Technical Paper ISSN 1997-1400 Int. J. Pavement Res. Technol. 6(1):22-30 Copyright @ Chinese Society of Pavement Engineering 22 International Journal of Pavement Research and Technology Vol.6 No.1 Jan. 2013 Developing a Quality Measure for Evaluating the Contractor Performance in Asphalt Pavement Construction Ahmed Elyamany 1 , Magdy Abdelrahman 2+ , Fu-Chih Cheng 3 , and Curt Doetkott 4 ─────────────────────────────────────────────────────── Abstract: Most construction agencies have a quality management system to control and manage the quality of their final product. Once the project is over, the tests results are usually kept in archives with no use. Project bidding is a time when the quality information could be very helpful to identify qualified contractor. The proposed model is an effort to use quality tests in evaluating the contractor performance. The composite percent defective is proposed as the quality measure which combines percent defectives of a number of selected quality characteristics. The trend analysis is used to provide a better look at the contractor performance. The significance of the proposed measure is in its ability to reflect the true quality of the contractor work being calculated based on actual tests results. Key words: Acceptance; Pay factor; Percent defect; Probability; Quality measure; Trend analysis. ─────────────────────────────────────────────────────── Introduction 12 The current practice of a number of State Highway Agencies (SHAs) is to evaluate and rate the contractor performance. Most of these cases do not have a well established measure or index to track the contractor performance. Rather, they use subjective questionnaires filled by project engineers and managers. This is mainly because in most states public projects must be bid by traditional low-bid method [1]. Parties involved in construction think the low-bid method decreases the value of the construction. In low-bid contracting, the contractor focuses on how to lower the cost rather than developing a quality product with a low cost. Some owners think quality has no price and only look for the lowest bid which, in fact, is responsible for the cost and time overruns in their projects. Saving the taxpayer money is an acceptable reason to believe in the low-bid; however this belief must change as most of the studies show losses in the long run [2, 3]. The problems associated with the low-bid system encourage contractors to implement cost-cutting measures instead of quality-enhancing measures. Therefore, it is less likely that the contracts will be awarded to the best-performing contractors who will deliver the highest quality projects with minimum cost [4]. The significance of having the performance measure for quality is to compensate the good contractors by allowing them to bid on more projects or at least let the other contractors account for their bad performance. Several attempts to develop a rational system of rating the quality of construction were documented. Most of these ratings were used for qualification and bidding processes. The construction industry lacks the existence of a standard rating system for quality based on measurable, not subjective data. Actually there are a lot of data 1 Department of Construction Engineering, Zagazig University, Zagazig-Egypt. 2 Department of Civil Engineering, North Dakota State University, Dept. 2470, PO Box 6050, Fargo, ND 58108, USA. 3 Statistics Department, North Dakota State University, USA. 4 Information Technology Services, North Dakota State University. + Corresponding Author: E-mail m.abdelrahman@ndsu.edu Note: Submitted April 3, 2012; Revised August 1, 2012; Accepted August 6, 2012. especially for quality control and quality assurance purposes. However, the major challenge is how to combine multiple distributions of different quality characteristics. Study Objective The main propose of this study is to develop an analytical model for combining the effects of multiple deficiencies of quality characteristics distributions, for contractors working in the pavement industry, using basic statistical concepts. The inputs of the model are the individual percentage defectives of multiple quality characteristics, while the output is the composite percent defective. The Composite Percent Defective (PD C ) is based upon assuming specific underlying distributions for individual quality characteristics, as well as the relationships among the individual quality tests. When the underlying assumptions are applicable, statistical principles from mathematical statistics, such as the form of the joint distributions of independent random variables, may be used to derive the expected results for the proposed quality measure. In such cases, simulations may be used to supplement our understanding and to verify the expected theoretical results. The basic statistical concepts are used to classify the contractor performance in two categories: first is the work accepted with full pay and second is the work accepted with partial pay. This classification is common in the pavement industry where defective work can be accepted with partial pay. Literature Review Quality Control/Quality Assurance testing results are available in most DOTs, which represent a valuable indication of past performance. As one part of the quality assurance process, there is a need for comprehensive methods to evaluate a contractor’s eligibility to engage in work from a quality perspective; thus, there is a need for examining quality performance measurement techniques and approaches. The main purposes of rating the contractor performance are for qualification, bidding, or payment schedules. Weed proposed a method for developing pay schedules based on the need for a rational method to relate As-Build quality to