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.
31 TYPICAL CONTRACTING ISSUES Various contracting procedures are used to procure pavement condition data. Some procurement comes through advertised contracts, requests for proposals (RFPs), and requests for qualifications (RFQs), which might lead to negotiated con- tracts. In addition, warranties are used by a few agencies. The various approaches, as well as some of the pricing in- formation developed, are evaluated and summarized in this chapter. In regard to the synthesis questionnaire, 30 agencies pro- vided at least some responses relating to contractual issues, and 9 agencies provided copies of typical contract documents. Some of those documents are contracts, whereas some are RFPs and others are requests for expressions of interest, etc. A list of agencies submitting contract documents is given in Table 11 (questionnaire responses on contracts are summa- rized in Table B6 in Appendix B). The most popular basis for a contract award was an RFP, which was used by 18 agencies. Seven agencies use an RFQ approach and eight use advertised contracts and a low-bidder approach [some of the Canadian provinces use a Terms of Reference (TOR), similar to an RFP]. Several agencies make use of more than one approach. On the other hand, the California DOT (Caltrans) reports in its questionnaire response (see Table B8), âCaltrans has seen a higher quality product for IRI collection by doing the work ourselves.â Typical contract items range from imaging to distress identification and quantification, roughness, rut depths, and joint faulting. In some cases, where image collection and dis- tress reduction from those images are contracted, collection and processing are two separate items. In other cases, the two are combined. By far the most popular approach, cited by 17 agencies, is to contract for sensor-measured data items (IRI, rut depth, and joint fault), whereas surface distress data are collected either under a separate contract or in-house by agency personnel, usually manually. Some contracts contain separate data management items, usually for hardware and software, and for maintenance of those items. Such items typically appear when the agency does separate QA work in the form of an acceptance process applied to the vendorâs deliverables. In such cases, the agency may contract for workstations for internal review and use of the deliverables. The most popular items appearing in data collection con- tracts are summarized in Table 12. Although some of the tabulated items appear in other tables, they are repeated here as contract items for reader convenience. Not reflected in Table 12 is that some agencies rely on a combination of agency and vendor collection, usually depending on the high- way system under evaluation. Several agencies contract for only sensor-measured data. In comparison, Quebec contracts only for distress data reduction from images collected by the agency. Although used in collecting pavement-related data, the most popular peripheral item collected by contract is right-of- way images, contracted for by 14 agencies. In addition, a few agencies contract for signage or drainage structure inventories, or for shoulder images. A typical contract period is 2 years, although some agen- cies use a 1-year period and one state has awarded a 4-year contract. Several agencies provide for up to 5 years of negoti- ated extensions of shorter-term original contracts. Six agencies have warranty provisions in their contracts, whereas only one requires a performance bond to be posted. A total of 22 agencies have QA provisions, and 12 have price adjustment clauses. QA provisions are discussed in chapter six. Typical price adjustment clauses relate to delivery dates and accrue in the form of penalties for late delivery. No agency mentions paying a bonus for early delivery. AGENCY EXAMPLES Nine agencies provided documents relating to contracted data collection, although it is not possible to address every- thing in detail in this synthesis. However, several agencies have been judged as fairly typical, but with at least some unusual features. These experiences are summarized here. Louisiana Department of Transportation and Development In 2001, the Louisiana Department of Transportation and Development (LADOTD) awarded a contract to the Road- ware Group, Inc., for 32,000 lane-km (20,000 lane-mi) of comprehensive data collection, including GPS, digital right- CHAPTER FIVE CONTRACTING ISSUES
32 delivered per month. The contract did not address the reason for this stipulation, but it is presumed to relate to the time required for LADOTD evaluation of deliverables. Liquidated damages of $100 per day were to be assessed for each day the data for the required number of districts were not delivered on time and $500 per day for each day the data for all nine dis- tricts were not delivered on time. Finally, damages of $300 per day were to be assessed for each day that the final report was not on time. The Louisiana contract is ongoing, and no assessment of its success or failure has been released. The LADOTD esti- mated the cost of delivered data at approximately $34 per km ($54 per mi) and did not note, in the questionnaire response, any special problems with the contract. The LADOTD, with its experiences with this contract and related issues, has been chosen as one of the case studies for this synthesis, and the contract will be discussed in greater detail in chapter eight. Ontario Ministry of Transportation Ontario released a TOR (or RFQ) for contracted pavement roughness measurements in April 2001 (59). The project was to cover the years 2001 and 2002 and apply to the deter- mination of IRI values on some 18 000 km (11,000 mi) of pavement. Before bidding, prospective vendors were required to con- duct precontract qualification tests on a 12-site calibration cir- Agency Type of Document Contract Items Alberta Ministry of Transportation British Columbia Ministry of Transportation Louisiana Department of Transportation and Development Manitoba Department of Transportation and Government Services Mississippi Department of Transportation Oklahoma Department of Transportation Ontario Ministry of Transportation Quebec Ministry of Transport Vermont Agency of Transportation Terms of reference Request for proposal Contract Contract Contract Contract Request for quotations Terms of reference Contract IRI, rutting Cracking, IRI, rutting Cracking, IRI, rutting, joint faulting Cracking, IRI, rutting, joint faulting Cracking, IRI, rutting, joint faulting Cracking, IRI, rutting, joint faulting IRI Cracking data reduction only Cracking, IRI, rutting, forward video Notes: IRI = International Roughness Index. Agency Type Pavement Images Distress from Images IRI Rut Depth Right-of- Way Images State Province FHWA Total 16 2 2 20 15 3 2 20 18 3 2 23 17 2 2 21 11 2 1 14 Notes: IRI = International Roughness Index. TABLE 11 AGENCIES SUBMITTING PAVEMENT DATA COLLECTION CONTRACT DOCUMENTS TABLE 12 POPULAR CONTRACT ITEMS (Number of Agencies) of-way and pavement images, IRI, faulting and rutting mea- surements, and pavement distress evaluation (58). Required peripheral data included signage and signal inventories. Con- tract provisions addressed the vehicle and test equipment configuration and delivery and payment schedules, as well as QA procedures and acceptance criteria. A workstation for review of images and validation of distress data was to be provided at an LADOTD office. The contract called for the completion of four major tasks: ⢠Preliminary activities, including training of raters and workstation delivery; ⢠Collection of clear digital pavement images and profile data for each district; ⢠Distress quantification for all roads tested; and ⢠Final documentation of the project. Completed pavement condition data were to be delivered on a district-by-district basis with the provision that data for no less than one district or more than two districts were to be
33 cuit. These tests were to be conducted with the exact same ASTM Class II profiling equipment as proposed for produc- tion testing, with each calibration site to be tested at least three times. The vendor calibration tests were required to meet two criteria. First, the IRI values determined must have the same rank ordering as Ministry tests on the same sites. Second, the statistical coefficient of determination (R2) between the ven- dorsâ IRI values and those determined by the Ministry was expected to exceed 0.85. Once the contract was awarded, the vendor was required to again conduct the aforementioned calibration tests and to meet the same criteria. Similar calibrations are required midway through each testing year, at the end of each testing year, and at the end of the 2-year contract period. The agency offers this explanation: The final calibration is required to ensure year-to-year consis- tency of the survey data. If the Vendor will choose to bid for the similar survey work in the Year 2003, and if the Vendor will pro- pose to use the same IRI-measuring device or devices as those which the Vendor used for the 2001 and 2002 survey, the final calibration may be accepted by the Ministry as both the final cal- ibration for 2002 work and the pre-contract qualification cali- bration for 2003 work (59). In addition to calibration testing, the vendor is required to submit and adhere to a Quality Control Plan using 30 monitoring sites throughout testing. That plan is described in chapter six. Delivery dates by district are spelled out in the RFQ, as are substantial penalties for failure to meet delivery dates. The penalty varies by deliverable; that is, it is lower for deliver- ables due early in the contract than for those due toward the end of the contract. Also, being 1 day late is penalized for the same amount as 1 month (5% to 10% of the project price for the year, depending on the deliverable) and increasing by either 5% or 10% for each month or part thereof, depending on the deliverable. The agency provided no cost per kilometer or mile information in its response to the questionnaire. Again, there is no real indication of how matters have tran- spired with the contract resulting from this RFQ; however, the Ministry indicated that penalties for failure to meet specified deadlines for deliverables have been invoked. Quebec Ministry of Transport The Quebec Ministry of Transport awarded a contract to Path- ways Services, Inc., solely for the collection and analysis of surface distress (cracking) on 13 000 km (8,100 mi) of road- way in April 2002 (60). An unusual feature of that contract is in the preliminary work. On signing the contract, a first lot of approximately 500 km (300 mi) of roadway was sent to the contractor for analysis. That analysis must be completed and the result submitted to the Ministry within 1 month and must be approved before the contractor can proceed with the remainder of the roadways in the contract. The Ministry also has a minimum production rate require- ment of 300 km (190 mi) of roadway per week, with a penalty equivalent to 5% of the contract unit rate applied for every 2 weeks of delay for each kilometer not submitted. For QA, the Ministry selects from 2% to 5% of kilometers rated for analysis and applies a penalty depending on the percentage of the production accepted. This QA feature is discussed further in the next chapter. The Ministry reported that most contrac- tual problems can be solved with human intervention and application of the fairly elaborate QC plan. Vermont Agency of Transportation In June 2000, the Vermont Agency of Transportation (VAOT) entered into a contract with Roadware Group, Inc., to collect pavement condition data for 4 years (61). Some 3800 lane-km (2,400 lane-mi) of state and urban highways were to be evalu- ated in the years 2000 and 2002, and some 1000 lane-km (640 lane-mi) of the Interstate were to be evaluated in the years 2001 and 2003. Pavement data to be collected included fatigue cracks, transverse cracks, block cracks, rutting, IRI, and texture (on the Interstate only). Protocols specified were ASTM E950 for IRI and VAOT for the other items, including a 5-sensor rut measurement. Contract prices amounted to roughly $34 per km ($55 per mi) for Interstate, $22 per km ($35 per mi) for state, and $106 per km ($170 per mi) for urban highways. Roughness data are verified by checking five separate sites each year selected by VAOT. Each site is run three times, and the results are reported separately for evaluation by VAOT. No specifics of this evaluation are given. The collection of cracking data is controlled through the use of five precollection sites used for the calibration of WiseCrax data reduction algorithms. In production, 3% to 5% of the net- work is randomly sampled by VAOT staff and compared with vendor-delivered data. The acceptance process is unique and summarized in Table 13. Note that the table provides for the reporting of invalid data or for retesting, depending on the amount of work considered and on the degree of invalidity encountered in the random sample evaluation. No definition of invalidity is given in contract documents. Payment for each year of the contract is based on the number of lane-miles actually tested. Twenty-five percent is paid at the completion of field testing, 25% upon delivery of the data, and 50% upon acceptance of the data. The agency reserves 30 days for review and verification of the data. SUMMARY Clearly, there are numerous approaches to contract pavement data collection in use by the various highway agencies in
North America. Because there are a relatively small num- ber of vendors, it logically follows that a more standard- ized approach could result in economic benefits. Although a âone type fits allâ is not a practical or even desirable approach to contract development and execution, it is evi- dent that there could be a great deal more standardization than currently occurs. The most obvious area in which standardization might reasonably occur is in the protocols applied. If vendors did not have to prepare for different protocols in going from agency to agency, they might eas- ily pass on some economic gains to their customers. Effec- tive protocols would address data collection methods as well as QA and acceptance plans. 34 Some of the major issues one would want to address in a pavement data collection contract are as follows: ⢠Pavement system definition, that is, what is to be eval- uated; ⢠Data items to be delivered and the delivery format; ⢠Start-up pilot test requirements; ⢠Location-referencing requirements; ⢠Delivery time schedules; ⢠QA features including QC, calibration requirements, and acceptance criteria for each deliverable; and ⢠Payment schedules, including any liquidated damages and how they would be assessed. Amount Tested Data Quality Action One 0.1 mi Section Ten or More Contiguous 0.1 mi Sections Node to Node, >1 mi and <5 mi Node to Node 5 mi >50% Invalid >25% Invalid >75% Invalid >25% Invalid >75% Invalid >25% Invalid >50% Invalid Report reason for invalid data Report reason for invalid data Retest sections Report reason for invalid data Retest from node to node Report reason for invalid data Retest from node to node TABLE 13 VERMONT CRITERIA FOR REPORTING INVALID DATA OR RETESTING OF SECTIONS
