Construction Manager-at-Risk Project Delivery for Highway Programs (2010)

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.

85 INTRODUCTION A number of conclusions regarding the state of the practice of Construction Manager-at-Risk (CMR) were reached in this synthesis. The following protocol was used to reach a conclu- sion. To be included, the conclusion needed to be the result of triangulation between three or more of the sources used in the report. The source study instruments were the literature, the national survey, the solicitation document content analysis, and the case study projects with structured interviews of owners and contractors. The contractor interviews were used in two ways. First, they were a source of information such as the agency interviews, and second, the contractor interviews served as a loose validation of the conclusions and effective practices developed in the synthesis. The contractors confirmed that the information and opinions collected in the agency interviews, the literature, and the content analysis were actually present in practice. They also served as a second opinion with regard to the relative effectiveness of each practice. A number of lessons learned and effective practices for implementing CMR project delivery were also identified. These effective practices and lessons learned are drawn from the intersection of the literature and one of the other study instruments. Thus, the rationale for developing conclusions and effective practices is to be able to map them back to two or three sources on which all agree and further validate them by indicated construction industry confirmation. Finally, gaps in the body of knowledge were also identified and are pre- sented in this chapter as recommendations for further research on CMR project delivery. CONCLUSIONS The first conclusion aligns with the one reached by the Utah Department of Transportation (DOT) in its “2007 CMGC Annual Report” on CMR contract performance. This is fur- nished first because the Utah DOT is the leading DOT in developing and implementing CMR project delivery, having decided to use it on 13 federally funded highway projects and an additional 16 state funded projects. Those DOTs that intend to implement this project delivery method would do well to review the Utah experience when developing individual policies and procedures. One DOT employee stated: Use CMGC as the primary delivery method unless schedule is the principle driver. When a shortened delivery schedule is the primary motivation design build is preferable. If the contractor cannot deliver the project for a fair price then the fall back position is [Design] Bid Build [DBB]. I make this recommendation because we should always want the contractor’s input to reduce risk, cost, and construction time. We should only go to [Design] Bid Build [DBB] when we cannot get a fair price. We should not go to CMGC to reduce schedule because it commits us to the contractor too quickly and drives up our costs. The intent here is not to advocate the use of CMR as the default project delivery method, but rather to demonstrate that it can be implemented successfully by a typical DOT and that there are benefits, such as “the contractor’s input to reduce risk, cost, and construction time.” Therefore, as aptly put by the above interviewee, CMR is not a one-size-fits-all solution, but rather a solution for highway project issues that require the constructor’s input during design. That input can and does take the many forms discussed in chapter five. The important point here is that involving the construction con- tractor in the design process has proven to be beneficial. It was cited in the literature (12 times, Table 1). The impact of all the 12 possible preconstruction services furnished during design had an average rating between “high impact” (four of five) and “very high impact” (five of five) in the agency inter- views. Constructability reviews were the most often cited preconstruction service in transportation Request for Quali- fications/Request for Proposals in the content analysis. Early contractor involvement was rated as “of highest value” in six of ten case studies and the owner’s response on this item was validated by the same rating from all the interviewed contrac- tors. Thus, the output from five study instruments intersects to confirm the earlier DOT employee’s conclusion: CMR project delivery’s major benefit is contractor input to the preconstruction design process. The other synthesis conclusions are more specific in nature. Each, where appropriate, is presented with the study instru- ments that intersected to form the conclusion noted. The conclusions developed in this study are presented in the order of the CMR project delivery process chronology. They are not numbered in order of importance. 1. The following characteristics are reported to apply to agencies seeking to use CMR project delivery: • Sophisticated construction experience; • Adequate level of resources for current projects; and • Minimal policy, statutory, and political constraints on implementing CMR project delivery method. CHAPTER NINE CONCLUSIONS

2. Based on the literature and case studies, the project characteristics that make it a good candidate for CMR project delivery are shown in Table 16 and are as follows: • Sophisticated construction experience—agency has built complex or phased projects of the same order of magnitude as the project in question and has expe- rienced personnel. • Benefit from integrated design—agency’s design- bid-build (DBB) design contract can be modified to coordinate it properly with the preconstruction ser- vices contract. • Constructability essential for project success—design is not fixed. • Financial constraints—tight budget/sufficient funding authority. • Needs flexibility during construction—minimize disruptions to traffic and other impacted parties/has authority to begin construction activities before 100% completion (permitting, etc.). • Phased construction. • Opportunity to explore innovation or technology— design is not fixed; agency can deviate from standard materials or design features. • Opportunities for value engineering—funding rules allow transferring funds from one use to another within the project. • Third-party concerns—issues early in process (util- ities, railroads, business impact, public relations, etc.)—project has more than the “usual” stakeholders and they are willing to facilitate progress. 3. The case study agency and contractor ratings shown in Tables 33 and 34 lead to the conclusion that imple- menting CMR project delivery will not change design quality and may actually improve it. 4. The qualifications of the CMR’s personnel and its past project experience are the aspects that have the greatest perceived impact on project quality. 5. The timing of CMR selection relative to designer selection is important. The literature and interviews indicate that selecting the CMR at a point in time where it can influence fundamental design decisions before they are made adds value to the project. Table 1 contains ten citations each for “ability to fast-track,” “early knowledge of costs,” and “ability to bid early work packages,” all of which require bringing the contractor on board early in the design process. Four content analysis documents cited design completion as between 30% and 50%. Case study output found that four of ten projects selected the CMR as soon as possible after the designer and the remaining six cases made the selection before 30% design completion. This was validated by the contractor interviews. 6. The agency DBB design contract can be modified to specifically require the designer to coordinate its efforts with the CMR. 86 7. If cost or fee factors are going to be included in the selection process, then selecting the CMR after enough design progress to reasonably quantify and price the preconstruction scope of work would be indicated. If direct construction cost information is required, then CMR selection is better timed to happen when the agency has sufficient design detail to permit reason- able numbers to be submitted for the indicated costs that do not contain excessive contingencies. 8. Protests of CMR selection decisions are reportedly rare. Three protests were identified in the study and all were unsuccessful. 9. Preconstruction services are a distinct benefit to the project’s cost, schedule, and ultimate quality. The cost of preconstruction is a reasonable investment that accrues tangible returns. 10. CMR services furnished during preconstruction reduce design costs by permitting the designer to produce an appropriate design rather than a biddable design. In other words, the CMR can tell the designer when it has sufficient design detail to properly construct a given feature of work. Achieving these savings requires a high level of collaboration and strong spirit of partnering. 11. The study found that nine of ten case studies allowed the CMR to self-perform work packages of its choice. 12. The study found that nine of ten case studies allowed the CMR to have unlimited ability to self-perform as little or as much work as it finds appropriate. 13. The study found that eight of ten case studies did not constrain the CMR’s ability to prequalify and/or select subcontractors for the following reasons: • To get real-time pricing information, the CMR is allowed and expected to communicate with the sub- contractors it knows during preconstruction. • To obtain real-time technical information about best practices for subcontractor trade means and methods, the CMR is allowed and expected to communicate with the subcontractors it knows to access subject matter experts during preconstruction. • Studies have shown that competitive pricing is “preserved” without competitive bidding; therefore, requiring the CMR to award subcontractor work packages to an open field of competitors does not appear to save money. 14. Progressive guaranteed maximum prices (GMPs) are reported to add value to the CMR project by reducing the total amount of contingency in the GMP and by allowing an orderly method to price early work pack- ages and/or construction phases. 15. Incorporating a shared savings clause does not create a significant incentive to the CMR and may add a layer of administration whose cost is not recovered in its actual benefit. 16. Early work packages can be used to mitigate cost risk by locking in the cost of the materials, and services associated with those packages were cited by ten of

87 the papers listed in Table 1. This ability was also cited as “of highest value” by all but one case study project owner and all the interviewed contractors. 17. The major barriers to CMR implementation are confu- sion about its mechanics and whether enabling legis- lation exists to use it. LESSONS LEARNED There were three primary lessons learned. All deal with the administration and approval of fees earned by the CMR and the designer. These three were selected because they all furnish solutions to three major issues that an agency seeking to implement CMR will face. Thus, it is hoped that these will allow agencies new to CMR to start a bit higher on the learning curve based on the experiences discussed in this section. Coordinate Design and Construction Manager-at-Risk Contracts The first lesson comes from the Memphis Airport case study project and deals with modifying the DBB contract and struc- turing the design fee in a manner that creates an incentive for the engineer to willingly participate in the preconstruction process. Memphis did not modify its standard design contract for its first CMR project and found that obtaining the necessary collaboration with the CMR was not as effective as was hoped. The root of the problem was that the designer had not priced the number of design reviews required and was using up burning billable hours faster than it was used in a DBB design contract. The agency modified the design contract for the next phase of the project to coordinate design milestones with budget review points. It added an explicit requirement to coordinate the design work with the CMR’s construction work packages and mandated joint coordination with third parties. This gave the designer a chance to propose a fee that reflected the changed scope of design coordination that is present in a CMR contract. Memphis also created an incentive for the engineer to willingly cooperate with the contractor during preconstruction. Its modified contract included a clause that put 10% of the design fee at risk for the final quality of the construction documents. The at-risk portion was divided into 5% for design quality and 5% for construction issues owing to design quality problems. The contract clause extends from preconstruction and rates the outcome and resolution of construction problems, such as change orders and delays that result from poor design quality control (QC). This changed the relationship from one of unwanted interference to one where the consultant saw the CMR reviews as a benefit to its design QC process and solved the issue of preconstruction collaboration in a manner that not only facilitated the budget and schedule but more importantly enhanced the design QC program. Arizona State University’s CMR design contract has a clause to coordinate the efforts of the CMR as a part of the designer’s product and requires the designer and builder to jointly present the design submission. The lesson learned here is summarized as follows: Preconstruction collaboration does not happen automatically by merely deciding to use CMR project delivery. The requirements to collaborate are included in BOTH the design and precon- struction services contracts. Empowering Third Parties to Incentivize the Construction Manager-at-Risk The Utah Transit Authority (UTA) Weber County Commuter Rail case study project furnished an innovative way to increase the trust of third-party stakeholders and hence increase their cooperation on a major CMR project. The project ran through ten different municipalities and had to share active track and right-of-way with the Union Pacific Railroad’s most heavily trafficked freight route. There were significant numbers of utilities to deal with and many environmental and drainage issues to be solved, not to mention impacts on the public during construction. To deal with the monumental challenge created by the sheer number of third-party stakeholders, UTA assigned the responsibility of coordinating to the CMR and created a stakeholder panel to furnish feedback on a monthly basis. The panel included the impacted municipalities, the state environmental quality agency, and representatives from the railroad and the FTA. UTA empowered this panel through a clause in the CMR’s contract that put half the CMR’s con- struction fee at-risk. The panel met monthly to review the issues that arose in the past pay period and made a recom- mendation to UTA as to how much of the at-risk fee would be awarded in the monthly progress payment based on the contractor’s performance. This created an incentive for the contractor to maintain good public relations and be respon- sive to the concerns of the affected third parties. The system was effective in not only inspiring the CMR to act promptly and aggressively on corrective action to resolve issues as they arose, but it was also effective in giving the third-party stake- holders a sense of substantive involvement, which greatly reduced the number of third-party issues that needed to be solved. The lesson learned here is summarized as follows: The CMR contract can be used to not only create an incentive for good performance by the contractor but also can assist in third-party issue resolution by giving these parties a voice in the CMR’s actual profit. Simplify the Guaranteed Maximum Price Negotiation Process Up Front UDOT is the most experienced DOT in the country at deliver- ing highway projects using CMR. As a result, its CMR process has undergone a series of revisions as UDOT incorporated the lessons it learned and experimented with different variations on the CMR theme. When the case study agency interviews

were complete, the study team was left with the impression that the least understood portion of the process was negotiating the GMP. UDOT uses different GMP structures for projects with different requirements. Its suite of CMR options includes a unit price GMP, unit price with no GMP, and preconstruction services only followed by hard bid for construction. In a unit price GMP, there is no reason to negotiate a construction fee because the contractor’s profit and general conditions are contained in the unit prices. Therefore, this leaves the GMP negotiation to consist of verifying quantities of work and deter- mining if the unit prices are reasonable. UDOT further sim- plifies this process by requiring the unit prices for four or five major pay items in the CMR’s proposal. This nails down a sig- nificant piece of the GMP before the CMR contract is awarded. Attempting to estimate a given contractor’s profit is impos- sible for anyone but the contractor itself. Profit is a business decision that reflects everything from how much risk a given contractor sees in a project to how badly they need the work. The same thing can be said for estimating the general condi- tions as a separate line in the GMP. UDOT’s solution is elegant in that it uses the best features of its DBB project delivery process to simplify its CMR project delivery process. It also manages risk by applying the unit pricing in a progressive GMP. The effect is to reduce the contingencies by not requir- ing pricing for work packages that have insufficient design detail. Therefore, for an agency that is trying CMR for the first time, the UDOT CMR approach is a relatively minor shift from DBB. The only essential differences are the preconstruction services contract and not getting all the contract costs in a single bid envelope. The lesson learned here is summarized as follows: Simplify the process of establishing a reasonable and realistic GMP as much as necessary by putting many of its components into unit prices. Effective Practices Several effective practices were also reported. 1. The case study interviews noted that agencies can develop a documented procedure for selecting CMR as the project delivery method based on project char- acteristics. Additionally, a similar policy can be devel- oped for selecting the CMR contractor based on the same project characteristics. 2. A CMR selection process that is transparent, logical, and defensible appears to be less likely to be susceptible to protest. 3. Eight of ten case study agencies use the same quality assurance (QA) program for CMR as they do for DBB. Therefore, it appears that no modification is necessary to a DOT’s QA program to implement CMR project delivery. 4. The two most often cited preconstruction services in transportation projects were design reviews and con- 88 structability reviews. Both of these are essential com- ponents of the design QC program. Thus, detailing the roles and responsibilities for design QC for both the designer and the CMR in the procurement phase facil- itates collaboration. 5. Joint development of the preconstruction service cost model before commencing design allows the designer and the CMR to be able to leverage it to make design decisions and to benchmark value engineering savings. 6. Splitting the contingency between the owner and the CMR appears to make accounting for contingency allocation less onerous. 7. An open books approach to contingency calculation and allocation enhances the spirit of trust between the owner and the CMR. 8. Detailing the specific preconstruction services the agency wants to be provided in the preconstruction services contract in the solicitation document leads to responsive proposals. This is critical to getting a rea- sonable proposal if costs are included in the selection process. 9. Including the submittal of an outline of the proposed CMR project quality management plan with the state- ment of qualifications or proposal allows the agency to evaluate each competitor’s understanding of the QA challenges in the project. 10. Assigning the CMR the duties of scheduling for both design and construction during the preconstruction phase enhances collaboration between the parties. This service was rated as the second most valuable pre- construction service by both the case study agencies and contractors, and the ability to fast track was cited by 10 of the 15 papers shown in Table 1. 11. The agency can furnish a list of the cost categories to be used in preconstruction and where it wants various costs, such as fees and contingencies, to be accounted for in the CMR contract. Doing so eliminates confusion as to where each cost is to be allocated and facilitates the GMP negotiations. Table 29 provides a typical format for these costs. FURTHER RESEARCH One purpose of this synthesis was to identify gaps in the knowledge about CMR and indicate possible research to address them. Five major research needs are presented. • Development of agency understanding and knowledge of CMR versus DBB and design-build (DB). • Investigating effective use of in-house design assets in CMR project delivery. • Development of a guide for CMR preconstruction cost modeling. • Estimating CMR preconstruction services fees. • Documenting and administering the GMP in CMR.

89 Increased Department of Transportation Construction Manager-at-Risk Knowledge In the national survey, 15 respondents indicated that they did not know what CMR project delivery was. The answers given by the two that used “other” as their reason for not using CMR appear to indicate that the real number is 17 of 47 states, or 36% of the sample. Another 22 indicated that they do not have legal authority. Although the study cannot determine why the authority does not exist, because DB is permitted in some form in 48 states infers that a lack of knowledge about the potential benefits and the mechanics of CMR project delivery may play an important part. Therefore, a short-term project to develop a “Primer on CMR Project Delivery for DOTs” would furnish a means to quickly disseminate the necessary information to transportation managers to make an informed decision as to whether or not the effort necessary to obtain authority was worthwhile. Two reports reviewed in the synthesis (Doren et al. 2005 and Strang 2002) fur- nished pragmatic political motivations to pursue the proposed research. • “Quasi-public and government organizations predomi- nately use the DBB method, but clearly many have tried other methods and most would consider either CMR or DB to be the best-value alternatives. Changing the deliv- ery methods used, in the case of these organizations, will often require changing laws and politics, but that is happening too, because the public is best served when it gets the best value for its tax dollars . . . CM-at-risk will likely become the more dominant delivery method for this group as long as the experience is positive” (Doren et al. 2005; italics added). • “The great advantage to CM-at-Risk for most public owners is that their governing bodies accept it . . . The choice then for most public Owners is between CM-at- Risk and the traditional [DBB] system. If they do not want to use the traditional method because of past poor results, and are not bold enough to try DB, they are encouraged in this direction” (Strang 2002). Once the barrier to implementation resulting from agency confusion about CMR is addressed by this short-term project, a second project may be developed to furnish the quantitative information that will be needed to remove legislative barriers to implementing CMR project delivery by those agencies that choose to add it to their procurement toolbox. This project would compare CMR with DBB and DB project delivery for transportation in the same manner as Sanvido and Konchar (1996) did for the vertical sector. That study developed a robust set of project performance metrics and then applied them to a large sample of projects drawn from the three project delivery methods. To achieve the level of statistical significance nec- essary to present data that will be used to change legislation, the study will need to extend itself to include the horizontal portions of rail transit and airport experiences as well as the standard highway projects. The prime deliverable will be a comparative analysis of project performance by delivery method that can be reduced to a form that is suitable to inform the legislative and agency decision makers that will ultimately decide whether or not to authorize CMR project delivery for their jurisdictions. The two studies could be combined into one, with the CMR primer as an early deliverable. Incorporating In-House Design in Construction Manager-at-Risk Project In the literature, there was no information regarding the use of in-house design professionals to accomplish the design. Most models found assumed that the owner would outsource the design. UDOT does both depending on the project’s requirements. Most DOTs have a robust in-house design capacity and use it on DBB projects. The synthesis found that there appears to be no need to change the construction QA program to implement CMR project delivery; much of that work is done by DOT designers. Therefore, there appears to be no reason to not use these design professionals to complete the entire design as well. Finally, the anecdotal information found on the topic indicates that the major barrier to overcome is developing a method where state employees can be held to the same scheduling standards as a design consultant. The research would be based on the California DOT’s design-sequencing program, which appears to be the “next best” analog outside of UDOT’s actual experience. The study would reach out to the municipal engineering community, as well as look overseas for possible solutions that occur in Alliancing and Early Contractor Involvement programs in use in the United Kingdom, Australia, and New Zealand. To make sure that the entire cost of the process is determined, the research would specifically look at the cost of in-house design and perhaps develop a methodology to estimate its costs. Additionally, it would document the changes necessary to implement in-house design with respect to scheduling and design product turnaround. The major deliverable would be a guide for using in-house design on CMR contracts. Preconstruction Cost Modeling Ladino’s “cost model then design” is a new process to most DOTs. Although contractors have extensive experience with this process, agencies typically depend on their bid tabulation- based estimating systems to do estimate after design. This study found that robust cost modeling actually drives design decisions and facilitates the value engineering process. Addi- tionally, it provides a foundation for controlling scoop creep and assisting design engineers’ understanding of the impact of design assumptions such as factors of safety. Therefore, a standardized procedure for developing CMR cost models could also be extended to DBB projects during design as well. The proposed project would take the cost modeling theory developed for DB projects and adapt it to CMR projects. It

could build on the work done by the Washington State DOT’s Cost Estimate Validation Process, which has been found to be very effective for developing early estimates. Additionally, there could be an investigation of the use of the stochastic estimating methods recommended by FHWA (Life Cycle Cost Analysis in Pavement Design 1998) to apply Monte Carlo simulation to quantify the uncertainty of construction costs. The deliverable would be a guide for developing these models for CMR projects, along with a commercial spreadsheet tem- plate that could be adapted by DOTs for use in their CMR programs. Preconstruction Fee Estimating The final proposed research need could develop a rational methodology for estimating preconstruction service fees. The contractors in this study were asked how they estimate the cost of preconstruction and each had its own method. However, regardless of the method, all stated that they then looked at the number and altered it to a value that they thought would be acceptable to the owner. This business decision was more pronounced in projects that evaluated fees as part of the selection process. One of the case study agencies (MDOT) stipulates the preconstruction services fee in its solicitation and treats it much like a DB proposal stipend. Michigan and Oregon both stipulated a fixed percentage of costs as well. When the contractors that competed for these projects were asked their feelings on this feature, they offered no complaints. Another case study contractor that had worked for non-case study agencies with similar policies indicated that it preferred this policy because it focused the selection process on “quality and value” rather than money. Therefore, the notion of developing a methodology that could be used by public agencies to estimate a fair and rea- sonable preconstruction services fee would appear to not only be helpful but also welcomed by industry. Most DOTs have a system to estimate the cost of design services for outsourced engineering contracts that could be adjusted to serve this purpose if they had reliable input data to populate their cost database. Thus, the study would produce a parallel method and representative input data to allow this fee to be estimated. Additionally, the research could do a broad survey of the construction industry to confirm the impressions found in 90 this synthesis, as well as to identify the full suite of possible preconstruction services. The study might also include a pilot of the method with an experienced DOT such as Utah. Documenting and Administering Guaranteed Maximum Price in Construction Manager-at-Risk The synthesis did not collect information on how the various case study agencies documented and administered their GMPs. In addition, the literature was silent on that subject. Therefore, it is suggested that a synthesis be conducted on this topic. The synthesis would need to take a multi-modal scope to find sufficient CMR experience and uncover the various options for dealing with this issue. The study would look at not only how the GMP is configured and contractually promulgated but also on how payments are made to the CMR from this source. It could also review requirements to audit the contingencies. Finally, the synthesis could do a more detailed study of contingencies that extends the finding of this study. SUMMARY CMR project delivery is not the proverbial “silver bullet” any more than DB was. However, it furnishes an attractive option for public highway agencies to deliver their projects in a manner that is less adversarial and more constructive by involving the contractor during design. It also furnishes an option where the agency does not have to relinquish control of the details of design to be able to accelerate the schedule or receive the benefits of real-time cost-estimating data. This report found that CMR can be successfully implemented and that there are documented cost and time benefits with no degradation in quality. The barriers to implementation are primarily legislative and perceptional. A public agency that has an interest in adding this tool to their procurement tool box may investigate the possibility, communicate with those agencies such as UDOT that have experience, and begin to develop the policies and procedures that will permit their initial project to be successful. This synthesis report essen- tially finds that an agency that has the proper characteristics and selects a project where tangible benefits can be accrued will probably be successful.