Utilizing Random Performance Records in the Best Value Model


Most construction agencies have a quality management system in order to control and manage the quality of their final product. Once the project is over, the testing results are kept in archives in which they are rarely re-visited or utilized. Quality testing results carry much information about the contractor performance that could be useful during the contractor evaluation/selection process. Previous attempts to implement the Best Value (BV) used the average performance records as the expected performance, which was utilized to evaluate contractors. The objective of this research is to develop, based on random data obtained from the contractor's performance records, a methodology that provides decision makers with the level of confidence or risk associated with the contractor selection using the BV model. Simulation technique is used to develop the BV model and analysis. Field performance data have been used to obtain the Percentage Defective, which indicates the contractor's performance in the BV model. The analysis of data indicates that performance follows a normal distribution. Sensitivity analysis of the BV model illustrates the significance of the weights in the BV model, which demands special attention when selecting the parameters' weights. The developed methodology provides the decision makers with the confidence and risk associated with their selection decision.


Civil, Architectural and Environmental Engineering

Keywords and Phrases

Asphalt Pavements; Contractors Selection; Performance Characteristics; Procurement; Quality; Sensitivity Analysis; Simulation

International Standard Serial Number (ISSN)


Document Type

Article - Journal

Document Version


File Type




Language 2



© 2012 Vilnius Gediminas Technical University (VGTU) Press Technika, All rights reserved.

Publication Date

01 Jan 2012