Cover Image

Not for Sale



View/Hide Left Panel
Click for next page ( 44


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 43
33 of the contract were significantly below the performance stan- tives, authorizing legislation that promotes publicprivate dards in the contract, the contractor had between 3 months and partnerships, downsizing, and a fixed level of staff faced 12 months to bring the assets up to the required standards. with rapidly growing maintenance needs. Other reasons include a focus on customer-oriented outcomes, potential Because the first contract proved to be successful, Monte- cost savings, the possibility of shifting more risks onto the video pursued two additional contracts, one involving gravel contractor, and achieving increased predictability of future roads (Zietlow 2005b). maintenance expenditures. Shortly thereafter, other Latin American countries, such The other evolutionary path consists of the steps federal as Brazil, Chile, Colombia, Ecuador, Guatemala, Mexico, agencies have taken to implement performance-based con- and Peru, also started adopting or planned to adopt a perfor- tracting. In response to a variety of legislation and regula- mance-based approach. tory changes, the amount of performance-based contracting has increased rapidly among federal agencies, including Chad the U.S. Department of Defense (DOD) and the U.S. DOT. Indeed, the Procurement Executives Council and the Office Chad's first performance-based road management and main- of Acquisition Management of the U.S. Department of Com- tenance contract occurred in 2001 with the assistance of the merce set a goal that by 2005 at least 50% of all eligible World Bank. The 4-year contract addressed 440 km of unpaved federal contracts be performance-based service contracts roads. A French firm was the contractor and used mainly local (Hyman 2003; "Increasing the Use of Performance-Based labor. A Cameroon engineering firm provided contractor over- Service Acquisition" 2005). sight. Elements of the contract included the following: The remainder of this section focuses on the experiences Management and maintenance as well as self-monitor- of state DOTs with PBMC and also addresses performance- ing of the contractor based warranty contracts. Rehabilitation over the first 21 months of the contract Rebuilding or replacement of drainage structures and Virginia signs Emergency help for those in accidents In 1995, the legislature of the Commonwealth of Virginia Rain erosion and axle load control; and enacted the Public and Private Transportation Act. The law Other emergency works as needed. stated that a private entity could submit a proposal to any responsible entity within the state, including the Virginia The contractor received a fixed monthly fee based on a DOT, to design, construct, finance, and operate facilities for lump-sum for the contract period. However, the contractor any mode of transportation. The public entity was required was paid on the basis of unit price for emergency work. A to evaluate and decide whether to accept the proposal after performance guarantee was equal to 10% of the contract it posted a description of the proposal for public comment. value. The contractor received an advance payment of 20% Under the legislation, VDOT also could solicit public to perform the rehabilitation. The total cost for road man- private partnerships (Virginia General Assembly 1995). agement, rehabilitation, and maintenance was estimated at US $5,740/km. Thanks to the performance-based contract, Not long after the enactment of the Public and Private the road was upgraded to excellent condition. Users appreci- Transportation Act, a construction and maintenance firm ated that the road was kept in good condition after the reha- submitted an unsolicited proposal to VDOT to maintain bilitation was completed. Cars and trucks were able to use parts of I-95 and I-81 and all of I-77 and I-381, totaling 251 the road during the rainy season, which was not possible miles (404 km), representing 20% of Virginia's Interstate before the roadwork occurred. A negative side-effect of the highways. In December 1996, VDOT awarded the firm a improved road quality was that the accident rate increased performance-based contract to maintain this portion of as travel speeds increased. As of 2004, no other significant Virginia's Interstate system. The value of the contract was difficulties were reported and the contract was judged to be $131.6 million and the initial term of the contract was for five a success (Zietlow 2005b). and one-half years, with an option to extend the contract for another five years. DOMESTIC EXPERIENCE This was a lump-sum contract. There were no deductions for failing to meet performance targets and no liquidated Within the United States, two evolutionary paths have damages expressed in terms of LOS. resulted in increased performance-based contracting. The first is growing use of PBMC by state DOTs. The motivation The contractor paid a company that specialized in asset for adopting PBMC includes governor and legislative direc- management to develop a maintenance management system

OCR for page 43
34 that would monitor achievement of performance targets and soundness of the analysis could not be verified ("Review of serve other purposes such as issue work orders. Although VDOT's Administration ..." 2001). Another VDOT study the contractor monitored its own performance, VDOT concluded that the contractor provided the interstate mainte- conducted an annual audit and issued a report card noting nance for 25% less than the state could (Hall 2005). Also, the achievement of the performance targets. contractor prepared an estimate of savings in terms of cost per mile: $22,400 for the contractor compared with $29,500 The contractor was responsible for nearly all types of for VDOT (Hall 2005). Virginia Tech ultimately found that assets and features from fence to fence, including repair and cost savings ranged from $6.5 million to $22 million, the rehabilitation of pavements, structures, and tunnels; drain- result of a sensitivity analysis based on varying the over- age such as culverts and ditches; roadside vegetation man- head rate of the contractor relative to VDOT, after escalating agement; and guardrail, signs, and fences. In addition, the costs according to a price index. Virginia Tech's approach contractor was responsible for snow and ice control as well as began with comparing bid tabs, because the contractor sub- incident response (Segal et al. 2003; Stankevich et al. 2006). contracted 80% of the interstate maintenance work (De La Garza and Vorster 2000). JLARC acknowledged that the The contractor was able to achieve the LOS targets for Virginia Tech study might shed light on the cost-effective- more than 90% of the items rated. In 2001, VDOT renewed ness of this contract. However, JLARC expressed concern the contract for five years. Figure 6 presents the annual about the narrow scope of the study and believed that the report card for VDOT's performance-based contract. The findings might not prove conclusive ("Review of VDOT's evaluation was performed by in independent third party. The Administration ..." 2001). report card shows that the contractor received a grade of A for shoulders, roadside, and drainage-related maintenance Washington, D.C. on all mainline sections but received a couple of B's and a C regarding traffic. In contrast, control sites that VDOT main- In 1998, in association with FHWA, the District of Columbia tained generally received B's and C's. Department of Public Works awarded a 5-year $69 million FIGURE 6 2005 report card for VDOT performance-based maintenance contractor (Source : Bryant 2007). VDOT projected cost savings of $23 million before it performance-based contract to a contractor for maintenance embarked on its first round of PBMC on Virginia's Inter- of 75 miles of the NHS within the District. This was the state highways (Segal et al. 2003). According to the Joint first performance-based contract of its type in an urban area Legislative Audit Review Committee (JLARC) of Virginia's of the United States. Contract selection was based on best General Assembly, VDOT's cost savings were based pri- value. Payments to the contractor included incentives and marily on estimates and projected costs that were compared disincentives, which depended on achievement of perfor- with future contractor payments. JLARC staff stated that mance standards. The contract covered the following (Rob- documentation did not exist to support the savings and the inson and Raynault 2005):

OCR for page 43
35 Pavement structures The objective of the project was to significantly improve Drainage asset condition and maintenance operations on the NHS. Roadside Cost savings was not an objective in light of the infusion Traffic safety such as guardrail and attenuators of funds required to significantly improve the LOS. Indeed, Roadside cleaning none of the published sources examined in this study offered Traffic safety evidence of cost savings regarding the District of Columbia's Bridges experience with PBMC. Curb and gutter Tunnels Texas Roadside vegetation Snow and ice control The Texas DOT (TxDOT) initially performed most of its Pavement markings maintenance work using in-house staff. Its outsourcing con- Traffic signs tracts involved just a limited number of single maintenance Highway lighting activities. However, over time, the legislature enacted laws Miscellaneous (e.g., weigh-in-motion and pedestrian requiring TxDOT to outsource an increasing amount of main- bridges). tenance. A 1989 bill required TxDOT to contract 25% of its routine maintenance, provided that the practice was cost-ef- Five LOS were defined for roughly 170 maintenance fective. In 1991, the Texas legislature raised the requirement elements grouped in various categories. Performance for maintenance contracting and directed TxDOT to increase measures reflected both LOS for assets and operations maintenance contracting to 50% by 1996, once again pro- and response times to address needed work. Performance viding that outsourcing was cost-effective. The legislature targets or standards were set between Levels 3 and 4 and required TxDOT to set targets for each of the districts to transformed to a scale of 0 to 100. The contractor was ensure compliance was met by 1996. As a consequence of responsible for monitoring its own performance daily this legislative direction, TxDOT embarked on two total based on its QC plan. maintenance contracts and developed a performance-based contracting program for its rest areas (Graff 2001). Each month an independent third party, along with District and contractor staff, inspected each maintenance Total Maintenance Contracts element. The independent third party provided for each ele- ment a rating of poor, fair, or good. Then a composite score In 1999, TxDOT entered into two performance-based main- was calculated. This complementary grading system was tenance contracts unprecedented in their size, scope, and risk. developed to facilitate communication with stakeholders These contracts involved sections of Interstate highway with regarding the outcomes that were being achieved. Poor per- some of the heaviest traffic in the state. The first involved formance received a 0, fair performance received a 50, and 120 miles of I-35 in the Waco District and the second was 60 good performance received 100. If the contractor equaled or miles of I-20 in the Dallas District. exceeded all performance standards, then its score would be 100 (Robinson and Raynault 2005). The impetus for these contracts was a presentation on the VDOT asset management contracts at the Texas Qual- At the end of each year, an objective, comprehensive eval- ity Initiative held in 1998. Partnering was a key to suc- uation was conducted by the independent third party. The cessfully initiating these total maintenance projects. The contractor was able to earn a variable award fee for excellent collaborative effort of the department and potential bidders performance. was essential to calm fears about this novel endeavor and establish the performance measures and standards (mini- A public policy institute wrote that, from the time the mum LOS) acceptable to all parties. Performance stan- District entered into its contract with the contractor, the dards were developed for every type of maintenance and District observed significant improvements in the condi- operations on these two sections of the Interstate. Selected tion of the road assets on the portion of the NHS maintained types of assets and operations that were addressed included under the performance-based contract. After the first year, the following: performance rose from the high 20s to the low 80s (out of 100). Based on input provided by a field engineer in FHWA's Pavements Office of Asset Management, the public policy institute Bridges observed that this improvement was attributable in part to Roadsides the specialization achieved by subcontracting to smaller Traffic operations companies or companies that the contractor created for an Traffic services area of maintenance (Segal et al. 2003). Incident response

OCR for page 43
36 Hazardous materials cleanup The total maintenance contracts were awarded based Emergency repairs. on the low bid. Payment was made in accordance with the contract payment schedule that set out monthly payments Contractor responsibility included coordinating with in terms of a percentage of the contract award. The period local government and law enforcement officials. Contrac- of performance was 5 years with a possible extension of 3 tors were also required to process damage claims and obtain years. Monthly payments were calculated by multiplying the disaster reimbursement from federal agencies. winning contractor's lump-sum bid by the monthly payment schedule percentage and making any deductions dictated by To make before-and-after comparisons of performance of contractor performance. contractors and in-house staff, TxDOT developed a Main- tenance Assessment Program. A LOS rating system simi- In addition to the benefits of lower bid costs than antici- lar to the process described in the NCHRP Web Document pated, TxDOT found the following additional advantages to 8: Highway Maintenance Quality Assurance (Smith et al. its total maintenance contracts: less inspection was required; 1997) was the key to the performance assessment process. less documentation of quantities applied was needed because This process so impressed upper management that it asked of the lump-sum nature of the payment, and the contractor for an evaluation of the LOS of the whole Interstate system. was encouraged to be innovative because it was not tied to The following is a description of the Texas Maintenance method specifications (Graff 2001). Assessment Program rating system: What were the outcomes in terms of service levels of the Random 1.6-km (1-mi.) sections are rated every 16 km (10 mi.). Bridges are rated in a separate evaluation. Each total maintenance contracts? LOS initially declined on I-20 element (for example rutting, failures, striping, signing, and I-35 and then started to rise (see Figure 8 in chapter 4) mowing, litter) is rated at each location on a scale of 5 to (Graff 2007). The contractor provided higher-quality snow 1, where 5 is excellent--new or like new; 4 is good--no and ice control than the agency previously did (A Guide for work needed; 3 is fair--minimum acceptable condition; 2 is poor; needs work; and 1 is failed. A 1 to 100 score is Methods and Procedures in Contract Maintenance 2002). determined....Each element is given a priority multiplier Although a Washington State DOT white paper on outsourc- depending on its relative importance based on the ing cited newspaper articles about poor contractor response following priorities: safety, protect the investment, user to icy roads (Ribreau n.d.), the articles did not acknowledge comfort, and esthetics. The elements are then combined by multiplying the resulting scores in each component, and that these were the worst ice storms in years. The contractor dividing by the maximum possible score. The components deployed its personnel and equipment; however, in numerous are combined to give an overall score (Graff 2001). instances, equipment became stuck in traffic and could not get to the areas needing treatment (Segal and Montague 2004). TxDOT developed detailed specifications, such as perfor- mance measures and standards, with the assistance of a large Rest Areas number of stakeholders, including headquarter maintenance personnel, districts, and potential bidders. This part of the As the year 2000 approached, Texas found that the condi- process was essential to achieve consensus and ensure that tion of its 740 picnic areas and 102 rest areas had seriously the procurement documents would attract bidders. eroded. The condition of buildings, grounds, pavement, water, and wastewater systems had deteriorated. To reverse TxDOT was concerned with a variety of risks and needed this deterioration, TxDOT decided to let and enter into four to develop a strategy to mitigate each one: 2-year performance-based contracts in each quadrant of the state to upgrade and maintain the rest areas. Performance- Calming employee fears. Fears were calmed through based maintenance contracts were valued at $6 to $8 million good open communication and the professionalism of each. Upgrades involved new construction, reconstruction, the potential contractors. major renovations, and possible conversion of some sites to Cost and insufficient competition. TxDOT believed truck parking. that contractors for the PBMC could not provide ser- vices for lower costs than state forces and the tradi- To measure performance, TxDOT developed rating guides tional smaller contractors that had done much work that provided pictures of acceptable and unacceptable condi- in the past. However, the successful contractor's bid tions for every component of these facilities. Each contractor was below the department's estimate. Also, the suc- was required to submit an enhancement plan explaining the cessful contractor subcontracted much of the work to repairs or improvements necessary to bring each component smaller contractors that historically did maintenance up to an acceptable level or better. TxDOT established an eval- for TxDOT. uation process, a rating system, and a combination of incentive Emergency response. The acquisition team put special and disincentive payments to ensure that conditions improved emphasis on performance standards concerning emer- and goals and standards were met. TxDOT conducts formal, gency response. unannounced inspections to keep subjectivity to a minimum.

OCR for page 43
37 Rating scores can range from 0% to 100%. At the begin- Mowing ning of the program, rest areas had scores that averaged Signs 73% and ranged from 15% to 99%. TxDOT established a Guardrail goal of increasing the average score across the state to 85%. Pavement striping Facilities with lower scores failed and scores of 85 or above Replacement of raised markers passed. For each day that a contractor scored more than Fence repair 92%, it received a 15% incentive payment of the normal Shoulder maintenance daily pay. The incentive was paid until the facility's score fell Cleaning drainage systems below 92% in another evaluation. Incentives and disincen- Environmental compliance tives were based on a rest area's overall score. Contractors Incident response that scored 84.49 or lower received deductions in daily pay Natural disaster preparedness according to declining thresholds. Failure to meet the rest Inspection of bridges area maintenance standards for two consecutive evaluation Highway lighting periods could result in an additional fine of $5,000 per day. Motorist aid service patrols. After the first year of these contracts, TxDOT had paid Contractors strive to achieve performance targets that are incentives and assessed disincentives and deductions of integral to a Maintenance Rating Program (MRP). Perfor- nearly an identical amount of about $246,000. Average mance measures are grouped by five elements: pavement, statewide ratings of facility conditions increased from 73% roadside, traffic services, drainage, and vegetation and before the performance-based maintenance contracts to 91% aesthetics. Each element includes unique characteristics; at the end of the first year (Sims 2004). for example, the roadside element is composed of unpaved shoulder, front slope, slope pavement, sidewalk, and fence. Florida Each characteristic is measured against a set of standards The Florida DOT (FDOT) has an extensive program of asset outlined in the MRP handbook. The standards for certain management involving PBMC. This program is termed characteristics vary by facility type: (1) rural limited access "Asset Maintenance." FDOT entered into 22 Asset Main- roads, (2) rural arterial roads, (3) urban limited access roads, tenance contracts from July 2000 through November 2005. and (4) urban arterial roads. Three times a year, each cost These contracts have a total value of $672 million, an aver- center evaluates 30 sample points for each of the 4 facil- age of nearly $100 million per year (Florida Department of ity types. These evaluation results are used to calculate the Transportation 2006). annual MRP score; however, FDOT is moving more toward performing MRP evaluations on the contractor, rather than According to a description of FDOT's asset manage- allowing the contractors to self-evaluate. FDOT has set to ment program, PBMC includes planning, programming, achieve MRP ratings of 70 for each characteristic, 75 for each administration, management, performance, and inspection element, and 80 overall. Failure to achieve the performance of routine maintenance involving 6- to 10-year lump-sum standards results in reductions in monthly payments. contracts. Criteria for contract selection give 40% to techni- cal considerations and 60% to price. FDOT has asserted that it has realized performance improvements and cost savings by transferring the responsibil- FDOT uses four types of asset maintenance-based ity for maintenance, daily management, and inspection of work contracts: to contractors (Florida Department of Transportation 2005). Corridor contracts focused on limited access highways; Based on extensive timeseries data on the state's MRP Geographic (e.g., regional) contracts with multiple published on FDOT's website (Florida Department of Trans- types of transportation facilities; portation 2000/20012006/2007), it is evident that FDOT's Facility contracts focused, for example, on rest areas or in-house staff and contractors, including those doing PBMC, weigh stations; and have achieved service levels that generally exceed minimum Fixed and movable bridges. performance standards in each maintenance area and on the four types of facilities rated. Performance-based contracts cover these maintenance activities, as well as specific contracts focused on such ele- FDOT indicated that it intended to further increase its ments as aesthetics or drainage. Asset Maintenance program expenditures in 2009 to cover 40% of the entire maintenance program. The balance of the Among other assets and activities, performance-based program will be 40% traditional contracts and 20% in-house maintenance contracts typically address the following contracts. Looking to the future, FDOT offers the following (Florida Department of Transportation 2005): perspectives on possible expansion of asset maintenance: