Legal Aspects of Performance-Based Specifications for Highway Construction and Maintenance Contracts (2013)

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 requirements such as hammer size, cushion replace- ment, jetting limitations, and maximum stress levels in driving the piles. These combined performance and de- sign specifications are often referred to as “composite specifications” or “mixed specifications.”32 While liability will be discussed in depth later, it is worthwhile for readers to note that composite specifica- tions are often the most complicated to resolve when problems arise, as contractors and owners will each view the specification as being “more” one way or the other—depending on what it is in dispute. For example, in the bridge scenario described above, assume that the specification precluded any type of pre-jetting. When the contractor starts driving the piles, it finds that it is exceeding the maximum stress levels and that the piles are starting to crack at elevations well above the mini- mum tip elevation. Assume further that the owner and contractor agree that the solution to this is to pre-jet to within 5 ft of the minimum tip elevation, which opera- tion costs the contractor more money and time than it had planned. The contractor would likely argue that the specification was defective, in that the design require- ments led it to believe that it need not price any pre- jetting operations. The owner would likely argue that it had drafted a performance specification and that the relaxation of the jetting requirements was to accommo- date the contractor. As discussed in more detail in Sec- tion VIII, the answer to this situation is very much de- pendent upon how much discretion the contractor truly had in meeting the performance specification, and whether the owner’s design specifications were so “cookbook” as to make the owner responsible for the consequences of the change in approach. F. Proprietary Specifications The term “proprietary specifications” is intended to identify desired products or processes by manufac- turer’s name, brand name, model number, or other unique characteristic. Even if a manufacturer is not explicitly stated, a specification is considered proprie- tary if only one manufacturer can meet the specified requirements. These are used when an owner wants to closely control product selection and, in some circum- stances, develop a higher level of design based on more precise information obtained from the manufacturer’s data. Their use introduces the potential disadvantage of unnecessarily eliminating/narrowing competition. They might also require products with which the contractor has perhaps had little or poor experience (e.g., slow de- livery); this can lead to higher bid prices or charges of favoritism.33 “Brand name or equal” clauses fall within this cate- gory of specifications, with the presumption that there is an “equal” to the “brand name” specified. While 32 This example is derived from a Maryland DOT project constructed in the mid-1980s, known as “Removal and Re- placement of Bridge No. 3097 on Maryland Route 150 (Eastern Avenue) over the Back River,” Contract No. B 752-501-471. 33 FHWA 2010 Technical Advisory, supra note 5. FHWA allows the use of a proprietary product (i.e., a “sole source” or “brand name with no equal”), there are severe constraints on this practice, and the agency is generally required to demonstrate that there is no suit- able alternative to the specified product.34 G. Application of Specifications to the Compaction Process While it is beyond the scope of this digest to discuss the technical aspects of specifications in any detail, it is useful to see how the specifications apply to a common element of highway work such as compaction: • A method specification would establish all of the steps for the compaction process. This would include the type and mass of the compaction equipment, the num- ber of passes, the moisture content of the fill, and its layer thickness. • A performance specification using an “end product” approach would specify compaction in terms of a re- quired value for properties of the fill when placed, such as density, moisture content, and air voids. • A performance specification using a “warranty” approach would specify behavior over a period of time in terms of maximum permissible settlement. As discussed above, all of these approaches lead to a series of questions that the agency needs to consider related to measurement, verification of compliance, and recourse against the contractor for noncompliance. III. CURRENT HIGHWAY INDUSTRY USE OF PERFORMANCE SPECIFICATIONS As described previously, there are myriad reasons why highway agencies have expressed dissatisfaction with method specifications. FHWA’s 2004 publication, Performance Specifications Strategic Roadmap: A Vi- sion for the Future35 summarizes them well. Method specifications “could not deal with rewarding a contrac- tor for better-than-minimum practice” and “could not consistently deal with work that was outside the bounds of ‘reasonably close conformance.’” Method specifications “inhibited innovation” and provided no incentive to a project manager to consider “departures from standards” and “departures from procedure.” These observations are similar to those made in other construction sectors, even though other sectors generally afford their construction contractors more latitude in terms of implementing the means and meth- ods of construction. Building owners concluded that the broad use of prescriptive specifications impeded general and trade contractors from using their vast knowledge of the construction process to develop better solutions. Owners also concluded that they faced exposure to 34 23 C.F.R. 635.411(a); see generally FHWA 2010 Technical Advisory, supra note 5. 35 Available at http://www.fhwa.dot.gov/construction/ pssr04tc.cfm, supra note 5.

10 change orders when the prescriptive specification was inadequate. Consider the area of fire protection sprinkler sys- tems. These systems were historically designed by the owner’s architect/engineer. However, many of these designs had to be totally redesigned, at great expense to the owner, to deal with field conditions and the vagaries of local code officials. This prompted the building indus- try to shift to the use of performance specifications for this work, leaving it to the fire protection trade contrac- tors to determine the precise design that would be needed to satisfy code requirements. While the move to performance specifications in other construction industry sectors was originally trig- gered, to a large extent, by risk considerations (i.e., eliminating change orders for defective design specifica- tions), the interest in using them has gone far beyond managing change order exposure. Owners discovered that contractors often had better technical approaches than designers to meet the owner’s ultimate objectives, particularly when factoring in price and constructabil- ity. Performance specifications were convenient vehicles for letting contractors provide this value. The growth of construction management at-risk and design-build, which place a high value on integrating the design and construction teams early in the design process, has also prompted a greater use of performance specifications in other industry sectors. Performance specifications are routinely used on buildings that seek Leadership in Energy and Environmental Design (LEED) certifica- tions and energy savings.36 While the highway sector has been discussing the use of performance specifications and performance- based contracting for quite some time, progress in tran- sitioning to a wider use of these techniques has been relatively slow. There has been progress in using per- formance specifications in areas driven by technology, such as traffic management systems, tolling facilities, and other systems using instrumentation and commu- nications platforms. The same is true with operation and maintenance services, particularly on concession contracts. However, the design and construction of physical structures, such as roads and bridges, has not routinely used performance specifications, other than for pavements. 36 See, for example, Nadine M. Post, Fee Holdback Raises Eyebrows, ENGINEERING NEWS RECORD, May 14, 2012, avail- able by subscription at https://enr.construction.com/ engineering/subscription/LoginSubscribe.aspx?cid=22847. A. Pavement Major progress has taken place in moving pavement from a predominately method specification process to one where the state of the practice is replete with ex- amples of end result and statistically-based QA specifi- cations. This move started with pavement research supporting the benefits of performance specifying, and has resulted in their use around the country. In recent years, pavement research has largely fo- cused on developing prediction models for both rigid pavement (i.e., portland cement concrete (PCC) pave- ment) and flexible pavement (i.e., hot-mix asphalt (HMA) pavement) that can support the development and use of PRS. This has resulted in a number of trial PRS projects in the United States and the development of the PaveSpec software to help state highway agencies develop PRS for their state,37 as well as FHWA’s crea- tion of the WesTrack facility in Nevada.38 Several states have developed PRS for PCC pave- ment, including Indiana, Wisconsin, Florida, and Ten- nessee. The PRS for each state was developed using PaveSpec software and local climatic conditions (for pavement performance prediction) and local costs for maintenance and rehabilitation. Relative to HMA, a number of tests have occurred at WesTrack, including the use of full-scale accelerated load testing, to support the adoption of PRS. As evident by the examples cited above, warranty specifications have been broadly used for both PCC and HMA pavements to address actual performance over time. A warranty specification has the advantage of being able to cover certain types of distresses that could not be predicted under a PRS (e.g., corner cracking and scaling). It can also cover certain functional characteris- tics that would be difficult to predict using predictive 37 PaveSpec (currently version 3.0) is a software program available through FHWA that enables transportation agencies to develop PRS and predict the performance of a constructed pavement. It is also considered a technology transfer tool to enable contractors and highway agencies to get a better under- standing of what it takes to construct highway performance pavements. According to FHWA, some of its specific capabili- ties include: • Simulation of pavement in terms of 1) transverse crack- ing, 2) transverse joint faulting, 3) transverse joint spalling, and 4) pavement smoothness over time. • Application of a user-defined maintenance and rehabilita- tion plan to compute life-cycle costs. • Development of pay factor charts for the following accep- tance quality characteristics: 1) strength, 2) thickness, 3) air content, 4) smoothness, and 5) consolidation around dowels. • Computation of contractor pay factors from actual con- struction test results for the above five acceptance quality characteristics. • Assistance in executing sensitivity analyses on a given developed PRS. More information is available at http://www.fhwa.dot.gov/pavement/pccp/pavespec/index.cfm. 38 WesTrack was FHWA’s test facility in Nevada for devel- oping PRS for HMA pavement construction.

11 models, such as texture/texture loss and skid resis- tance. Warranties have been more widely applied to HMA pavements than to PCC pavements. This may be due to the fact that short-term (e.g., 5-year) warranties on PCC pavement are not useful—even poorly designed and constructed PCC pavements will often last 5 years before showing significant distress and deterioration. As discussed later in this digest, while longer-term war- ranties can overcome this, there are commercial prob- lems with obtaining them given the perspective of the surety bond industry. This issue is not as significant on HMA pavements, where typical performance character- istics (e.g., ride quality, rutting, friction, and cracking) can be observed in a short-term period. B. Bridges There are a number of research studies on the use of performance specifications for bridges, with the focus being on structural concrete and bridge decks. The typi- cal quality characteristics for these specifications were strength, stiffness, permeability, and air content. Some of these research studies addressed optimizing mix de- sign, sampling and testing, selection of specification limits, and pay factors. However, while there has been research in these areas, it appears that bridge specifica- tions have remained relatively prescriptive, requiring that concrete be batched, mixed, placed, and cured in accordance with the plans and specifications. There is also little evidence that performance specifications are being used robustly on long-term contracts (e.g., design- build-operate-maintain (DBOM) and concession con- tracts), as underlying design requirements for those contracts referenced agency or other FHWA-approved standards. There are several practical challenges with using performance specifications for bridges. There is a gen- eral reluctance by “safety-conscious bridge engineers” to give contractors decisionmaking responsibility— creating few opportunities for innovation and risk transfer. The long service life of most bridge compo- nents also makes short-term warranties or mainte- nance agreements ineffective from a risk-transfer per- spective, as there is a likelihood that the contracting entity will dissolve over time or its initial costs to put in place life-cycle costing would make the initial costs of the construction too high. C. Geotechnical Features Most geotechnical-related specifications on highway projects seem to be currently based upon either method specifications (e.g., number of roller passes for earth- work construction) or end-result specifications (e.g., achievement of 95 percent compaction). Method specifi- cations have been relied upon because of the high corre- lation between construction methods and performance, and the absence of widely-accepted methods to evaluate performance characteristics during construction. While end-result specifications are used, they are often com- bined or substituted with prescriptive specifications. There is a practical challenge in balancing geotech- nical variability and test methods that could validate performance over the life cycle of the geotechnical fea- ture—i.e., during construction, after project completion, and long term. Low-frequency testing is the state of the practice, and a move to PBS requires, among other things, statistically valid assessments of performance characteristics based on high-frequency testing and monitoring. D. Work Zone Traffic Control The majority of work zone traffic control specifica- tions currently in use are method specifications that require the contractor to perform to a set of clear, spe- cific steps for work zone management, with no opportu- nity to deviate or innovate. This is in spite of industry findings that these traditional method-based specifica- tions for work zone traffic control do not provide an effi- cient and cost-effective means of managing the work zone. It appears that even those agencies that have in- cluded performance specifications for work zone traffic control have done so in title only. For example, some of these performance specifications will identify perform- ance goals that are aspirational (e.g., “Provide a safe travel corridor”), but do not tie these aspira- tions/objectives to a quantitative measurement strategy (e.g., “Limit work-zone crashes to two per month.”). Agencies have been more successful in effecting posi- tive changes to work zone traffic management by im- plementing innovative contracting techniques—as op- posed to strict performance-based traffic control specifications. Among the techniques that have reduced construction durations and minimized traffic disruption are A+B bidding and lane rental.39 E. Intelligent Transportation Systems and Operations and Maintenance Performance specifications are routinely used by agencies when they contract with an entity to develop intelligent transportation. This is similar to what oc- curs in other industries where innovative technology and management approaches are the centerpiece of the contract and the party seeking the approach is looking for innovation from the system’s developer. Consider a recent request for proposal (RFP) issued by the Virginia Department of Transportation (VDOT).40 The RFP contains a broad set of performance 39 See, e.g., Washington State DOT discussion of A+B bid- ding method (http://www.wsdot.wa.gov/Projects/delivery/ alternative/ABBidding), and lane rental (http://www.wsdot.wa. gov/Projects/delivery/alternative/LaneRental.htm). 40 See VDOT’s July 10, 2012, RFP (RFP # 150401) for “Transportation Operations Centers and Statewide Advanced Traffic Management Systems Services.” The purpose of the RFP is to establish a contract to provide VDOT with, among other things, Transportation Operation Center Floor Opera- tions, Intelligent Transportation System Infrastructure and Field Network Maintenance, and design of a Statewide Ad- vanced Traffic Management System (ATMS) Solution and Technology Support, http://www.virginiadot.org/business/