Quality Assurance in Design-Build Projects (2008)

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.

INTRODUCTION Quality issues are always present during the design and construction phases of a project. DB, however, also adds the option of allowing financing and post-construction services to be included in the DB contract. Following the success of SEP-14, Innovative Contracting, the FHWA in 2004 issued SEP-15, Public Private Partnerships, to “encourage tests and experimentation in the entire development process for trans- portation projects” (Blanding 2006). The most common examples of additional services in DB are design-build- maintain, design-build-operate-maintain, and finance-design- build-operate-maintain (also known as concession or PPP). These delivery methods that extend beyond the completion of construction provide DOTs with more options to deliver qual- ity projects that meet long-term financial goals. At face value, the impact of adding post-construction op- tions to a DB project would appear to have a positive effect on quality, because the design-builder is financially tied to the cost of operating and maintaining the project. Therefore, if the constructed product is of high quality, it would follow that its post-construction operation and maintenance costs would be minimized. The report of the AASHTO Construc- tion Management scan tour to Canada and Europe (DeWitt et al. 2005) found that five of the six countries visited “rely on the private sector for essentially all highway maintenance. This is accomplished through a series of term maintenance agreements in which routine maintenance and repair is done in accordance with performance contracts.” The team also observed project delivery methods that “allocate more risk to the private sector and/or create more motivation for total life cycle maintenance and operation solutions from the private sector” (DeWitt et al. 2005). Apparently, the Canadians and Europeans have found ways to realize the potential benefit of linking DB projects with post-construction operations and maintenance through their performance contracts. U.S. Experience Things are not quite as clear cut in the United States. An in- terview with the project manager of the Hudson–Bergen design-build-operate-maintain (DBOM) project, a $600 mil- lion light rail project in New Jersey, indicated that with the post-construction operations and maintenance portion of the contract, the agency did not feel the same pressure to inde- pendently verify quality during design and construction that it 62 did in other project delivery methods (Touran et al. 2007). However, these options are rare in the United States, and Nancy Smith, a well-known construction attorney who was involved in a number of the large DB projects, puts it this way: It is unclear whether a 5-, 10-, or 15-year maintenance/warranty obligation provides the contractor with a real incentive to ensure project quality. Since there are very few contracts in place that require long-term maintenance or warranties, it remains to be seen whether such an approach will ultimately be successful in assuring project quality (Smith 2001). Therefore, it appears that the jury is out regarding the rel- ative impact on U.S. project quality that post-construction options might create. The literature is sparse on this subject, as are RFPs for these types of projects. The Finnish Road Ad- ministration published a report detailing its strategy for se- lecting project delivery methods and reviewed its experience with DBOM and finance-design-build-operate-maintain (FDBOM). It found that “from the standpoint of the quality assurance of contracts, specification of functional or perfor- mance requirements will become very important . . . require- ments that are poorly specified or difficult to objectively measure, appear in tenders that produce high prices from contractors or even make it impossible to achieve the desired level of service” (“Procurement Strategy of the Finnish Road Administration” 2003). Here it appears that an experienced public owner has discovered that being able to adequately de- fine and perhaps, more importantly, measure quality in the post-construction phase is of primary importance. Another report that detailed the lessons learned by the U.S. Depart- ment of Energy in its privatized projects stated that the “team responsible for managing the contract also includes engi- neers and managers who ensure conformity to product spec- ifications, schedule adherence, safety, and quality assur- ance.” Therefore, it can be concluded that how quality is defined in these long-term post-construction contracts is just as important as how quality is defined in design and con- struction contracts. Synthesis Survey Results The national survey yielded some interesting results for this particular issue. Four questions directly addressed the use of post-construction options and their impact on the quality of the project. One question asked those with experience if the quality management plans required in projects with post- CHAPTER SIX VARIATIONS ON DESIGN-BUILD QUALITY ASSURANCE FOR POST-CONSTRUCTION OPERATIONS, MAINTENANCE, AND FINANCING ARRANGEMENTS

63 construction options were different than those normally used in typical DB contracts. Interestingly, only 35% of the re- spondents said “yes,” most likely because this system is so new to the United States that DOTs have not yet gained enough experience with its results that they would feel com- fortable modifying their tried and true systems. Another question asked the respondents to rank the impact on project quality of three common post-construction options. (Note: those without experience could answer “N/A.”) The results were as follows: 1. Warranty provisions: 56% rated this option as having either “high” or “very high” impact on project quality. 2. Follow-on maintenance agreement: 62% rated this op- tion as having either “high” or “very high” impact on project quality. 3. PPP/Concession: 58% rated this option as having ei- ther “high” or “very high” impact on project quality. In the last two questions that relate specifically to follow- on maintenance and PPPs, the most popular answer was “very high impact.” Thus, the respondents confirm the idea that the addition of a post-construction option to a DB project will in- fluence the quality of the constructed product. This finding is validated by the results of the perception survey in which the majority of the respondents held the belief that using DB proj- ect delivery positively affects final project quality. DESIGN-BUILD-MAINTAIN The idea of attaching a post-construction maintenance con- tract to a construction project as a mechanism to create a financial incentive to enhance construction quality has been debated for years. It is based on the theory that putting the construction contractor at risk for the cost of maintenance after completion will cause the contractor to be more careful during production. It is intended to “(a) reduce the amount of highway agency resources required on a highway project; (b) reallocate performance risk; (c) increase contractor inno- vation; (d) increase the quality of constructed products; and (e) reduce life cycle costs of highway projects” (Queiroz 1999). Thus, it is merely a natural extension of the concept of linking a post-construction maintenance contract to a DB project to create design-build-maintain (DBM) project deliv- ery. This practice has been done regularly in Europe, where warranties have been used to defuse the issue of supposed lower quality on DB projects (Carpenter et al. 2003). U.S. Design-Build-Maintain Experience One of the early experiments with DBM was a project to widen two-lane New Mexico State Highway 44 to a four-lane road redesignated as US Route 550. The project became known as NM-44 and consisted of the reconstruction of 118 miles of state highway. The use of DBM project delivery with public-private financing allowed the project’s delivery period to be reduced from the original estimate of 27 years to 3 years (“Contractor to Honor Road Warranty” 2004). The project came with a 20-year warranty at a cost of $62 million. The New Mexico DOT was complimented in the press for “not telling Koch [the DBM contractor] how to build the road . . . avoiding micromanagement allows the contractor to use in- novative practices often not available to the public sector” (“Contractor to Honor Road Warranty” 2004). Several years after the road opened to traffic, it began to reveal distresses that were not associated with normal wear and tear, starting a controversy that was avidly followed by the local press. A New Mexico DOT official told a pavement symposium that the “DOT should have been more involved in quality control” (“Contractor to Honor Road Warranty” 2004). Therefore, even though there was a post-construction contract for main- tenance, this project had quality problems associated with either design deficiencies or construction quality. Adding maintenance to a DB contract was also used by the Utah DOT when it issued the RFP for the I-15 makeover in an- ticipation of the 2002 Winter Olympics (Postma et al. 2002). Initially, the Utah DOT decided to request that a 20-year maintenance period be included in the proposals; however, some contractors believed that this was not possible at a “reasonable cost” (Postma et al. 2002). Taking this into consideration, the Utah DOT changed the requirements to in- clude only 10 years of maintenance on specific parts of the project. The requirement was for the “Contractor to provide a cost option to provide maintenance for the first five years and options for five one-year periods up to a total of ten years of maintenance” (Postma et al. 2002). Included in the con- tract for the maintenance option was the requirement that the contractor provide a “Maintenance Quality Management Plan” that would be approved by the Utah DOT if the option were to be exercised. In the end, the Utah DOT did not exer- cise the maintenance option because it was “very comfort- able with the quality of the completed project and felt that they could perform the required maintenance more effi- ciently and at a lower cost with their normal maintenance program” (Postma et al. 2002). There were some issues associated with this project that led the Utah DOT to not exercise the maintenance option. The contract for I-15 was written so that the contractor would receive a cost adjustment based on the Federal-Aid- Highway Construction Urban (Composite) Index. The index was very volatile at the time of the contract and “UDOT [Utah DOT] was reluctant to proceed with the adjustment” (Postma et al. 2002). The Utah DOT recommended that a different cost index be used that takes into account the time value of money (Postma et al. 2002). Another issue written into the contract was the sharing between the Utah DOT and the contractor of some of the catastrophic risks that may occur over the life of the project. Simply stated, “UDOT was not willing to take this risk” (Postma et al. 2002). Another issue from the Utah DOT experience is the importance of having a maintenance plan that clearly defines what the

64 DOT wants in the RFP. Because of how the specifications were written for the project, . . . an effective Maintenance Plan would be difficult to develop. The requirements for maintenance were spread throughout the contract, which made it difficult to identify and monitor. The specifications also had some ambiguities, which made it difficult to interpret the requirements and assure UDOT would get what they intended (Postma et al. 2002). In spite of the issues, the major outcome was that the Utah DOT was satisfied with both the design and construction qual- ity and made the business decision to not exercise its option to invoke the follow-on maintenance agreement. Thus, this proj- ect stands in contrast to the New Mexico project regarding the impact of a post-construction maintenance warranty on the constructed quality of a major transportation project. In addition to these examples of DB projects, it is worth mentioning the results of some DBB projects that included warranties. A previous survey of DOTs in Arizona, Florida, Michigan, Ohio, Virginia, and Washington State (Final Report: Six-State Survey . . . 2001) “found that using war- ranties resulted in improved quality in the highway work con- tracted” (Carpenter et al. 2003). A similar study done by the Wisconsin DOT also found that warranted projects performed better than similar nonwarranted projects when measured with the Performance Distress Index and the International Roughness Index (Asphaltic Pavement Warranties . . . 1998). Maintenance Contract Transfer Criteria Including a post-construction option to operate and/or main- tain a DB project adds another set of quality factors that must be included in the project’s quality management planning. These entail the quality performance criteria for operations, maintenance, and transfer after the contract period is com- pleted. As previously stated, the DOT is probably trading the ability to “relax” its QA activities during design and con- struction for having to evolve an entirely new PQA responsi- bility for conducting quality management activities that measure the quality of maintenance and operations. “Mainte- nance quality assurance is a process that uses quantitative quality indicators and statistical analysis to assess the perfor- mance of maintenance programs” (Adams and Smith 2006). This study found that, even though the experience with main- tenance quality assurance is evolving and more DOTs are gaining some experience, no published standard exists from which owners can benchmark their own programs. There is also a requirement in post-construction QA to develop a set of criteria for the quality of the project at the end of the concession period. The agreement with the con- cessionaire on the Texas DOT SH 130 project included a 15-year maintenance agreement and contained 36 separate transfer criteria on everything from pavement condition to traffic signals (SH 130 Turnpike Project Exclusive Develop- ment Agreement 2002). These quality performance criteria must be connected to the compensation arrangement associated with the post- construction option. A study in Australia confirmed this statement when it found that “payments will also be related to performance criteria” (Pelevin 1998). A similar arrange- ment is in effect in Portugal where payment to the conces- sionaire is made based on criteria for safety and lane closure performance each month (Fernandes and Viegas 1999). Therefore, it can be seen that although making the design-builder financially liable for the post-construction operations and maintenance may alleviate some quality management issues, it brings an entirely new set of quality management requirements for the owner, and because this process is new in the United States, there is not much expe- rience available for DOTs to leverage in developing these new PQA criteria and processes. DESIGN-BUILD-OPERATE-MAINTAIN DBOM is a system that is widely used in Europe as a primary way to deal with QA issues. A report by Cox et al. (2002) cites several advantages for this project delivery method: Adding maintenance or operation and maintenance has numer- ous advantages. The primary advantage is that these contracts create a lifecycle responsibility for the design-builder. The same company that designs and constructs the highway is also respon- sible for maintaining quality over a period of years. This situa- tion provides an incentive to deliver better quality in the initial design and construction of the project because the design-builder will be responsible for additional maintenance and repair costs if the initial quality is inadequate. Another report addresses this issue as follows: Design quality is ensured because the team must operate what they build. The process is not necessarily driven by the lowest bid; therefore, quality and robust design become more important over the life of the project life and the DBOM team’s commit- ment period (Wiss et al. 2002). Nevertheless, the addition of follow-on maintenance and operation contracts create a new level of quality management for the public agency. The Cox report states it like this: The addition of operation and maintenance to the design-build contract solves the problems of design administration, construc- tion administration, quality control, and use of warranties. How- ever, the drafting and enforcing of operation and maintenance performance criteria creates new issues that are not commonly dealt with in most highways agencies in both Europe and the United States (Cox et al. 2002). The conundrum articulated previously is solved by the use of measurable performance requirements. These are then used as benchmarks against which the public agency can evaluate the design-builder and/or maintenance con- tractor’s quality of performance. The idea is to require that the entity that designs the project is accountable for its product because “it owns the design and construction

65 methods used to achieve the performance requirement” (Cox et al. 2002). FINANCE-DESIGN-BUILD-OPERATE-MAINTAIN (CONCESSIONS—PUBLIC-PRIVATE PARTNERSHIP) FDBOM project delivery is increasing in use in the United States owing to the deterioration of the nation’s infrastruc- ture and the limited amount of funding provided by govern- ment. This is often called a concession or a PPP to credit the private participation in financing. The U.S.DOT defines a PPP as “a contractual agreement formed between public and private sector partners, which allows more private sector participation than is traditional” (Saunders 2006). This is an- other tool that DOTs can use to deliver needed infrastructure projects when there are not sufficient public funds for project financing. PPPs are commonly used in Europe, most notably the United Kingdom, with Australia, New Zealand, and South Africa also leading the way (Carpenter et al. 2003; Saunders 2006). “The French national highway system is al- most exclusively operated by concessionaires . . . . Portugal is the most aggressive employer of concessions” (Cox et al. 2002). In Portugal, the method can be described in the fol- lowing terms: The [PPP] scheme merges two decisions—one involves the al- location of responsibility and risk for the production and main- tenance of a motorway to a private consortium, and the other the reimbursement for this effort over a period of years by the state (Fernandes and Viegas 1999). The inclusion of private financing changes the equation with regard to project quality. In projects that are fully funded with public funds, the owner can take the moral high ground and insist on getting the project for which it is paying. In PPP, the owner is no longer furnishing 100% of the fund- ing, but it still has a constituency, the traveling public, to pro- tect. The contract creates a legal structure between the public agency and its private partner (often called a developer or concessionaire in these projects). The public agency must develop a system “to monitor the performance of the con- cessionaire adequately, ensuring that [the] public interest is defended properly throughout the concession period” (Fernandes and Viegas 1999). Quality Assurance Organization in Privately Funded Projects The shift of financial input that occurs in a PPP is from the owner providing 100% of the financing during design, con- struction, and operations to the concessionaire furnishing a significant amount of funding. Although this may seem inherently to invalidate the need for owner-controlled QA activities, FHWA Technical Advisory 6120.3 requires that DOTs perform design and construction quality oversight to satisfy their legal responsibilities to the public. As a result, this causes a shift in the DB project QA model introduced in chapter two to accommodate the shift in financial liability that accompanies a PPP project. The Texas DOT is currently in the process of delivering a number of PPP projects using the authority of FHWA Special Experimental Program 15 (SEP-15) (SEP-15 Program—Public Private Partnerships 2006). To receive FHWA approval for use of federal funds for portions of a project under SEP-15, the DOT must con- summate an “early development agreement” (EDA) with the FHWA that details the DOT’s approach to assuring quality in the project. The EDA for the Trans-Texas Corridor ties the post-construction warranty to the quality when it states that “TxDOT anticipates that a long-term general warranty will be a critical part of any Facility Agreement . . . to ensure quality in design and construction of the Facility” (TTC35 Early Development Agreement 2005, italics added). The DOT carries this approach further when it develops the final agreement with the concessionaire in which quality man- agement roles and responsibilities are spelled out. Another example from Texas, the “Exclusive Development Agree- ment” (not to be confused with the previously defined EDA) for the SH 130 Turnpike Project (SH 130 Turnpike Project Exclusive Development Agreement 2002) illustrates this principle. Developer shall complete and manage such portions of the De- velopment Work . . . in a manner satisfactory to the TTA [Texas Turnpike Authority] and in accordance with the Project Schedule, including . . . Providing quality control and quality assurance with respect to the Development Work, subject to TTA oversight, involvement and directives; TTA shall carry out certain design and construction related activities, including . . . Performing certain quality assurance oversight services in connection with the Development Work (italics added). Figure 23 results from combining an analysis of the above project’s contractual quality management require- ments, information found in the literature (Yuan et al. 2007), and consultations with appropriate members of the Texas DOT staff. It illustrates how the DB project QA model that is developed in chapter two applies to a PPP project. In this specific case, Figure 23 provides an exam- ple of the PQA model as applied to a PPP by the Texas DOT (“Quality Assurance Program Components” 2007). It must be noted that in this form of project delivery, the design- builder is called the concessionaire and will operate and maintain the completed project for some concession period after construction completion. This adds an additional set of quality requirements for post-construction service, opera- tion, and maintenance to the DB PQA universe. Addition- ally, the concessionaire brings its own financing to the project and, as a result, the model’s use of the design and construction acceptance decision points is modified be- cause the concessionaire becomes financially liable to correct any design or construction deficiencies during the operations and maintenance period. Thus, in this case, the owner has chosen to retain an independent engineering con- sultant to perform what the Texas DOT terms as “owner verification testing.” This supplements the IA and QA

66 FIGURE 23 Design-build project quality assurance model for a public-private partnership. activities, which are now assigned as the concessionaire’s responsibility. Additionally, this entity may perform the QA activities necessary to verify that the concessionaire is meeting its contractual requirements for the quality of ser- vice, operations, and maintenance after construction is complete and the project has been opened to traffic. Financial Incentives for Public-Private Partnership Quality To influence the concessionaire’s quality-related design and construction decision, the owner in a PPP must create a finan- cial incentive for the developer to aspire to a high level of qual- ity. Portugal has extensive experience with this type of project delivery and has found that good quality can be obtained “by making them the objectives of the concessionaire . . . there are bonuses and penalties based on performance, so the con- cessionaire has pecuniary incentives to present high stan- dards” (Fernandes and Viegas 1999). This process is also used in the Netherlands, where a “fixed lump-sum payment [is made] every 3 to 6 months if the desired performance cri- teria are achieved” (Cox et al. 2002). In Canada, the incentive is provided through the use of a specific pay item for quality management performance. A paper presented at the 55th Annual Quality Congress details the process as follows: A novel feature of the DDB [PPP] Agreement is the identifica- tion of Quality Management as a “line item” in the Guaranteed Maximum Price (GMP) of the project. This item, approximately 4% of the GMP, is scheduled as a series of maximum monthly amounts (varying with season and certain project milestones) to be paid out based on an independent assessment of the effec- tiveness of the QMS [quality management system]. . . . The method of assessment introduced by the IA [independent audi- tor] at the commencement of the agreement used a random auditing of an exhaustive list of requirements in the DDB [PPP] agreement. Random samples proportional to the budgeted or estimated cost of fulfilling each requirement were audited each month. The degree of conformance was reported as the total conformances divided by the total month’s sample size (% of conforming observations) (Collier et al. 2001). The Canadian public agency then uses the degree of con- formance to determine the percentage of the monthly QMS payment earned in accordance with Figure 24. Thus, the developer has essentially 4% of its project cash flow at risk and the aggressive payment algorithm creates a financial incentive to ensure satisfactory quality. The authors of this report go on to conclude: To compete . . . in today’s environment, a company must be willing to agree to world-class requirements for quality. By aligning quality requirements . . . and incorporating them di- rectly into the agreements, the advantage went to the proponent [proposal] who would accept extra responsibility and seek to delegate it effectively. This arrangement also allies the Owner and the Developer closely in pursuit of their common interest— the Public-Private-Partnership (Collier et al. 2001). A similar performance payment scheme has been used in public transit projects in Utah and Oregon (Touran et al. 2007), although these projects were not PPP projects. Inter- estingly, both of the U.S. projects used the same 4% formula as was used in Canada, although their monthly payment al- gorithm was not as aggressive as the Canadian one. Both

67 FIGURE 24 Percentage of monthly payment as a function of quality performance score (Collier et al. 2001). public agencies reported that they were satisfied with the re- sultant quality of their constructed facilities. Additionally, they believed that the scheme drove the design-builder to be proactive in identifying potential issues and correcting them before they were assessed by the owner. This impression was confirmed through an interview with one of the project design-builders. SUMMARY The relative newness of post-construction options for opera- tions and maintenance makes it difficult to draw any conclu- sions on their impact on the quality of transportation projects. Intuitively, the idea of holding the design-builder liable for the potential costs of inadequate quality through either an extended warranty or an operational period would seem to cre- ate an incentive toward a robust design and well-constructed project. However, the experiences in the United States are decidedly mixed. The trends observed in previous studies of warranties seem to confirm the notion, but the experience on the New Mexico 44 project graphically refutes it. Although this may have been an anomaly, the one conclusion that can be made from this portion of the analysis is that DOTs would ben- efit from investing a significant amount of creative energy and thorough investigation when deciding to deliver a project using some form of post-construction option. The issues iden- tified in the Utah I-15 project are very instructive, and the results of the AASHTO European scan report point to the potential adoption of methods and measures used overseas, where they have a decent amount of post-construction option experience, rather than to the few U.S. projects. Finally, if the decision to employ a post-construction option is made, the DOT may recognize that although it may have found a mech- anism to mitigate the risk of design and construction quality, it has now created a situation in which it must manage the new set of quality factors associated with quality during post- construction operations and maintenance.