Procedures and Guidelines for Validating Contractor Test Data (2020)

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.

5 Use of contractor construction materials test results in acceptance decisions of state highway agencies’ (SHAs) projects dates back to the 1970s (1). This acceptance decision is typically coupled with a validation process. Validation is defined as, “The mathematical comparison of two indepen- dently obtained sets of data (e.g., SHA acceptance data and contractor data). For example, using statistical tests such as F- and t-tests to determine whether it can be assumed the data sets came from the same population. (Validation of test results includes the verification test results.)” (2). Title 23 Code of Federal Regulations (CFR) Part 637 Subpart B (23 CFR 637B) permits the use of contractor test data for construction materials acceptance, as long as SHAs validate the contractor data with independent test results (3, 4). A combination of hypothesis tests is commonly used to statistically determine if contractor and SHA test results are from the same population. Implementa- tion Manual for Quality Assurance (5) and Quality Assurance Guide Specification (6) were developed to assist SHAs with specification development and making efficient use of resources and improving product quality. The implementation manual includes two validation procedures, F- and t-tests, and a simple comparison test between a single SHA test and a range of contractor tests (5). Recent FHWA reviews and surveys show that only about one-third of SHAs’ comparison pro- cedures are performed according to the first procedure described in the AASHTO implementation manual, and well over half of SHAs use contractor test data in acceptance processes, most of which use nonstatistical methods (7–9). The common deviations from the AASHTO implementation manual statistical procedure are: eliminating the F-test, using split rather than independent mate- rial samples, having an inadequate number of SHA test results, using comparison inappropriate methods [such as the single SHA test method or the Difference Two-Sigma Limit (D2S Limit) method], and performing unwarranted retests. These deviations could lead to erroneous accep- tance and payment decisions (10). The literature suggests a general distrust of contractor test data reported by some SHA personnel, and investigations have found misconduct and fraud (11–14). 1.1 Problem Statement NCHRP Project 10-100 was initiated to identify or develop statistically sound and practi- cal procedures for validating these data and prepare a recommended practice for using these procedures. 1.2 Objective and Organization of the Report The objectives of this research were to recommend procedures for validating contractor test data for construction materials and to prepare guidelines for their application. This report documents the research effort in two parts. Part I summarizes the research performed in four Background C H A P T E R 1

6 Procedures and Guidelines for Validating Contractor Test Data chapters; further details are provided in five appendices. Chapter 1 provides background infor- mation. Chapter 2 presents the research approach. Chapter 3 discusses the findings and applica- tions of the research and presents examples to illustrate use of the recommended procedures for different scenarios. Chapter 4 presents the summary and recommendations for future research. Appendices A and B present the survey of SHAs and its results, respectively. Appendix C presents the numerical simulations used to evaluate procedures. Appendix D presents the SHA data analysis, and Appendix E contains the statistical tables used in the report. Appendices A–E can be downloaded from the project summary page at https://apps.trb.org/cmsfeed/TRBNet ProjectDisplay.asp?ProjectID=4192. Part II, contained in this report, is a proposed AASHTO practice for validating contractor test data.