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.

9 BACKGROUND This chapter summarizes current practices used to design and evaluate pavement on which ADAB designs are based. Pave- ment design and analysis is a complex subject that cannot be presented here in its entirety. This synthesis presents a brief summary of influential factors on pavement design, introduces common pavement design methodologies, and presents an eco- nomic analysis framework that leads to the decision of whether ADAB is suited for a pavement project. In pavement-type selection, each project is evaluated individually. Many states provide pavement selection guidelines, but when no clear pavement design alternative exists for a project, the agency may design rigid and flexible pavement alternatives for bidding the project as ADAB (Hallin et al. 2011). ADAB concepts are founded on developing pavement design alternatives that maintain a similar level of performance over the same analysis period (FHWA 2012). The decision to use ADAB is a process that begins with an evaluation of pavement- type selection. Project-specific factors, as well as economic and performance issues, are evaluated during pavement-type selection (Hallin et al. 2011). When a preferred pavement design alternative is not clear, ADAB provides an avenue for agen- cies to competitively bid the different alternatives. The anticipated benefits for the ADAB are increased competition among contractors, reduction in overall costs, and a process that competitively selects the most economical alternative based on bid material prices (Wimsatt 2009). To successfully implement the ADAB process, agencies have emphasized the need for indus- try involvement, support of agency leadership, and a clearly defined, transparent process (Temple et al. 2004). The purpose of this chapter is to introduce the concepts of ADAB programs by • presenting the current practices based on a recent survey of state highway agencies; • introduce pavement design concepts and current methods for developing equivalent pavement design alternatives; • illustrate the importance of rehabilitation timing and performance data; and • provide a discussion of engineering considerations that factor into the decision to use ADAB on a pavement project. AGENCY USE OF ALTERNATE DESIGN/ALTERNATE BID FOR PAVEMENT PROJECTS The survey found that a majority of the ADAB pavement projects are awarded through the design-bid-build (DBB) process and about half the states that use ADAB also apply it to design-build (DB) projects. A majority of the states that use ADAB have a defined ADAB process with specifications and policies. The majority of states that use ADAB responded that the ADAB process is an integral part of the pavement-type selection for their agency. The framework of an ADAB program includes the processes for pavement design and LCCA. Eighty percent of the states that responded yes to using ADAB procedures also used an LCCA-based methodology using deterministic inputs for pavement-type selection. PAVEMENT DESIGN CONSIDERATIONS IN ALTERNATE DESIGN/ALTERNATE BID Pavement design methodologies provide engineers with a framework for collecting project-specific information to engineer pavement structures that have performed well based on prior experience or expected performance. This section briefly CHAPTER TWO ALTERNATE DESIGN/ALTERNATE BID DESIGN ASPECTS “The SEP-14 work plan permits MDOT to develop concrete (PCC) and hot mix asphalt (HMA) pavement cross sections for a project that is structurally equivalent. HMA and concrete paving contractors are then allowed the opportunity to competitively bid on the project. This process is expected to increase competition, which may result in more favorable bids for MDOT.” Youngs and Krom SEP-14 Report (2009)

10 presents key factors in pavement design, commonly used pavement design methodologies, and the most common pavement design methods used in states using ADAB. The available literature shows that fundamental pavement design procedures are not substantially altered to accommodate ADAB (Ahlvers 2010; Roark 2011; Duncan and Holtz 2012). Pavement Design in States Using ADAB and Developing ADAB Alternatives In the survey, agencies were asked which flexible and rigid pavement design methodologies that are currently in use for flexible pavement design, 47% of agencies are using AASHTO 1993 and 26.7% are using AASHTO-MEPDG (Mechanis- tic–Empirical Pavement Design Guide). The remaining respondents use state-developed methodologies or a combination of AASHTO 1993 and AASHTO MEPDG. For rigid pavement design, one-third use AASHTO 1993, one-third use AASHTO MEPDG, and one-third use a combination or a state-developed methodology. Fourteen states responded that pavement design is adjusted for equivalent designs when performing ADAB. In general, the states that adjusted their design for equivalency noted they are currently using AASHTO 1993 in their pavement design process but use AASHTO MEPDG when perform- ing equivalent pavement designs. Several agencies mentioned they are currently performing MEPDG model calibration and refinement of prediction models. Many respondents explained that they are transitioning from AASHTO 1993 to MEPDG. Implementation of MEPDG often includes calibrating the prediction models to meet local conditions and materials. MEPDG provides a way to base a pavement design on material performance characteristics. One survey respondent illustrated the need to incorporate performance specifica- tions to ensure that the pavement materials placed in the field match the design input parameters. Survey responses about the use of MEPDG combined with the information found in the literature led to the conclusion that MEPDG implementation provides a common platform for pavement design equivalency evaluations, which could potentially enhance the ADAB process. Common Pavement Design Methodologies Pavement design methodologies require numerous design inputs to develop a structural design that will perform at a designed serviceability level over a chosen design life. The literature review found that pavement design has been slowly transitioning over the past 20 years from purely empirical design guides to the MEPDG. Commonly used pavement designs are presented in Table 2 (Pierce and McGovern 2014). Similar pavement designs were reported in the survey. TABLE 2 AGENCY USE OF PAVEMENT DESIGN METHODS Method New Construction Rehabilitation Number of Agencies Asphalt Concrete Asphalt Concrete AASHTO 1972 7 2 5 1 7 AASHTO 1986 1 0 2 0 2 AASHTO 1993 35 23 31 19 39 AASHTO 1998 Supplement 4 11 4 8 13 AASHTO MEPDG a 12 10 10 7 13 Agency Empirical Procedure 7 1 9 3 13 WINPAS (ACPA 2012) 0 5 0 4 7 MS-1 (AI 1999) 1 0 3 0 3 ME-based Design Table or Catalog 1 3 0 2 3 Other ME Procedure 8 3 6 2 11 Other 5 7 7 8 14 a A number of agencies indicated that the MEPDG is currently being used or under evaluation; however, only three agencies indicated that the MEPDG has been implemented. Pavement design is rooted in empirical design methods. Increased computing power capabilities have facilitated main- stream integration of computer models into pavement design to simulate stress-strain material behavior, traffic loading, and climate over the design life. This pavement design methodology is called mechanistic empirical pavement design. Computer modeling has introduced mechanistic components into pavement design but ongoing model calibration, empiri- cal factors, and engineering judgement still play a significant role in the final structural design of a pavement (Souliman et al. 2010).

11 The AASHTOWare® MEPDG is a pavement design software package that incorporates mechanistic models as well as empiri- cal factors, climate, and forecasted traffic levels to simulate pavement performance over the design life. MEPDG uses laboratory performance test results of pavement materials to model and predict performance based on anticipated traffic and environmental conditions during the design life (ARA 2004). One study looked at using predicted performance from the MEPDG for pavement management decisions. A similar framework could be investigated to determine future rehabilitation timing based on MEPDG predicted performance rather than a fixed rehabilitation schedule. This would be useful in comparing alternatives in LCCA. The MEPDG uses material characterization models as inputs into pavement distress models that are used to predict pavement performance (ARA 2004). The MEPDG provides a common platform for flexible and rigid pavement design and allows agen- cies to consider material properties in predicted pavement performance (Velasquez et al. 2009). Studies comparing actual versus predicted performance have been performed for both asphalt and flexible pavements (Mogawer et al. 2011; Vandenbossche et al. 2011). According to the survey and literature, MEPDG research and development in the form of model calibrations are an ongoing effort for agencies. A recent study looked at using performance testing as a pay factor adjustment to provide a way for agencies to establish relationships between material properties, performance, and cost (De Jarnette et al. 2013). De Jarnette et al. showed that it is possible to develop performance-related specifications based on MEPDG predictive models and set pay adjustment factors related to predicted performance at the time of construction. The study conclusions also emphasize that sensitivity analyses be performed when establishing the performance targets and development of the adjustment factors be done transparently (De Jarnette et al. 2013). This type of approach might address some of the specification challenges that were cited by agencies in the alternative pavement-type SEP-14 program. Pay adjustment factors might incentivize contractor innovation and competition. Agencies are encouraged to allow incentives when they lead to long-term cost savings (Hallin et al. 2011). Studies similar to De Jarnette et al. will be beneficial in developing implementation plans for incentivizing innovative materials and could lead to long-term benefits. Influential Factors in Pavement Design Common pavement design methodologies have universal factors that significantly influence the pavement structure. Traffic level and functional class of the roadway are key considerations. Typical functional classes include residential, local roads, collectors, arterials, and interstates. In general, the functional class will influence the traffic/truck distribution. The traffic level is often converted to an equivalent single axle load (ESAL) for developing a pavement design. The percentage of truck traffic will greatly impact the total number of ESALs (Huang 2004). In the MEPDG, the design moves away from using ESALs and requires detailed traffic-load spectra. Soil conditions will greatly influence the carrying capacity of the pavement system. Soil stiffness, uniformity, drainage, and seasonal changes resulting from freeze-thaw conditions are all important considerations in the design phase (AASHTO 1993). The need for soil stabilization can factor into the decision to use ADAB. For example, a project in Kansas that required soil stabilization for only one of the alternatives created differences in earthwork calculations for the two design alternatives and complicated the ADAB process to the point that it was difficult to determine whether pavement costs or earthwork costs drove the final winning bid (Gisi 2009). Weather and climate will have an important influence on the mechanical properties of the pavement system because soils are subject to bearing-capacity changes (FDOT 2013). Many agencies develop material specifications to prevent damage from in-service environmental conditions (ARA 2004). Several SEP-14 project reports recommend further work on specifications (Youngs and Krom 2009; INDOT 2011; Roark 2011). The materials can play a substantial role in the performance of the pave- ment, and new pavement design methods are using material properties as inputs into design parameters. Project Factors That Influence the Use of ADAB The survey asked about project-specific characteristics that determine whether a given project is suitable for ADAB procurement. The responses indicated that the characteristics that make a project suitable for ADAB are projects with high levels of traffic and a high percentage of truck traffic. Respondents indicated that ADAB is not suitable either where adjacent lane continuity was required or where there are issues due to unstable subsoils. A majority of respondents pointed out that vertical geometry, availability of local materials, pavement noise, safety, and traffic control through the construction zone have little or no influence on ADAB use. Project designs bid using ADAB typically complete an analysis that includes reconstruction or major rehabilitation (MDOT 2013; West et al. 2013). The pavement-type selection generally applies to projects that are greater than 1 mi in length (FDOT 2013), and many agencies set a minimum project length requirement. For pavement projects bid using ADAB, Michigan DOT recommends that paving typically be the controlling operation for the construction schedule (MDOT 2013). Shoulders can be

12 a factor when considering ADAB. One study reported that the MoDOT Design Division calculates a separate LCCA factor for shoulder pavement when considering pavement-type selection (Wimsatt et al. 2009). During an alternate bid design, agency effort is perceived to increase because additional designs are performed, additional plans may be required, and each alternative may require different construction staging (Wimsatt et al. 2009). Pavement eco- nomics influence the decision to use ADAB, and major factors include the timing and cost of future maintenance and reha- bilitation strategies. The rehabilitation plan and design will also influence the equivalence of the two alternatives. A study in Oklahoma (Jeong and Abdollahipour 2012) looked at developing the timing of activities and costs by studying the Oklahoma DOT’s pavement management system. It found these factors to have a significant influence on the LCCA because the future maintenance and rehabilitation costs determine the adjustment factor for the bids. Based on a review of rehabilitation practices, many states have recommended timings for the first rehabilitation as well as the rehabilitation’s service life (Luhr and McCullough 1983; Jeong and Abdollahipour 2012). This will also influence remaining service life values at the end of the analysis. The Oklahoma study recommends that more performance and treatment data be gathered for rigid and flexible pavements, which can be used to update the LCCA models accordingly (Jeong and Abdollahipour 2012). Addi- tional details for determining a bid adjustment factor and DOT variations on that theme are found in chapters three and six. PAVEMENT DESIGN ALTERNATIVES Survey respondents were asked about the critical factors that are considered when choosing between competing design alter- natives. The following are significant factors: • Traffic level and composition • Suitability for functional class • Pavement-type continuity • Work zone issues • Soil complexity. Other less critical factors reported in the survey include low tolerance for pavement noise and the availability of local materials. Pavement-Type Selection NCHRP Report 703 provides guidance on pavement-type selection, and many state agencies have a pavement-type selection guide that can be used to determine the preferred pavement type. DOT pavement-type selection guides typically outline both economic and noneconomic factors that influence the selection decision (Hallin et al. 2011). For instance, projects in congested areas typically seek pavement types that minimize disruption, create the lowest possible hazards to traffic, and have long initial service lives regardless of the relative economics (AASHTO 1993; FDOT 2013). Another promising source for pavement-type justification and for improving data-driven decisions is information routinely collected for agency pavement management sys- tems, which typically contains roadway performance data. Information can be analyzed at both the project level and the network level to develop a better understanding of the performance of pavement alternatives (Von Quintus and Moulthrop 2007). Conti- nuity with existing pavement sections and construction considerations will also have a significant influence on the selection and design process, such as an anticipated future widening or seasonal construction factors (Hallin et al. 2011). Recyclability and sustainability of the pavement can also have both economic and noneconomic influences on the pavement selection decision. Reuse is especially important in areas where only marginal aggregates are available. Other factors that may influence the decision are incorporating test sections for research purposes, agency preference, and the use of innovative meth- ods and materials. Hallin et al. (2011) posit that innovations from industry, academia, or abroad be considered by agencies for implementation when research has found them to be both cost-effective and able to provide equal or better performance. Analysis of Pavement Design Alternatives ADAB requires agencies to develop comparable pavement design alternatives. According to FHWA’s RealCost® User Man- ual, alternative can be defined as “the complete set of activities that define a means to achieve the performance goals of a project, including the initial and future activities. In LCC analysis, all alternatives being considered for a single project will equally meet the performance requirements of the project.” An activity is defined as “a specific action performed by the high-

13 way agency, such as initial construction or major rehabilitation. An activity is defined by its agency costs, its service life, and its effects on highway users. An activity is a component of an alternative.” To comprehensively evaluate the pavement design alternatives, an LCCA is performed. FHWA and many states have guidelines for conducting an LCCA. The pavement design alternatives are compared by selecting future reconstruction and maintenance activities to analyze the LCC of all potential alternatives over an analysis period. To calculate the design thickness and choose construction materials, a pavement design methodology is followed. It is important that design assumptions and calculations be documented and carefully prepared. As shown in Table 3, Washington DOT’s pavement selection guide provides a table of comparable design thicknesses. In the guide, additional assumptions and design inputs are provided. Agency documentation and communication of design inputs and assumptions with industry can be key factors to success in alternate bidding. TABLE 3 EXAMPLES OF STRUCTURALLY EQUIVALENT PAVEMENT DESIGNS Source: WSDOT (2011). The California DOT (Caltrans) has developed extensive recommendations on future rehabilitation and maintenance strate- gies (Caltrans 2013). Caltrans’ LCCA guide provides example rehabilitation and maintenance strategies that designers can integrate into their pavement alternatives. Overall, the fundamental development of pavement design methodology remains unchanged for ADAB, but many factors influence the decision for pavement-type selection. ADAB has been used successfully in multiple states, showing that the pro- cesses and policies presented in this chapter are being used successfully based on survey results. Similar fundamental pavement design processes are used in design-build projects. In addition to the design alternatives, design-build allows for alternative technical concepts (CTC & Associates 2014). Only limited information was available on the use of contractor pavement design in the context of alternate bidding practices per end product specifications and alternative technical proposals. The MoDOT case study in chapter six provides some guidance on how this has been implemented in the ADAB process, but future research and potential impacts to the pavement systems would increase the amount of available information and guidance for agencies. Evaluating ADAB Pavement Alternatives Using LCCA Projects suitable for alternate bidding are those where no clear pavement material is determined to be the preferred option. FHWA’s technical advisory recommends addressing the following components: equivalent designs, discount rate, uncertainty issues, maintenance and rehabilitation strategies, and noneconomic factors such as recyclability, appropriate application, and user delay costs (FHWA 2012). FHWA also recommends the use of a bid adjustment factor based on future maintenance

14 and rehabilitation costs. Throughout the pavement design and economic analysis, a number of engineering judgements and assumptions are required to provide a reasonable comparison between pavement alternatives. FHWA provides guidance for performing an LCCA with the intent to compare the total costs of design alternatives to make the most effective transportation investment decisions (Walls and Smith 1998). The LCCA primer developed by FHWA emphasizes that initial costs are only a part of the overall transportation investment, and LCCA provides a method for com- paring the long-term benefits and costs of competing alternatives (FHWA 2002). The LCCA analysis for pavements typically includes costs for initial construction, maintenance, rehabilitation, and user costs over an analysis period that incorporates at least one major rehabilitation (Hallin et al. 2011). The five steps in LCCA recommended by FHWA include (FHWA 2002): 1. Establish design alternatives 2. Determine activity timing 3. Estimate costs (agency and user) 4. Compute LCCs 5. Analyze the results. Figure 2 shows a typical expenditure stream diagram for an LCCA. According to the survey, some agencies also consider user and maintenance costs in their LCCA. FIGURE 2 Expenditure-stream diagram (FHWA 2002). The LCCs will be influenced by various input factors such as the analysis period, the selected discount rate, and the estimated timing for each activity. In order to study the influence of LCCA factors and assumptions on the cost analysis, a sensitivity analysis should be performed (Walls and Smith 1998). Once the LCCA is performed, many agencies use it as a basis for developing a bid adjustment factor. The bid adjustment factor is calculated based on future maintenance and rehabilitation costs for each alternative. Technical Advisory T 5040.39 recommends calculating the adjustment factor as the difference in the net present value (NPV) of anticipated future M&R costs. The cost difference is added to the initial bid price of the alternative with the higher net present value (NPV) for antici- pated future M&R costs to determine the lowest responsive bidder (FHWA 2012). In a similar way, the Iowa alternate bidding specification subtracts the adjustment factor from the initial bid price of the alternative with the lower NPV for anticipated future M&R costs, but the net result is the same (FHWA 2008).

15 SUMMARY Conclusions One conclusion was drawn from the information reviewed for this chapter. Based on the survey responses, most agencies use the MEPDG when performing equivalent pavement designs. It appears that the MEPDG provides improved tools and a com- mon platform for designers when developing the ADAB design alternatives. Effective Practices No effective practices beyond the implementation of the MEPDG design method were found. Implementation of MEPDG often entails calibration of material characterization models and pavement distress models to improve predicted performance. Future Research The information considered in this chapter identified five possible future research needs: • A number of SEP-14 project reports recommend future research to develop better specifications. Both the survey and one SEP-14 project report indicated a need for guidance on mixture tolerances. A recent study looked at using perfor- mance testing as a pay factor adjustment that provides a way for agencies to establish relationships between material properties, performance, and cost (De Jarnette et al. 2013). • Research to quantify ADAB performance and treatment data for rigid and flexible pavements is required to update the current LCCA models accordingly (Jeong and Abdollahipour 2012). • Research to quantify the cost to the agency for developing ADAB alternatives and plan development is needed. This was studied within the Indiana DOT but the program required only one plan set instead of two (INDOT 2011). • Research and additional guidance for contractor pavement design and impacts to an ADAB program is recommended. • Research in pay adjustment factors and end-method specifications is also recommended. Expansion of research similar to De Jarnette et al. (2013) to integrate MEPDG into material requirements, pay adjustment factors, and contractor incentives to enhance the performance of long-term infrastructure investments.