision (FHWA 2012). As such, ADAB can be a robust bidding mechanism that allows agencies to make pavement-type decisions using real-time market pricing and eliminates the need to make broad assumptions on market prices. The value of alternate bidding may be most effectively leveraged when uncertainty over construction costs is a primary factor in selecting a preferred design alternative. In some pavement-type selection cases, primary or secondary pavement-type selection factors may not point toward a clear design alternative. In addition, noneconomic factors, such as constructability, availability of local materials and construction expertise, conservation of materials, and continuity of adjacent pavements, may not be sufficient to justify a pavement-type decision. INDUSTRY PERSPECTIVES ON ALTERNATE BIDDING Both the concrete and the asphalt industries have supported wider use of ADAB practices. The Asphalt Pavement Alliance has supported alternate bidding “provided the guidelines are based on technical merits” (APA 2010). The Portland Cement Association (PCA) has been more vocal in advocating alternate bidding practices by drawing attention to improved competi- “Bidding pavements ‘head-to-head’ allows the open market to determine what is constructed …not outdated assumptions made during the evaluation/selection process years before letting.” Lenz (2010)

23 tion between the two industries, especially in agencies where a single pavement type is common practice. PCA stated that alternate bidding “gives the contractor a choice to bid on either a concrete or an asphalt option, thereby increasing the number of bidders on each job and enhancing competition” (PCA 2010). However, both industries often disagree on their assumptions of maintenance and rehabilitation strategies, and on whether the specified alternatives by the agencies amount to equivalent pavement designs. The assumptions made in calculating bid adjustment factors over the analysis period, such as the main- tenance and rehabilitation costs, service life, and performance levels, can also contribute to industry dissatisfaction, and in some cases may lead to disputes between the agencies and contractors (Temple et al. 2004; Ahlvers 2010). At least two such disputes were reported in the Ohio and Nevada DOTs’ responses to the survey. ALTERNATE DESIGN/ALTERNATE BID PROCEDURES In response to the growing adoption of ADAB practices among the state agencies, FHWA endorsed the use of alternate bid- ding methods in 2012 (FHWA 2012). It is now clear that many states that use ADAB procedures have experienced real ben- efits from this practice. The main benefits include reduced project costs and increased competition. The agency policies on ADAB procedures show a significant degree of variation of across states. The main source of this variation lies in the states’ pavement-type selection policies (Crawford 2014). Whether or not LCC-based pavement-type selection processes are used, the assumptions underlying the LCCA are important factors for the differences observed in ADAB practices. Two main issues must be addressed before alternates can be compared. First, the underlying assumption of all ADAB methods is the presence of equivalent design, without which competing alternates cannot be meaningfully compared. Adjust- ing for the differences in LCCs thus becomes an important consideration for alternate bidding practices. FHWA guidance on ADAB states that an LCCA is the fundamental framework for comparing competing alternatives (FHWA 2012). Second, ADAB is typically indicated when the expected LCCs of competing alternatives are reasonably close to one other and when there is no preferred pavement type among the competing alternatives. Although there is no consensus on a single threshold level among the state transportation agencies, thresholds in practice range from 10% to 20% difference in LCCs among equivalent design alternatives (Hallin et al. 2011). FHWA guidance, for instance, calls LCCs similar if the NPV of the higher cost alternative is less than 10% higher than the lower cost alternative (FHWA 2012). Equivalent Design and LCCA Although agencies have different approaches to achieving design equivalence among competing alternative pavement designs, the expected benefits of alternate bidding depends on the design equivalence of competing alternatives. Pavement-type alternatives are compared using LCCA and are considered equivalent if they provide similar level of service over the same period. Given design requirements for traffic level, reliability, and service life, the pavement service levels are expected to sustain comparable levels of ser- vice over the desired design life. Because competing design methods often have unequal traditional design lives, the LCCA period should be made equal by including at least one major rehabilitation cycle (FHWA 2012). FHWA also encourages basing equiva- lent design alternatives on similar performance thresholds at the end of the design period. Because of the inherent variation in the performance levels of different design alternatives, both the terminal and average yearly levels of pavement roughness and relative distress can be achieved through a careful consideration of alternative maintenance and rehabilitation strategies (Hallin et al. 2011). The specification of similar service levels over the common performance period depends on the underlying maintenance and rehabilitation strategy assumptions for each alternate. Each strategy must reflect realistic agency-level maintenance and rehabilitation costs, calibrated to simulate the pavement service levels with associated future costs. Maintenance and rehabili- tation activities are primary factors in the LCCA and development of adjustment factors. Guidance concerning maintenance and rehabilitation strategies can be found in NCHRP Report 703: Guide for Pavement-Type Selection (Hallin et al. 2011). Adjustment Factors Inclusion of an adjustment factor to account for the difference in the alternates’ future costs is the recommended practice (FHWA 2012) and has been commonly adopted by the DOTs that responded to the survey. One of the most widely used methods is the so-called A+B+C method. The method adjusts each contractor’s base bid (A) by adding two adjustment fac- tors. Construction duration and user costs are captured in component B, whereas expected future overlay, maintenance and rehabilitation costs, and associated user costs, if any, are included in component C. A variation of this methodology in which no adjustment is made of construction duration, A+C, is also commonly used.

24 In general, the winning bid in alternate bidding is determined by finding the lowest bid among the set of lowest bids for each pavement-type alternative after adjusting for LCCs, as shown in Equations 1 and 2 (Hallin et al. 2011). Min [Min (Alternate 1 Bid), (Min (Alternate 2 Bid) + C))] (1) Where: C (Bid adjustment factor) = NPV (Future Costs of Alt 2) – NPV (Future Costs of Alt 1) (2) LCC Threshold for Equivalent Designs When selecting LCC thresholds, there is a direct relationship between the expected number of bids to be awarded through alternate bidding and the potential project cost ranges for each alternative pavement type. Setting higher threshold levels results in a larger number of projects to qualify for alternate bidding, because more project LCCs would fall within the speci- fied threshold levels. Conversely, low LCC thresholds reduce the number of potential ADAB projects. Given the administra- tive and engineering bid costs associated with additional pavement types, each agency can balance the expected benefits from alternate bidding (such as receiving market prices for competing alternatives, increasing competition, and reducing costs) against the costs of adopting ADAB practices. It can be noted that although the latest FHWA guidance on alternate bid- ding process recommends a 10% cutoff level for similar LCCs (FHWA 2012), the survey results indicate that agencies have threshold levels as high as 20%. The decision to use alternate bidding can also be based on other factors, such as functional classification and the project size. For example, MoDOT uses 7,500 square yards of pavement surface area as the cutoff level above which all projects are let using ADAB. SURVEY RESULTS FOR AGENCIES PRACTICING ALTERNATE DESIGN/ALTERNATE BID The literature documented variations in the management of alternate bidding practices (Ahlvers 2010; Roark 2011; Duncan and Holtz 2012). The variations include whether pavement-type selection and alternate bidding processes are integrated, modifying alternative pavement types to ensure equivalent pavement design, the LCCA threshold below which alternates are deemed to pro- vide similar service levels, inclusion of user costs in LCCA, and how maintenance and rehabilitation strategies are determined. The survey results show that ADAB is used in 40% (16) of the agencies that responded to the survey. This finding confirms that alternate bidding is now an accepted decision aid in states’ pavement-type selection procedures. Reflecting the variation among states in their approach to alternate bidding, 31% (5) of the states that use ADAB indicated that they do not follow defined processes for projects awarded using ADAB procurement. Agency Practices and ADAB The agencies that practice ADAB identified ADAB benefits as detailed in Figure 4. The largest cited benefit was increased competition. Other potential ADAB benefits, such as cost savings to the agency, introduction of alternatives that contribute to the overall performance of the pavement, and accelerated project completion, also were cited, although there is no consensus among agencies on these benefits. Agencies have divergent expectations on the cost savings potential of the ADAB practices for the agency. Since the analysis of the survey results found that those agencies that tend to disagree with the statement of “ADAB provides cost savings to the agency” also tend to disagree with other expected benefits of the ADAB practices, a number of key characteristics of the agencies’ alternate bidding practices are shown based on their perception of the cost savings potential of ADAB procedures. The three (agree/strongly agree, neutral/unsure, and disagree/strongly disagree) groups shown in Figures 5 and 6 correspond to the survey response: “ADAB provides cost savings to the agency.” Figure 5 suggests that agencies that selected “agreed” to “ADAB provides cost savings to the agency” have more years of experience with ADAB procedures. A stronger association emerges when the agencies’ median number of ADAB projects is grouped by their perceptions on ADAB’s cost savings potential (Figure 6). Those agencies that expect higher cost savings tend to use alternate bidding more frequently. The agencies that agree with ADAB’s cost savings potential reported having 22 ADAB projects per year, whereas those that are either neutral/unsure or disagree on ADAB’s cost savings potential had 5.5 and 3.5 ADAB projects per year, respectively.

25 FIGURE 4 Expected ADAB impact on cost and pavement program improvements. FIGURE 5 Median years of ADAB experience. FIGURE 6 Perceived ADAB cost savings vs. median number of ADAB projects per year.

26 The survey’s also found that increased HMA-PCC lane-mile percentages are associated with agency adoption of defined pavement-type selection processes. Of the 13 ADAB practicing agencies that reported their pavement-type shares in lane miles, the four states with no defined pavement-type selection processes had a lower average PCC pavement types (4%) than the nine states that reported practicing formal selection processes (7%). Additionally, those agencies with defined ADAB processes, specifications, and policies demonstrated higher median levels of ADAB use when compared with those that did not have formal ADAB procedures, as shown in Figure 7. FIGURE 7 Formal ADAB processes vs. median number of ADAB projects per year. ADAB and Variation of Agencies’ Pavement-Type Selection Decisions The survey results also suggest that the variation in the agencies’ lane-mile pavement-type compositions may be associated with the agencies’ perceptions on ADAB’s cost savings potential. Figure 8 shows that agencies with a higher percentage of PCC pavement type in lane miles tend to attribute higher cost savings benefits from ADAB use. This may result from agency perceptions of more competitive local markets among the concrete and asphalt industries, which in turn may perceive more cost savings benefits from ADAB practices. Alternate bidding in states with established concrete and asphalt industries, when compared with those where a single industry dominates the DOT market, could increase the average number of bids received and may influence agencies’ perceptions on cost savings. FIGURE 8 Median share of PCC pavement type.

27 SURVEY RESULTS FOR AGENCIES NOT PRACTICING ALTERNATE DESIGN/ALTERNATE BID The survey found that 24 of the 40 responding agencies did not use ADAB practices. Of the 22 states that answered the question of whether they have interest in adopting ADAB practices in the future, nine (41%) expressed interest. When asked whether ADAB use would benefit their pavement-type selection practices, 65% were unsure about potential benefits of ADAB on pavement-type selection and 15% did not see a potential benefit to their program. Figure 9 shows that the most common reasons for agency interest are increased competition and cost savings, followed by flexibility in design and construction, accelerated project completion, and improved use of local construction materials. FIGURE 9 Reasons for interest in ADAB use in the future. The reasons reported by the agencies that did not express interest in potentially implementing ADAB included insufficient experience with the process, lack of resources, and industry resistance (Figure 10). FIGURE 10 Reasons for no interest in ADAB use in the future.

28 COMPARING PRACTICING AND NONPRACTICING AGENCIES The survey results suggest that the agencies that practice ADAB procedures are different from their nonpracticing counter- parts in important ways (Figure 11). A large majority of ADAB-practicing agencies have documented pavement-type selec- tion processes. Agencies that practice ADAB procedures not only have higher adoption levels of LCCA methodology for their pavement-type selection processes but also tend to have a higher percentage of documented LCCA and pavement-type selection processes. The other two notable differences between the two groups are their approaches to modifying their design methodology to achieve equivalent design and the inclusion of user costs in their LCCA. Although a clear majority of the prac- ticing states report changes to their typical design methodology, less than a third of the nonpracticing states make any changes. This finding may suggest that agencies that have alternate bidding practices may also pay closer attention to equivalent design by changing their typical pavement-type design methodology to enhance design equivalence. In addition, practicing agencies include a major rehabilitation cost in their LCCA more frequently compared with the nonpracticing agencies. Both groups of states are similar in their inclusion of user costs in LCCA. Although all practicing and nonpracticing agencies have the about same (5%) median share of PCC pavements, considerable variation exists between the agencies that expressed an interest in ADAB practices and those that do not (Figure 12). The survey results indicated that the agencies with higher shares of PCC pavements also have higher levels of interest for future ADAB FIGURE 12 Median share of PCC pavement (percentage of total lane miles). FIGURE 11 Life-cycle cost practices and ADAB use.

29 use. One plausible explanation for this relationship is that as agencies chose among different pavement types with increasing frequency, the potential utility of alternate bidding in aiding such decisions becomes more tangible and gives agencies higher incentives to practice alternate bidding. As such, agencies with little or no potential in deviating from their default pavement type also tend to express no interest in adopting alternate bidding to supplement their pavement-type selection practices. USE OF ALTERNATE DESIGN/ALTERNATE BID POLICIES OVER TIME The results of the survey supplement the widely reported alternate bidding benefits in the literature. NCHRP Report 703 investi- gated states’ pavement-type selection processes that contained a comprehensive survey about pavement-type selection and a smaller survey that targeted alternate bidding (Hallin et al. 2011). To provide a summary of the changes in ADAB use by state transporta- tion agencies between 2011 and 2016, the results of NCHRP Report 703 and this synthesis were compared with study the trends of ADAB use over time (Table 5). A more detailed agency-level comparison is also provided in Appendix A (Table A13). TABLE 5 COMPARISON OF SURVEY RESULTS Survey Question: Does your agency use ADAB contract award processes for pavement construction contracts? NCHRP Report 703 Results (Hallin et al. 2011) NCHRP 47-02 Survey Results ADAB in use (yes) 17 16 ADAB not used (no) 25 24 No response (NR) 8 10 Total 50 50 ADAB Adoption Rate 34% 32% A comparison between these two sources’ data reveals that the agencies’ ADAB adoption rate has remained fairly stable between 2011 and 2016. NCHRP Report 703 found that 17 of the 42 responding states used ADAB (34%) in 2011. Similarly, this synthesis’ survey received 16 agency reports confirming ADAB use out of the responding 40 states (32%). The synthesis survey has also found that four new states (Alabama, Arkansas, Maryland, and Michigan) have adopted ADAB use since their response to the 2011 survey. On the other hand, three states (Kansas, Louisiana, and Montana) reportedly discontinued ADAB over this 5-year period. SUMMARY The results of the survey supplement the reported alternate bidding benefits in the literature. Yet the survey results also sug- gest that benefits from alternate bidding may actually depend on a few key agency characteristics. Among such key factors are the length of the agency’s ADAB program in use, the presence of formal pavement alternate bidding and pavement-type selection processes, and the variation of pavement-type decisions from agencies’ default pavement types (e.g., higher disper- sion of pavement types in the agency roadway network, as indicated by higher percentage of PCC lane miles, for instance). The following are general findings related to the use of ADAB at the agency level. 1. When a clear design alternative cannot be chosen with historical pavement material costs or when such costs are not reliable owing to material costs volatility, ADAB permits agencies to defer the final pavement-type selection decision to reflect current market prices on letting day, which may indicate that alternate bidding may be expected to be used more frequently during volatile periods in pavement material prices. 2. ADAB reveals true market pricing preventing decisions based on cost assumptions made months or years before letting. 3. There is a significant level of variation among agencies in the mechanics of how each has implemented ADAB. 4. The agencies that perceive a potential cost benefits from ADAB also tend to have formal processes in both alternate bidding and pavement-type selection, as well as longer histories of alternate bidding use and a higher number of ADAB projects per year.

30 5. The agencies that expect to use alternate bidding on a continuing basis tend to have better defined processes than those that prefer alternate bidding on an ad hoc basis. 6. Alternate bidding is expected to be more common in agencies where pavement-type selection decisions do not favor a default pavement type. 7. If the agency’s pavement-type selection process is not currently based on an LCCA, ADAB provides incentives to do so. 8. The differences in the expected initial construction costs of competing design alternatives and their associated future costs directly influence the size of adjustment factors. To the extent that initial costs are offset by future costs, a thor- ough sensitivity analysis of the underlying LCCA assumptions is important. 9. Two groups of agencies are expected to derive the most benefit from ADAB. First, those agencies that routinely specify different pavement types may be able to increase the number of bidders per contract by allowing both concrete and asphalt paving contractors to bid on the same projects. Second, states considering the introduction of previously unused pavement types may experience benefits by comparing the LCCs of traditionally used and new pavement types. 10. The earlier an agency adopts an ADAB and establishes the equivalence of competing alternatives in pavement-type selection using LCCA processes, the more pavement selection decisions could be made using actual bid day market conditions, ultimately accruing a benefit of the most economical alternative. Effective Practices No effective practices were observed in this chapter. Future Research No suggestions for future research were identified in this chapter.