Improving network-level quality management plans during flexible pavement condition assessment

Abstract

Since the inception of the Pavement Management System (PMS), manual road condition assessment and the subsequent ad-hoc estimation of Maintenance and Rehabilitation (M&R) needs have remained susceptible to human bias due to their subjective nature. During network-level road condition assessments, a Visual Condition Index (VCI) is calculated not only to estimate M&R needs but to predict future road performance trends or deterioration curves. It is, therefore, critical that distress ratings be of “acceptable” accuracy. Road agencies worldwide are adopting semi-automated or fully automated road condition assessment methods to enhance distress ratings, integrated with advanced data analysis techniques. Illustrated through a case study, significant efforts were made focusing on enhancing condition data analysis through a statistical method called the “t-test”. Developed by the Western Cape Government in South Africa, agencies are using the Student’s t-distribution (“t-test”), integrated into the Quality Management Plan (QMP). The study demonstrates the importance of automation in the QMP, highlig