Development of a Sustainable Quality Assurance Program for Transportation Infrastructure

One of our nation’s most valuable assets is our roads and bridges, and its ability to move people and goods. The prosperity of our nation is dependent on the quality of our infrastructure and system, which is directly dependent on the condition of our highways. State and federal departments of transportation realize the importance of Quality Assurance (QA). From experience they have learned that non-conforming material or construction practices can result in premature failure of highway components. Major attention and resources have been devoted to the development of QA programs to address this concern. State Highway Agency’s (SHA) across America are faced with the challenge of addressing a deteriorating infrastructure system under constrained financial budgets, reduction in staffing levels, increasing public demand for better and faster construction of projects and the public scrutiny of how State funds are spent. Demands on state work forces have never been greater. With limited resources and ever-increasing demands for services, SHA’s are implementing new technologies and innovations for the purpose of improving and optimizing their QA programs under existing conditions and available resources. The objective of the dissertation is to provide SHA’s with recommendations for the development of an effective, efficient and sustainable QA program. 0428 To achieve the project objective a comprehensive literature review was conducted, with a focus on the ingredients in which SHA’s differ most, Quality Control (QC) and Acceptance. The purpose of the literature review was to determine the state of practice of SHA’s QC and Acceptance practices and policies. An evaluation into the use of Contractor Performed Quality Control (CPQC) test results to supplement agency Acceptance testing was performed. A detailed investigation was conducted on the use of consultant engineering testing services to supplement agency QA staffing. The cost effectiveness was evaluated through a cost analysis of RIDOT in-house acceptance testing versus consultant engineering testing services. The findings indicate that the evolution of QA programs which started back in the 1960’s is still very much ongoing today. The result is a large spectrum of QA programs, resulting in QA programs which differ significantly from one state agency to another. The area where QA programs differ most is in the QC and Acceptance arena. How SHA’s have delegated QC roles and responsibilities to the contractor significantly impact the overall efficiency, effectiveness and sustainability of a QA program. QC policies and requirements, including a QC plan requirement, should be consistently implemented, monitored and enforced on each and every project that a SHA puts out for bid. The contractor’s QC role and responsibilities cannot begin and end at the plant. It was found that the use of CPQC is essential to a successful QA program. The use of CPQC test results to supplement agency Acceptance testing will reduce the number of test that an agency must perform. The use of consultant engineering testing services to supplement agency QA staffing will allow SHA’s to meet peak workload demands more cost effectively. The recommendations derived from this dissertation can help SHA’s improve the efficiency, effectiveness and sustainability of its QA program.

policies. An evaluation into the use of Contractor Performed Quality Control (CPQC) test results to supplement agency Acceptance testing was performed. A detailed investigation was conducted on the use of consultant engineering testing services to supplement agency QA staffing. The cost effectiveness was evaluated through a cost analysis of RIDOT in-house acceptance testing versus consultant engineering testing services.
The findings indicate that the evolution of QA programs which started back in the 1960's is still very much ongoing today. The result is a large spectrum of QA programs, resulting in QA programs which differ significantly from one state agency to another. The area where QA programs differ most is in the QC and Acceptance arena. How SHA's have delegated QC roles and responsibilities to the contractor significantly impact the overall efficiency, effectiveness and sustainability of a QA program. QC policies and requirements, including a QC plan requirement, should be consistently implemented, monitored and enforced on each and every project that a SHA puts out for bid. The contractor's QC role and responsibilities cannot begin and end at the plant. It was found that the use of CPQC is essential to a successful QA program.
The use of CPQC test results to supplement agency Acceptance testing will reduce the number of test that an agency must perform. The use of consultant engineering testing services to supplement agency QA staffing will allow SHA's to meet peak workload demands more cost effectively. The recommendations derived from this dissertation can help SHA's improve the efficiency, effectiveness and sustainability of its QA program.   Table 2 Xbar-S Constants ……………………………………………………72 Table 3 RIDOT Acceptance Temperature Data………………………………80 Table 4 RIDOT Acceptance Slump Data ……………………………………80 Table 5 RIDOT Acceptance Air Data ……………………………………….81 Table 6 Contractor Quality Control Temperature Data ……………………...85 Table 7 Contractor Quality Control Slump Data …………………………… 85 Table 8 Contractor Quality Control Air Data ………………………………..85 Table 9 RIDOT Acceptance Air Test Data for F-Test and t-Test ……………86  Today, most SHAs are moving towards shifting the QC responsibility from the SHA (owner) to the contractor. Throughout this dissertation the term owner is used in place of SHA. The SHA agency is owner of the final product and as such, has and must maintain final say in the acceptance of the work performed by the contractor. How this transfer of QC from the owner to the contractor has evolved varies significantly from one SHA to another. Most SHAs are in agreement that QC is the contractor's responsibility but the development and implementation of QC requirements and policies have been a low-key effort by SHAs. It is for this very reason that contractor QC roles and responsibilities vary significantly from one SHA to another. Even more problematic is the finding that contractor QC roles and responsibilities vary from one project to another within the same SHA. This results in confusion and the intermingling of QC responsibilities between the SHA and the contractor.
SHAs across America are faced with the challenge of addressing a deteriorating infrastructure system under constrained financial budgets, increasing public demand for better and faster construction of projects and the public scrutiny of how State funds are spent. QA programs have a significant impact on the SHAs budget and overall reputation. A program not properly staff or managed will have a negative impact in the overall QA operating cost. In addition, a program not properly managed or staff will also likely result in inferior quality work. Since the agency must operate and maintain the completed project, inferior quality work will significantly impact the agencies future budgets as the work will need to be maintained or replaced sooner than the expected service life. When a SHA fails to deliver quality projects within an established budget there is a detrimental effect on future programs and loss of faith on the agencies abilities to provide quality work and manage public funds. For a QA program to be efficient and effective all three components, QC, Acceptance and IA, must be founded on clear and concise policies and the delegation of responsibilities must be understood by both the owner and the contractor. For a QA program to be sustainable it must be developed, implemented and managed in a cost effective manner. One of our nation's most valuable assets is our roads and bridges and its ability to moves goods and people. The success of our nation is dependent on the condition of our highway system, which is directly dependent on the quality of construction. With the administration and implementation of QC and Acceptance being left to the individual SHAs there is a need for information exchange on QC and Acceptance policies and procedures.

Summary
A major concern for SHAs has always been the actual quality of the work performed. SHAs have devoted major attention and resources to QC and QA activities to address this concern. It is the goal of this study, through the evaluation of existing SHA QA programs and through the synthesis of best practices derived from the literature review, to provide SHAs with recommendations that can be implemented to improve the efficiency, effectiveness and sustainability of a QA program. With these changes and developments, it is understandable, that one of the first issues encountered in this literature review was the lack of consensus regarding critical definitions. The words, terms and phrases used in QA programs and specifications are specialized vocabulary whose meaning and definition often differs from one SHA to another. This is a concern that was noted by S. Hughes where he wrote "One problem associated with QA programs and specifications since their inception have been differing interpretations of the specialized vocabulary used in these programs." (Hughes et al. 1999

Quality Assurance Elements
 Quality Assurance (QA). All those planned and systematic actions necessary to provide confidence that a product or facility will perform satisfactorily in service.
 Quality Control (QC). Also called process control. Those QA actions and considerations necessary to assess and adjust production and construction processes so as to control the level of quality being produced in the end product.      Type II error. Erroneous acceptance of the null hypothesis.

2-2 Status of Knowledge
The evolution that has taken place in QA programs has been driven by several factors. Two of the major factors were the AASHO Road Test and the construction of the Interstate Highway. The AASHO Road Test project revealed unexpectedly large variabilities in measured properties of highway construction materials and products. Methods projects, where the SHA has total control and contractor has little to no control to DBOM projects where the SHA has little to no control and the contractor has total control. This is the change in the philosophy of project management that is leading the QA evolution. This research has found that even though there is abundant evidence of an increase in contractor involvement in QA and overall project management there is still a lack of confidence among STAs and contractors. As with the transfer of QC responsibilities from the owner to the contractor many see when STAs are looking to develop efficient and sustainable QA Programs, the reluctance to transfer responsibilities and roles, due to lack of trust, still remains a major concern and obstacle. When STAs first started to delegate the QC role to the contractor the analogy of the Fox guarding the hen house was widespread. Today, the majority of STAs have delegated most QC responsibilities to the contractor. What was questionable and a major concern then is now standard practice. SHAs across the country are incorporating more contractor involvement in the design, construction, management, operation and maintenance of projects. The practice of a SHA hiring a contractor for the equipment and manpower to build a project is now evolving into more of a partnership between the SHA and a contractor. These types of partnership can result in better quality projects because it is to the advantage of both the agency and the contractor to design and build a quality product since both have stakes in the operation and maintenance of the final product. These changes do not come easy. With Method and Materials specification projects STAs have become accustomed to having total control over the manufacturing and placement of the materials and product, total QC control. Through this control STAs believed that they would achieve the required quality of the final products. It has now been tested, proven and accepted by STAs that it is the contractor, the producer and installer of a product or material that has the most control of the overall quality of the final product. by the contractor to assure quality of the material and final product.

2-3 Current QA Programs
To understand why so many variations of QA programs exist, specifically in QC and Acceptance, this research looked at how STAs addressed the many elements and factors that comprise a QA program. 23 CFR 637.207 sets forth the requirements within a STAs QA Program as follows: (a) Each State transportation department's (STD's) quality assurance program shall provide for an acceptance program and an independent assurance (IA) program consisting of the following: (1) Acceptance program.
(i) Each STDs acceptance program shall consist of the following: (A) Frequency guide schedules for verification sampling and testing which will give general guidance to personnel responsible for the program and allow adaptation to specific project conditions and needs.
(B) Identification of the specific location in the construction or production operation at which verification sampling and testing is to be accomplished.
(C) Identification of the specific attributes to be inspected which reflect the quality of the finished product.
(ii) Quality control sampling and testing results may be used as part of the acceptance decision provided that: (A) The sampling and testing has been performed by qualified laboratories and qualified sampling and testing personnel.
(B) The quality of the material has been validated by the verification sampling and testing. The verification testing shall be performed on samples that are taken independently of the quality control samples.
(C) The quality control sampling and testing is evaluated by an IA program.
(iii) If the results from the quality control sampling and testing are used in the acceptance program, the STD shall establish a dispute resolution system. The dispute resolution system shall address the resolution of discrepancies occurring between the verification sampling and testing and the quality control sampling and testing. The dispute resolution system may be administered entirely within the STD.
23 CFR 637 provides STA's with flexibility in the development of its QA Program.
The decisions that a STA must address when developing a QA program are many.
These decisions are made based on what works best for the individual STA QA program. For an understanding of the differences in QA programs from one STA to another, the following are a small sample of the decisions agencies must consider in the development of its QA program:


The development of frequency guide schedules for verification sampling and testing,


Type of specifications to be used.
 What attributes will be used for QC and acceptance.
 What types of test will be used.


Who will perform the testing?
 What will be the acceptance /rejection criteria, will there be pass or fail or pay adjustment factor.
 What are the qualification requirements for QC and Acceptance inspectors?
Qualified or certified?
 How much of the QC role will be given to the contractor.

-Types of QA Programs for Aggregate Base and Subbase
Of the 45 respondents, 15 of the STAs QA consist of materials and methods provisions, 14 of the STA's use QA programs with the agency controlling quality and performing acceptance, 21 use QA programs with the contractor controlling the quality and the agency performing the acceptance, and 10 use QA programs with the contractor controlling the quality and the agency using contractor test results in the acceptance decision ( Figure 5). As with soils and embankments the variability of the material lends itself to the use of materials and methods type of QA. The attributes most often used for QC and Acceptance for aggregate base and subbase are gradation (15), compaction (20), and moisture content (14)

-Types of QA Programs for HMA
Responses from the Questionnaire clearly showed that the majority of STAs QA programs use statistically based specifications. Only 2 out of the 45 STAs that responded still use materials and methods provisions for HMA. Materials and methods specifications are rarely used for HMA. Figure 7 shows that out of the 45 STA that responded 25 reported using a QA program with the contractor performing QC and the agency using contractor QC test results for Acceptance. That is more than 50% of the STAs using contractor QC test results for Acceptance. properties of HMA and the fact that HMA statistically based specifications have been developed and in use longer by more agencies than for any other material. There is more confidence in the validation of contractor QC test results as a result. As more and more STAs move in the direction of performance based specifications and end result specifications, it is expected that the use of contractor QC test results for Acceptance will also increase. This is a sign of the evolution that is currently taking place in the development of STAs QA programs. The importance of testing must be based more on the selection of attributes to test and type of test to perform rather than who performs the actual testing. Once STAs develop confidence in the validation of contractor QC test results, the use of contractor QC test results for Acceptance will be as common as is the practice of contractor performed quality control (CPQC) is today.

-Types of QA Programs for PCC Paving (PCCP)
The use of statistically based QA programs with PCCP is at the early stages with most STAs but its use has increased in the past decade. Studies conducted by Chamberlin for the Performance-Related Specifications for Highway Construction and Rehabilitation indicate that "the use of performance-related specifications for PCCP is on the increase and appear to be ahead of the use of this type specification for HMA" (Chamberlin, W.P., 1995). STAs have reported an increase in PCCP for intersection improvements and construction. As seen in figure 9 of the 40 STA that responded 16 STAs use QA programs with Contractor QC and Agency Acceptance and 13 STAs use

Types of QA Programs for PCC Structures
QA programs for PCC structures are very similar to the QA programs for PCC paving.
The use of statistically based specifications is in the increase but is currently not used as often as with HMA. The survey responses showed that of the 43 responses, 25 agencies use material and methods provisions ( Figure 11). The attributes most commonly used for QC and Acceptance are gradation, slump, air content, cylinder strength and water/cement (W/C) ratio ( Figure 12). The purpose of QC is to monitor quality characteristics of the product so that adjustments can be made as needed to assure that the process is producing material that meets or exceeds specification requirements. To assure that the process is in "control". Selecting the proper product characteristics to test and developing a testing frequency for testing/monitoring that will allow for timely implementation of corrective actions is the foundation of an efficient and effective QC process. A concern regarding the responses from this properties such as air content, slump and gradation are appropriate QC quality characteristics. Each of these characteristics can be tested, monitored and modified while the product is in production. Therefore these are good material properties for QC testing because they provide information that can be used to make corrections to the concrete mix while in the production and placement stage. The use of cylinder break strength tests as a QC attribute does not fit into this QC definition. Cylinder break strength test are conducted at 3, 7 or 28 days after the placement. The information learned from test conducted after the placement cannot be used to monitor or adjust the mix during production. While concrete cylinder break strength tests are beneficial for the development/testing of mix designs or for payment calculations, it is not a good QC attribute.

-Independent Assurance
The third component of a QA program is Independent Assurance (IA). The survey questionnaire sought to determine how the IA unit in each agency is organized and what its function is. There are two ways in which IA is used by STAs. The first is the narrower context which is, to provide an independent assessment of QC and Acceptance test results. The seconder, broader view, is one where IA performs an assessment of the overall QC and Acceptance process. Responses from the questionnaire and through phone calls with STA Officials supports that IA is being conducted by most STAs in compliance with 23 CFR 637; however, the manner in which IA is organized within an agency varies greatly, as does the level of staffing, even when normalized by agency budgets. Figure 13  the survey used the agency's construction and maintenance budget as a normalizing factor. The results can be seen in Figure 14 which show that staffing varies from 0.5 IA full time employees per hundred million dollars to 16 IA fulltime employees per hundred million dollars. agencies of 39 responding (44%) indicated that they contracted some QA testing outside of their workforce" (Smith R.G 1998). In less than one decade the percentage of agencies that utilize outside consultants within their QA programs has nearly doubled. Table 1 represents the wide range of products tested by outside consultant services for STAs.  Figure 15 shows the volume of CEI consultant work estimated as the percentage of total highway funding per state transportation agency. As depicted in Figure 15 the CEI budgets varied from zero to 35.5%. RIDOT reported a 6.55% budget use on outside consultant services. Figure   California DOT  For manufactured materials; when the department implements a materials management system, the department will no longer perform QA on a projectby-project basis, but will release material on a manufacturer's track record.
 Implement requirement for contractors to develop a QC plan with minimum acceptable frequency and observations including identification of a quality manager. Department QA will be "Did they follow the plan?" and perform statistically valid random sampling and separate tests.  The contractor test results will be used in the acceptance decision with agency verification testing at a reduced rate.

-Summary of Literature Review
The results from this literature review confirm that there is a broad spectrum of QA programs being implemented by STAs across the country. Every STA has tailored their QA program according to the agency's needs, goals and available resources. There is not a "one-size-fit -all" QA program. With the many variables involved in the development of a QA program this broad spectrum is to be expected. The concrete is then tested before it leaves the plant. On a project with CPQC there will be a contractor QC inspector or the consultant inspector hired by the contractor on site with slump cones, air meters and cylinders to test the concrete before it is placed.
In addition the QC inspector will assure that the ground where the material will be placed has been compacted to required density levels. On a project where there is no CPQC there is no contractor QC inspector on site. It is the STA Construction Inspector or Materials Inspector that will conduct slump test or run an air meter test on the mix. A scheduled Acceptance Test often turns into QC testing as required adjustments are implemented into the mix. These tests are typically classified as "Informational Only Testing". The information derived from these test is whether the product is in compliance with specifications. This is QC testing that must be performed by the contractor. It is the contractor that must monitor his production process so that necessary modification can be implemented in a timely fashion to insure control of the production process.
SHAs have been and continue to mix and mingle QC roles, from one material to another and from one project to another. This is the confusion that has arisen with

3-2-1 72 Inch Pipe Piles in Place of H-Piles
There were many innovations incorporated into this project. Piers #4, #5 &#6, located in the channel and deepest section of the Sakonnet River were designed on friction and end bearing 72-inch steel pipe piles. The "forced plugging " of these 72-inch pipe piles was achieved through the incorporation of a 2½ inch thick, steel plate welded inside the pipe, 40 feet from driving end of the pipe.
The plate contained a 14-inch diameter hole to relieve soil and water pressure. Figure   19 is a representation of this steel plate used to seal the driving end of the pipe pile.
The purpose of this steel plate is to incorporate end-bearing capacity to the pipe.
Without this plate, the bearing capacity of the 72-inch steel pipe pile will the bearing capacity achieved through friction. During the test pile project RIC 2007-CB-006, it was determined that a combination of end bearing and friction would significantly reduce the depths that the piles would need to be driven. In

Partnering
Partnering was implemented in the Sakonnet River Bridge Project as a pilot program. Partnering was a practice that neither the agency nor the contractor had much experience in. Partnering works as follows:  The contractor and the agency agree on and select a third party person/consultant to run the partnering sessions.
 All cost associated with partnering are shared equally between the agency and the contractor.  Daily meetings between the contractor's superintendent and the project Resident Engineer were agreed upon to discuss the days scheduled operations.


Weekly schedule meetings were conducted to evaluate project schedule, upcoming major events, actions required by the contractor and actions required by the agency.


The importance of teamwork was institutionalized on every SHA and contractor individual involved in the project.

Transfer of QC (CPQC)
One of the most significant innovations incorporated into this project was the as with other innovations such as Design Build Projects, there was the hurdle of relinquishing control and concern of loss in overall quality. There are several concerns associated with the reluctance to relinquish QC control to the contractor.
First is the lack of confidence in the contractor's test results. Opponents of the transfer QC to the contractor often used the analogy of the "Fox Guarding the Hen House".
The validation of CPQC test results is the key to the successful implementation of CPQC and the implementation of the use of CPQC test results for Acceptance.
Another factor is the misconception that the owner can control QC better than the contractor can. Time and experience has proven that it is the contractor, the producer and installer of the product, with the most control of the quality of the final product. It is from this understanding and acceptance that most STAs are shifting the QC role to the contractor. On the contractor's side, there was the concern of how would a QC program be implemented and monitored and at what cost. Most contractors were also aware of the fact that they were not equipped or staffed to perform overall project QC.

QC Plan
The key to the successful implementation of CPQC is the QC Plan. RIDOT understood this concept and therefore included in its Request for Proposal (RFP) the contractual requirement that the contractor must submit a QC Plan to the Department for review and acceptance prior to the start of any work. The objective of a QC plan is to measure and monitor those material properties characteristic's that impact the quality of production, thereby enabling timely corrective actions to prevent the incorporation of non-conforming material into the project. Agencies can specify the QC plan requirements in one of two approaches. One is for the agency to stipulate the minimum QC requirements and properties that the QC plan must contain. The other is to specify all the requirements and properties that require testing. The disadvantage to the first is that you may only get a plan that meets the minimum QC requirements.
There are two disadvantages to the second approach, first by stating all the testing requirements the contractor may view the QC plan as the agency's plan rather than the contractors plan. This can have great liability issues. The second disadvantage is that being too prescriptive opens the door for claims when something was inadvertently omitted. RIDOT selected the first approach. To insure that a proper overall project wide QC Plan was established, it was designated as a shop drawing submittal which required Departmental approval before any work could commence on the project.
This allowed the contractor to design and construct a QC plan that incorporated the testing and inspection of activities that the contractor specified to produce acceptable material. There were many iterations of the contractor's QC plan but after several submissions and revisions, the Plan was approved. A good Quality Control Plan will clearly indicate the use of random testing and event/ process testing. Both are essential to the effectiveness of a QA plan since both have very different functions.
The purpose of random testing is to evaluate the quality of the total population through random sampling. The purpose of event/ process testing is to assure the quality of the This was the first project where the Resident Engineer had contractor QC inspectors on site with a QC plan to monitor and perform field material testing. Figure 20 shows the contractor QC organizational chart f or this project.

Observed Increase in QC Field Testing
At the very start of construction operations, RIDOT field personnel noted a change in field inspections. A significant increase in the amount of testing was taking place throughout the project. testing. RIDOT, as with the majority of SHAs report that QC is the contractor's responsibility and that Materials Inspectors are there to perform Acceptance Testing when in reality, they are also performing QC testing. This is evident since contractor QC personnel are not on site on most RIDOT projects where CPQC is not a contractual requirement. The RIDOT Materials Section has a limited staff and Construction Inspectors are often required to oversee more than one operation.
Acceptance testing can be viewed as "spot checking" to assure that the contractors product is in conformance with specifications requirements. The frequency of Acceptance Testing is determined and established in accordance with RIDOT's Master Schedule of Testing Manual (RIDOT 2010). The Master Schedule of Testing Manual (MST) contains a template that specifies the material type, test type, test description (included frequency of testing) and method of testing. For example, In the MST template you will find that for the use of a material classified as "Fill Gravel Borrow" the tests required are "One 50 lbs. sample per 1000 CY or less for gradation testing" and "One (1) field density test per 1,000 CY or less." Specifications for a typical sidewalk construction operation require a 5-foot wide, 12 inch depth gravel base. One thousand CY would allow you to construct 5400 linear feet of sidewalk base. By MST requirements, only one RIDOT Acceptance test is required per 1000 CY or less. Without CPQC that equates to over a mile of sidewalk construction with only one field inspection, the Acceptance test. This is not a fault of RIDOT acceptance testing procedures. Acceptance testing is not QC testing. The purpose of Acceptance testing to provide assurance that the materials and workmanship incorporated into every highway construction project are in close conformity with the requirements of the plans and specification. The frequency of testing for any material is a direct relationship to the level of risk that the STA is willing to take. RIDOT Materials Inspectors and Construction Inspectors were not accustomed to seeing actual CPQC field testing taking place on a project. The fact that QC testing frequency far exceeds the frequency for Acceptance testing is because each serves a different purpose. The purpose of QC testing is to make sure that the process is in control. QC frequency of testing must allow the contractor the opportunity to incorporate changes, on a timely fashion, to assure that the product complies with specifications requirements. Acceptance testing is a spot check to assure that the contractors QC is serving its purpose, to assure material compliance with specifications and for use in payment. When the Contractor hired ATC to take on the QC role the contractor assigned all liabilities associated with the quality of the product to ATC. All material found to be unacceptable by our Material Acceptance Inspectors would require removal and or be subjected to a significant pay reduction factor. Any cost associated with unacceptable material was now the contractors QC consultants responsibility. The transfer of responsibility from the Contractor to the subcontractor increased the level of risk to the subcontractor. To reduce his level of risk the subcontractor established a QC plan that incorporated a testing frequency that provides a level of confidence that the material incorporated into the project will meet all specification requirements. It is important to note that the frequency of testing developed by the materials testing subcontractor far exceeded the frequency of testing in the original QC plan submitted by the Contractor. Simply stated, without CPQC there is no contractor QC in the field. The only testing taking place is that performed by the RIDOT. If RIDOT Materials Inspectors are responsible for only Acceptance testing, then where is the QC? CPQC places QC responsibilities on the contractor where they rightfully belong. The significant amount of QC testing that took place at the Sakonnet River Bridge Project is proof that CPQC will result in overall better quality and more productive and efficient use of RIDOT materials personnel.

Communication Problems
Several issues arose associated with the use of CPQC. The first problem incidents. This problem was resolved by meeting with the Contractor and ATC. The QC Plan notification requirement was clarified and brought to everyone's attention. In addition, two very simple practices were incorporated; one was the exchange of cell phone numbers between RIDOT's crew and the ATC crew. Second, it was mutually agreed that due to the constant changes that occur on a day-to-day basis, RIDOT Inspection staff would meet with the ATC inspection staff, at the start of every workday. Communication between RIDOT and ATC improved immensely. RIDOT and ATC both understood and agreed that they had to work closely together as a team to make CPQC work.

Non Compliance Test Report Submissions
Another problem encountered was ATC's non-compliance with the timely submission of QC test results to RIDOT. The QC plan was very specific on time requirements for the submission of QC test Results. This was the contractors own plan and it was the Contractor that established the time requirements. ATC was not getting test results to RIDOT in accordance with the Plan. This requirement was especially critical on this project because some of the existing soils throughout the site contained contaminates. The degree of contaminants varied significantly from one location on the site to another.
It was clearly stated in the contract that no soils shall leave the site prior to testing, classification and approval from RIDOT. The goal was to use as much of the on-site material as possible. To comply with this contractual requirement, the contractor had to establish stockpiled areas for all material excavated.

Concerns with Contractor QC Test Results
One of the most significant issues that arose from the implementation of CPQC was the concern with the validity of the contractor's QC test results. As previously noted, RIDOT field personnel noted a change in field inspections. A significant increase in the amount of testing was taking place throughout the project. At first, the increase in testing conducted by ATC provided RIDOT with a sense of assurance. As the level of operations increased so did the amount of ATC was an opportunity to show a SHA that outside consultant inspection services can be used to perform project QA testing. It was a win-win situation for the owner and the contractor. Figure 21 is a field photo of RIDOT and ATC testing personnel performing test on a concrete mix.

Control Charts, F-Test, t-Test
The data collected for performing simple comparison of ATC test results with RIDOT's Acceptance test results was later used to perform statistical analysis testing.     Figure 27 represent the Xbar-R and Xbar-S Control Charts for RIDOT Acceptance Test results. Figures 28 through Figure 33 represent the Xbar-R and Xbar-S Control Charts for the contractors QC test results. Tables 3 through Table   5 represent RIDOTs test result data used to develop Acceptance Control Charts. Tables 6 through Table 8  A sample size of 80 RIDOT Acceptance air percent test results and 80 contractor QC air percent test results were used to conduct the testing. Table 9 represents RIDOT Acceptance air test data and Table 10 represents the contractors QC air test data. With a sample size of n=80 normal distribution was assume in accordance with the central limit theorem. Table 11 represents the summary of RIDOT Acceptance test results and Contractors QC test results for the overall database established. Table 12 represents the results of the Two-Sample F-Test for Variance. All three test indicate that at a significance level of alpha (α) =0.05 the results are not statically significant.
We therefore fail to reject the null hypothesis that ợ (QA air)/ ợ (QC air) = 1.            both mean that you need to reject the null hypothesis.    To prevent confusion and the intermingling of QC roles and responsibilities QC policies must be clear, concise and consistent from one project to the next.

RIDOT In-House Acceptance Testing verses Consultant Acceptance Testing
Cost Analysis

4-1 RIDOT's Current QC and Acceptance Practices and Policies
The objective of this dissertation is to provide SHAs with recommendations on how to improve the efficiency and sustainability of its QA program. QC policies, practices and requirements that differ from one project to the next only create confusion to the contractor and to the state inspectors. Regardless on the type of operation or size of the project, QC needs to be contractually delegated to the contractor through policies that the contractor can follow and state inspectors can monitor. Every project from the building of a bridge to a small pavement marking operation must require a contractor QC Plan submission. It is this QC Plan that will tell RIDOT how the contractor will assure QC, both in the plant and in the field.

4-2 Implementation of CPQC Test Results for Acceptance Testing
SHAs are responsible for Acceptance Testing and IA. The policies and practices associated with these two components of RIDOT's QA Program have been evaluated to determine their efficiency and cost effectiveness. RIDOT's IA and Acceptance Inspectors consist of experienced, qualified and certified Inspectors who perform all IA and Acceptance Testing responsibilities. There is a clear and distinct separation between IA and Acceptance roles and responsibilities. There are two ways in which IA is used by STAs. The first is the narrower context which is, to provide an independent assessment of QC and Acceptance test results. The seconder, broader view, is one where IA performs an assessment of the overall QC and Acceptance process. RIDOT's IA section performs a hybrid of these approaches. It provides an independent assessment of QC and Acceptance test results while providing an assessment of the overall QC and Acceptance process. RIDOT will need to consider this alternative as it faces increase workloads with reduced staff and budget constraints.

4-3 Implementation of Outside Consultant Testing Services
Many STAs are implementing outside consultant inspection services to supplement their QA programs. Responses from the NCHRP Synthesis 346 survey questionnaire reported 78% of the responding STA's implement the use of consultants within their QA programs. Agencies implement consultant inspection services for a variety of reasons, including:  The ability to increase or decrease staffing levels to meet current workload.
STA funding for infrastructure work varies from year to year. Therefore the number of projects that an agency puts out for construction also varies from year to year. No agency can afford to staff to handle peak workload periods.
An agency that staffs to handle peak periods will find itself paying for to get a project out on time or to meet a peak demand period, from an administrative viewpoint, a consultant can hire more help much faster than a government agency can negotiate a contract or hire more staff.


To bring innovation and quality. The goal of every STA is to serve the public.
The consultant's goal is to survive in a competitive environment and make a profit. To do so the consultant must remain strong, lean and on the cutting edge.
Consultants compete to stay in business. They need to be the best in their game. To do so they hire and fire and keep only the very best. SHAs operate in very significantly different manner. A SHA decisions cannot simply be based on cost. A SHA is primary goal is to serve the public. A SHAs hiring and retaining of employees practices is also significantly different from that of the private sector. RIDOT hires full time employees with pre-established pay level step increments. One common complaint often heard from State workers is they have reached their top step pay grade level. This typically occurs when an employee has been with the State over ten (10) years. Once an employee reaches this pay grade level the only pay increases that the employee will see are those increases that all state workers receive when their union contracts are renegotiated every 2 to 3 years. This type of system does not provide any motivation or incentive for an employee to excel at his position. Consultant employees know that their employment is directly related to their abilities, skills, qualifications, dedication to the company and their wiliness to work.
Both the consultant inspection service and its inspectors want to bring innovation and quality to a project to assure future work from the Department of Transportation. If they cannot perform the work at a competitive price and deliver quality work, they will not survive in the business.
There Most STA's cannot accurately establish a project overhead cost. Another fault of this approach it that it does not take into account any long-term cost associated with performing the work in-house or outsourcing the work.
The second approach, referred to as the "Life Cycle " approach takes the "Direct Cost" approach one-step further and includes long-term cost in the analysis.
The "Life Cycle" approach is a more logical approach to use when performing a cost effectiveness analysis between a STA and an outside consultant because the STA cost associated with labor, equipment and overhead continue to accrue for as long as these resources remain with the STA. With outside consultants, once the work for which they were hired for is completed, all consultant cost end. There is no long-term cost with outside consulting. There are many difficulties in conducting an accurate cost effectiveness analysis between a STA and a private consultant service. The following are major difficulties and challenges reported from case studies and encountered in the development of a cost effectiveness analysis for the RIDOT: Therefore, we cannot accurately identify consultant supervision costs" (Renfrow 1992

Section 4-4 In-House verses Outside Consultant Testing Services Cost Analysis
The findings of this dissertation indicate that neither the "Direct" nor the "Life Cycle" approaches can account all the factors that affect the outcome in a cost analysis of outsourcing verses performing the work in-house. The research also supports the fact that the accuracy of the data used in either approach will significantly affect the outcome. To conduct a cost analysis of RIDOT's in-house Acceptance Testing verses consultant engineering testing services the "Life Cycle "approach was selected because it includes long-term cost in the analysis. A cost analysis between a public agency, such as a STA and a private business, such as a consultant engineering testing service must take into consideration long-term cost. To assure the accuracy and validity of the data used for this analysis the data will be received directly from the     RIDOT Tech III and Tech IV average base salaries will be converted into an average hourly rate for the purpose of performing a RIDOT hourly rate comparison to the average consultant's hourly average rate. To convert the average annual salary to an hourly rate I will divide the annual salary by 2080 (52 week per year * 40 hour work week=2080 hours/year). Table 20 represents the average Tech III and Tech IV RIDOT Material Inspectors total annual cost converted to an hourly rate.   Testing and improve the efficient and sustainability for the overall QA Program.
To keep this cost analysis on the conservative end the average base salary used in these calculations does not include overtime. As shown in Table 16   Assurance Procedures for Construction, SHAs were given more flexibility in designing their QA programs, including permitting the use of CPQC test results for Acceptance. The findings from this dissertation confirm that SHA's QA programs consist of the three main ingredients; QC, Acceptance Testing and Independent Assurance (IA). It is how these three ingredients have been blended that accounts for the differences in QA programs from one SHA to another. There is not a boiler plate or a one-size-fits -all QA program. Transportation agencies have developed QA programs that have been customized to meet their individual States needs and available resources. The importance of QA became evident as a result of the AASHO Road Test conducted in Ottawa, Illinois from 1956 to 1960. Today, over 50 years since the Road Test, the strategies and practices used by SHAs to ensure quality and to meet 23 CRF637 requirements encompass a wide variety of approaches which has resulted in a broad spectrum of QA programs.

Quality Control (QC)
For a QA program to be efficient, effective and sustainable all three components; QC, Acceptance and IA must be functional. Of the three components, IA is the component where SHA are most consistent on regarding the IA roles and responsibilities. The majority of SHAs assume the IA role and responsibilities in its entirety. In general, the literature review supports that IA is being conducted by most SHAs in compliance with 23 CFR 637.
QC is the component that differs mostly from one SHA to another. The reason for this is that QC is a component that requires contractor involvement. There is little disagreement that QC is the contractor's responsibility. However, how QC is delegated to the contractor is where SHAs differ most. It is widely accepted that QC should be the contractor's responsibility. It is the contractor that manufactures the product and it is the contractor that constructs the product. It is therefore the contractor with the most ability to control the QC process for both the manufacturing and the construction of the product. Where the disagreement exists is in the delegation of QC responsibilities to the contractor. The literature review clearly shows that SHAs are still assuming QC responsibilities, mainly so when it comes to field QC. The contractors QC responsibilities cannot begin and end at the contractor plant. Both field QC and plant QC are essential component of the overall QC process and both are the responsibility of the contractor. If the contractor does not have designated QC personnel in the field to perform the required field QC testing then these tests are routinely being performed by agency material inspectors under the umbrella of "Acceptance Testing". The delegation of QC responsibilities to the contractor differs from one SHA to another for many reasons including low keyed efforts from STAs in the development and implementation of QC policies, reluctance of the agency to relinquish control and lack of trust between the agency and the contractor.
The transfer of QC from the SHA to the contractor differs significantly from one SHA to another and in many agencies from one project to another. As a result QC responsibilities and roles are not clearly designated. This often results in the intermingling of QC responsibilities between the agency and the contractor. How QC is delegated to the contractor significantly impacts the agencies overall QA program.

RIDOT first incorporated Contractor Performed Quality Control (CPQC) back in 2008
for the Sakonnet River Bridge Replacement Project. The contractual language for this project was very clear and concise as to the delegation of QC responsibilities to the contractor. CPQC for this project resulted in a significant increase in field QC testing never witness before by the Department resulting in better quality project wide.
CPQC was a success for this pilot project and continues to be used on RIDOT major projects, such as; the Pawtucket River Bridge 550 Project and the Providence Viaduct Bridge Project At the time of this study the RIDOT has not incorporated CPQC on all of its projects. On projects where CPQC is not a contractual requirement the contractors QC roles generally start and end at the contractor's production plant.
In my 28 years as an Engineer for the RIDOT it was common practice to witness RIDOT Materials Inspectors performing testing that far exceeded those required by the RIDOT Master Schedule of Testing Manual. It is through this additional testing that RIDOT Materials Inspectors assure the quality of work on RIDOT projects. This additional testing comes at a cost to the Department and a strain on the Materials Section staffing.

CPQC Test Results for Acceptance Testing
At the time of this project the RIDOT Materials Sections retains and performs the entire acceptance function. As stated above, on projects where CPQC is not required the QC roles and responsibilities are intermingled and the Materials Acceptance Inspectors are burden with the task of performing field QC testing.
Before a SHA can implement the use of CPQC test results to supplement Acceptance Testing, the SHA must assure that the contractor has been designated all QC responsibilities. When a contractor is performing project wide QC the SHA will then realize an increase of field testing by contractor QC personnel. This increase in contractor QC testing, as a result of CPQC, will provide SHAs a valuable database resource of CPQC test results and agency Acceptance Test results. Through statistical analysis of these databases a SHA will be able to develop validation procedures for CPQC test results. Once the Department has achieved a level of degree of confidence with its validation procedures, then CPQC test results can be used to supplement agency Acceptance Testing. Every contractor quality control test result that can be validated and used to supplement agency acceptance testing equates to one less Acceptance Test that the agency needs to perform. The use of CPQC test results to supplement agency Acceptance Testing will improve the overall efficiency and cost effectiveness of a SHAs QA program.

Consultant Testing Services for Acceptance Testing
RIDOT Materials Section is experiencing a reduction in staffing level as a result of retirement and or employees leaving to the private sector. Efficient and cost effective staffing with full-time employees is a difficult task to accomplish. By nature, road and bridge construction work is extremely affected by weather conditions.
Here in the New England area, the peak work period is between the months of May and October. SHAs cannot afford to staff to meet peak workload demands. A SHA that staffs to handle peak workloads, it will find itself paying for inspectors when there is no work taking place. With funding for infrastructure projects varying from one year to the next, full time staffing becomes an even more difficult. SHAs are looking for alternative cost effective management strategies to meet construction inspection and material testing staffing needs. The use of outside consultant engineering testing firms is a solution. The use of consultant testing services provides the ability to increase or reduce staffing levels to meet current workloads (peak and low periods), bring expertize and quality to the project and reduce the current 12% Materials Inspectors overtime rate.
A life cycle cost analysis was conducted in this dissertation comparing the cost of RIDOT In-house Acceptance Testing verses outside consultant Acceptance Testing.
The outcome of the cost analysis showed that it is less expensive and more cost effective to use outside consultant testing services than in-house RIDOT Acceptance Materials Inspectors.
RIDOT will need to make staffing decisions in the very near future. SHAs need to evaluate whether supplementing its QA staff with consultant engineering testing services provides a more cost effective solution to its staffing requirements.
Contractors have benefitted from the use of consultant services since the early 70's.
Contractors have become extremely proficient at managing consultant services by optimizing their use while minimizing overall QC cost. SHAs can use the consultant engineering testing services as a "Force Multiplier" during the peak workloads periods. The use of consultant services can allow SHAs to optimize its current resources and provide a cost effect way of meeting peak workload demands without over staffing. As SHAs experience future reduction in full time personnel the agency should implement the cost analysis evaluation process to determine what is the most cost effective solution is for the Department, hiring additional full time personnel or using additional consultant engineering testing services to meet current workload demands.

1.
Clear, concise and consistent QC policies should be incorporated into every SHA project. QC policies, practices and requirements that differ from one project to another create confusion and often result in the intermingling of QC responsibilities between the Department and the contractor. This intermingling of QC responsibilities has a negative impact in quality, diminishes the overall effectiveness of QA Program and increases QA cost therefore affecting the sustainability of the overall QA Program.

Every transportation project should require the submittal of a Quality Control
Plan before the start of any work. A Quality Control Plan describes what actions the contractor will take to meet RIDOT QC requirements and policies.
It is therefore the best tool that RIDOT field inspectors have to monitor and enforce QC requirements.