A blind test of nondestructive underground void detection by Ground Penetrating Radar (GPR)

Abstract

Blind test/experiment is widely adopted in various scientific disciplines like medicine drug testing/clinical trials/psychology, but not popular in nondestructive testing and evaluation (NDTE) nor near-surface geophysics (NSG). This paper introduces a blind test of nondestructive underground void detection in highway/pavement using ground penetrating radar (GPR). Purpose of which is to help the Highways Department (HyD) of the Hong Kong Government to evaluate the feasibility of large-scale and nationwide application, and examine the ability of appropriate service providers to carry out such works. In the past failure case of such NDTE/NSG based on lowest bid price, it is not easy to know which part(s) in SWIMS (S – service provider, i.e. people; W – work procedure; I – instrumentation; M – materials in the complex underground; S – specifications by client) fails, and how it/they fail(s). This work attempts to carry out the blind test by burying fit balls (as voids) under a site with reinforced concrete road and paving block by PolyU team A. The blind test about the void centroid, spread and cover depth was then carried out by PolyU team B without prior information given. Then with this baseline, a marking scheme, acceptance criteria and passing mark were set to test six local commercial service providers, determine their scores and evaluate the performance. A pass is a prerequisite of the award of a service contract of similar nature. In this first attempt of the blind test, results were not satisfactory and it is concluded that ‘S–service provider’ and ‘W–work procedure’ amongst SWIMS contributed to most part of the unsatisfactory performance.+