Click for next page ( 82


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 81
81 ior TxDOT staff that evergreen contracts encourage the use of the highest growth in the country, with a population growth of the contractor on hand as a matter of convenience, as rate of 2.5 percent. (104) opposed to selecting the best contractor for a given job. It Utah's state highway system is composed of close to 41,000 is feared that expediency or practicality could endanger proj- centerline miles of highway roads. Of that, UDOT is responsi- ect quality. (101) ble for approximately 6,000 miles. (105) While state respon- Overall, a balance should be found between accelerating sibility accounts for roughly 14 percent of the total roadway project delivery and hampering competition. There is also miles, these roads carried roughly 67 percent of the total consensus that it is important to limit the use of evergreen VMT in 2007. (106) contracts to smaller projects, as they don't perform as well on Transportation infrastructure has not kept pace with the large and complex projects, which often require highly spe- booming population. From 19902007, Utah experienced cialized skill sets rather than the versatility required of firms a growth in VMT of 71 percent, but added only four percent selected for evergreen contracts. capacity to its highway system. In reaction to this pressure, UDOT has adopted a strategy with four goals: maintaining the existing system, improving system efficiency, improv- Conclusions ing safety, and building capacity. The department feels that Evergreen contracts are not a panacea for project accelera- building capacity is necessary to meet the needs of the popu- tion. Rather, these contracts are one tool that a state DOT can lation but notes that, because of financial constraints, high- keep in its acceleration toolbox. TxDOT has found that when ways cannot be built fast enough to keep pace with growing unforeseen design issues arise on larger projects, this form of congestion problems. (102) contracting can provide an accelerated solution. Utah is unique in terms of funding transportation projects TxDOT uses these contracts in conjunction with a wide in that the DOT relies heavily on state funding. While UDOT variety of financing and contracting methods. The use of mul- does take advantage of federal aid funding for qualifying proj- tiple tools and the organization's regional restructuring are ects, the department generally receives a significant portion emblematic of TxDOT's willingness to experiment. In these of its capacity expansion project funding through the state uncertain economic times, the department's past ability to legislature. That money is allocated from the state General adapt and find creative solutions to external forces portends Fund and fed into the Transportation Investment Fund, Crit- well for its future. ical Highway Needs Fund, or the Centennial Highway Fund. The Transportation Investment Fund is designated for maintenance, construction, and reconstruction. The major- Utah ity of the revenue for this fund is generated by the state sales tax, but it is also fed by legislative appropriations. Money Construction ManagerGeneral Contractor remaining from the latter two funds is also deposited here Utah Department of Transportation (UDOT) considers when their projects are completed with a surplus. The Centen- its charge to include a holistic approach to transportation in nial Highway Fund is an allocation of state and federal money the state. The department describes its mission as, "quality for use on 43 capacity-driven projects. Each of these projects transportation today, better transportation tomorrow," and is included in the Statewide Transportation Improvement recognizes a fundamental connection between the state's Plan (STIP). transportation system and the overall quality of life and eco- The Critical Needs Fund revenue comes from voluntary nomic prosperity. (102) contributions, legislative appropriations, and sales and use Utah is a large western state (13th in the nation ranked by tax. This fund is used for transportation projects around the area), with a projected population under three million. While state, as determined by UDOT, the state Transportation Utah features a very low overall population density, that pop- Committee, and the Executive Appropriations Committee. ulation is significantly city based, with 88 percent of its popu- This fund is generally used for capacity expansion projects that lation located in urban areas. The state's population is clustered are deemed high priority due to population growth. These largely around Salt Lake City along the Wasatch Front, with projects generally do not receive federal aid. some growth occurring in the St. George area. (103) The state has experienced tremendous population growth Organizational Structure over the past several decades. Between 1990 and 2007, Utah's population grew roughly 47 percent, with much of that growth UDOT's overall strategy has been to centralize some tech- concentrated in the urban areas. (102) During that time, Utah nical knowledgeintensive functions while decentralizing has consistently been one of the fastest growing states. For sev- functions that require local knowledge. This strategy has been eral years, including July 2007 to July 2008, Utah topped the list implemented under a matrix management model.

OCR for page 81
82 UDOT has divided the state into four regions (Figure 22); not all are advertised in the same year. The project manager Region 4 has been subdivided into three districts. Each region also administers contracts for consultant projects. (107) is assigned a region director. Project teams, as managed by the project manager, are com- The region director is at the top of the matrix model and posed of staff from any and all of the applicable position cate- oversees five position categories: gories listed previously. These teams are assigned on a project- by-project basis and are not static across all projects under a Administration specific manager. Project managers Program managers are centralized and assist project man- Pre-construction engineers agers with managing the region's transportation program from District engineers initial planning through design, construction, and project Materials engineers closeout (see Figure 23). Their responsibilities include secur- ing funding for projects and determining project funding pri- Project managers are regional staff members who are re- orities (with consultation from UDOT Planning and relevant sponsible for overseeing the scope, schedule, budget, and MPOs), and determining project priorities (with consulta- quality of a given project. They provide continuity of project tion from regional directors and project managers). They are knowledge throughout all phases of a project. Project man- charged with providing project managers with the resources agers oversee project-specific teams for each project and will they need to deliver projects. Program managers are also re- generally manage between 10 and 15 projects at a time, though sponsible for tracking program metrics. UDOT policy has been Source: UDOT Figure 22. UDOT regions and districts.

OCR for page 81
83 Figure 23. UDOT project delivery chart. to promote program managers from within UDOT, so that leadership can build projects quickly with all or a majority of each program manager has spent time as a region director. the required resources under their control. It is believed that The program manager position was created in the early this decentralization has assisted with project acceleration. 2000s. In UDOT's funding environment, where the depart- One possible downside suggested regarding this decentral- ment is reliant on state funding, the program manager is ization strategy is that when the central office wishes to pur- accountable to the state legislature. According to UDOT sue an innovative procedure, local staff may not be receptive. staff, the position has created greater accountability at high Resistance to innovations has stemmed from associated costs levels within UDOT. However, some feel that there has been as well as from the tendency of the local staff to perceive these a tendency for accountability to accumulate at those high lev- initiatives as intervention. els and diminish at lower levels. In this regard, it is felt that As a means of linking centralized and decentralized func- the creation of these positions has not had an appreciable tions, UDOT has implemented a proprietary software tool impact on the speed of project delivery. called Electronic Program Management (ePM) for scheduling, UDOT has sought a strategy to centralize functions re- tracking, and monitoring program management. The tool is quiring a high degree of technical knowledge. For instance, used by all staff involved in project delivery. The STIP is logged staff at UDOT headquarters in Salt Lake City performs into this system as well. Through sophisticated algorithms, the functions relating to structures, right-of-way, hydraulics program tracks and prioritizes critical steps in project delivery and geotechnical engineering for projects statewide. Func- and allocates staff time accordingly. Timesheets are logged into tions that require more local or on-site knowledge or the system and budgets are tracked. The tool brings a systems significant field time have been decentralized and are per- engineering approach to program management. formed by regional staff. Examples of these functions UDOT reports that the ePM system can be a valuable include project management, roadway design, utilities, and scheduling tool and in many instances does produce a stream- construction. lined, efficient project and program schedule. However, the High-level UDOT staff feel that decentralization of those system requires high-quality data and diligent upkeep through- functions is a strength of the department because the local out the project delivery process. UDOT has found that with a

OCR for page 81
84 large number of individuals using the system, faulty data inputs ing and thereby contributes to the definition of the method. are frequent and can hinder the efficacy of the tool. In UDOT's model, a contractor is selected as early as possible, At UDOT, perhaps because of the funding mechanisms in though the local FHWA office does not permit this selection place, the emphasis on project acceleration is strong. Access to take place until after the approved environmental review to legislators who allocate funds allows UDOT to reduce time documents are signed. (109) spent on programming. Political pressure from those same legislators creates an incentive to deliver projects quickly. The Model Initially, a design consultant is selected. Then, a construc- Best Practices for Utah tion contractor is selected. CMGC contracts are not awarded Background on a low-bid basis. Rather, the selection process incorporates technical merit and price into a best value formulation. Con- UDOT has demonstrated a willingness to experiment and tractors are evaluated on criteria ranging from design skills take on risk in virtually all facets of project delivery. Its three- to delivery schedule. Once a contractor is selected, that firm pronged procurement strategy exemplifies this willingness to appoints dedicated design staff to the project. UDOT then experiment. The department currently uses designbidbuild, works in cooperation with the design consultant and the designbuild, and construction managergeneral contractor contractor's designers to achieve the desired design. (CMGC) models. Over the past 20 years, the department has The design team works in a collaborative and iterative fash- shifted from primarily using the traditional designbidbuild ion. For example, UDOT will inquire what would be required procurement model to primarily (in dollar terms) using a for an accelerated schedule or to implement a given innova- designbuild model. Now, the department is attempting to tion. The contractor's design representatives provide feed- build on the strengths of designbuild by entering into an back regarding constructability and cost for suggested design agreement with the FHWA to implement and evaluate a pilot parameters. The design changes repeatedly, based on identi- CMGC procurement model. fied risks to the schedule or budget. Under this agreement, federal funding is authorized for When the design is finalized, the relationship with the 24 projects over a two-year period starting in 2006. Utah has contractor is severed. That contractor is then offered the first agreed to fund an additional 24 projects--6 projects in each opportunity to bid the construction. UDOT simultaneously of the UDOT's four regions--to test the model in geograph- hires an independent auditor/contractor (usually a seasoned ically diverse settings. UDOT has agreed to report to the individual retired from the industry) who also submits a blind FHWA on a variety of criteria, including budget analysis, bid. If the two figures are close and reasonable, the contractor schedule analysis, design, and constructability feedback. retains the job and enters a construction agreement. If the con- Though implementation of the model is somewhat limited, tractor's bid is not within 10 percent of the independent bid, there have been some preliminary findings that indicate that then the project becomes a traditional bidbuild project. (108) this model could be a useful tool for project acceleration and budget savings. (108) Implementation of the CMGC model is consistent with Early Findings UDOT's efforts to engage the construction and engineering In terms of project schedule, the CMGC model does appear industries. UDOT's efforts have included regular outreach to reduce the time frame for most projects. Involving a con- and consultation with industry representatives, such as the tractor in the design process creates efficiency in several ways. state's Association of General Contractors and the American First, during the design phase, having a contractor provide Council of Engineering Companies on issues such as RFP feedback in real-time allows for earlier identification of design language. The department has used input solicited through this errors and quicker problem solving. The transition from outreach to inform the contract formulation process and refine design to construction phases is also hastened and smoothed the contractor selection process. These groups are not involved as the contractor already has detailed knowledge of the design. in the process of writing specific contracts, but rather provide It has been found that there is less need for specificity and general guidance on what will make projects biddable. UDOT greater ease of communication between UDOT and the selected has made efforts to standardize RFP language based on this contractor. (108) input and successes from past projects. The UDOT experience has also shown that this model allows Outside of UDOT, the CMGC procurement method is not for a project to begin with a degree of risk that would be con- generally well understood and is not often employed. In some sidered intolerable under other procurement methods. In one locations, there are legal barriers to its use. UDOT's use of the instance, design began before ROW issues had been resolved, method is one of the largest applications of CMGC contract- and the two processes occurred in parallel. That project fin-

OCR for page 81
85 ished a year ahead of schedule by eliminating the time that Having the contractor involved in the design process also had originally been allotted for ROW acquisition. (108) opens the possibility for parallel phases. As long lead items are The CMGC model has been found to enhance working rela- identified, the contractor can begin procuring before the design tionships with external agencies. Utility and railroad compa- is complete. Additionally, if innovative methods are employed, nies have shown greater openness to solving ROW or relocation early involvement provides the contractor with additional time issues when a familiar contractor is performing the work. to learn and develop familiarity with the task. (108) This has again allowed for parallel processes in which these issues can be resolved alongside the design phase, thereby Lessons Learned decreasing the overall project schedule. UDOT found that though CMGC projects do benefit from shortened schedules, it is not the best contracting method to Change Orders use if there is a considerable time restraint and an abbreviated UDOT hoped that the CMGC model would reduce overall schedule is the principal driver. When schedule is the primary costs for a project. One of UDOT's primary goals for initiating driver, the project manager and the contractor tend to place the pilot program was to test if the model would reduce the pressure on the design consultant to speed the pace of design. number of change orders. The department had identified This pressure can increase the likelihood of design flaws going change orders related to design errors as a significant source unnoticed. Also, there have been experiences where early- of inefficiency with regard to both cost and project delays. phase contracts for materials have resulted in multiple change By involving the contractor in the design process, design errors orders and increased construction costs. In instances where were expected to be caught early and contractors expected to schedule is the primary driver, UDOT recommends using a absorb the risk associated with those errors. designbuild model. Early indications are that the model has indeed reduced the Project and program managers at UDOT believe that the number of change orders related to design errors. However, greatest gains in schedule under the CMGC model are derived UDOT frequently utilizes change orders to expend funds that from the ability to perform tasks on a parallel schedule. Cus- are either freed by negotiations or legislatively allocated. tomarily, design takes longer because of the iterative nature of UDOT employs change orders for this purpose on all projects, the process. Some acceleration occurs in construction due to regardless of contracting method. This procedure, in turn, efficiencies created by the contractor's familiarity with the complicates the analysis of the impact of the CMGC arrange- design. However, the greatest acceleration is achieved by the ment on change orders and overall budget. It requires analysis ability to perform parallel scheduling. One caveat is that par- on a project-by-project basis, which has not yet been com- allel schedules do increase risk; therefore, the model should pleted by the agency. (108) not be employed in a risk-adverse environment. UDOT found that an advantage of CMGC over design build is that it is simpler to select a contractor. Where design The Design Process build procurement can require a 500-page RFP, a CMGC RFP There has been consensus between program managers and can be as brief as 25 pages. By retaining control of the design Association of General Contractors representatives that this process in a CMGC contract, UDOT can greatly reduce the contracting method has resulted in reduced risk, improved need for specificity in the RFP language. For contractors, the design, and shortened project schedules. It is believed that cost of responding to a CMGC proposal request can be a tenth the better design stems from UDOT's active involvement in of the cost of responding to a designbuild RFP. the design process. In a designbuild model, the department Defining roles and responsibilities has proven to be an cedes design control to the contractor. In that process, the important task in CMGC projects, as many in the construc- contractor is motivated to maximize profits. That objective tion industry are unfamiliar with the model. UDOT has begun may run counter to accelerating the schedule, developing a partner training at the onset of a project to reinforce roles and high-quality design, or pursuing innovation. responsibilities. The department has found that one benefit In the CMGC model, the contractor is part of the design of the model is the spreading and sharing of risk. In the con- process insofar as providing feedback on constructability and ventional designbidbuild model, the department owns the cost, but UDOT staff guides the use of innovative techniques. risk; in designbuild, the contractor largely owns the risk; in The contractor advises on construction, and the design team CMGC, risk is distributed and shared, thus creating account- tailors the design to fit the contractor's strengths and abilities. ability for all the project delivery partners. While the contractor does not have the same kind of control The department also found that cost must be considered as in a designbuild contract, the design team still strives for a throughout the process. UDOT expressed concern that some design that the contractor will be able to construct deftly. (108) contractors may try to exploit the first bid opportunity for

OCR for page 81
86 construction. It is important for the contractor to know that demonstrate real project acceleration benefits, this contracting if it cannot deliver the project for a fair price, the project will method is likely to gain greater prevalence at UDOT. revert to a bidbuild model. Conclusions Future Developments UDOT does recommend CMGC as the primary delivery Recent months have seen great uncertainty arise in the process when schedule is not the primary driver. Its CMGC realm of transportation funding all across the United States. pilot projects seem to indicate that the contracting method States from coast to coast are finding significant funding short- can be an effective tool for accelerating projects. However, the falls as tax revenue drags with the overall economic slowdown. UDOT's efforts have been conducted in an environment Adding uncertainty in the coming year is the reauthorization where there is active engagement and cooperation between by the U.S. Congress of a federal transportation bill. This bill the DOT and the construction industry. UDOT's persistent has been speculated to feature significant alterations to fund- outreach to the state's Association of General Contractors ing formulas and mechanisms. may lay the groundwork for a unique mutual trust between At this time, the funding outlook for Utah is difficult to the department and participating contractors. In other envi- estimate. The state's heavy reliance on State funds for capac- ronments, where contractortransportation agency relation- ity building projects further clouds the ability to forecast. It is ships are less amicable, such a contracting method may prove unclear whether this model will fare better or worse than counterproductive. states that rely more heavily on federal funds. In the past few In addition to project acceleration, the CMGC contracting years, UDOT's efforts have been well funded, but the needs method has a number of co-benefits. Specifically, UDOT has continue to outpace resources. found the method is less intensive on the front end, with less According to senior UDOT staff, an estimated 70 percent of effort required to prepare an RFP and contract documents. transportation dollars are being spent in designbuild projects. Additionally, the model grants the agency significantly more That figure is likely due to the designbuild model's proven control in the design process and allows UDOT to pursue record of accelerating project schedules over the traditional innovations as it sees fit. This trait could help to drive inno- designbidbuild model. If the CMGC model continues to vation in Utah and around the country.