Analysis of Detection Repeatability and Reproducibility Trials Damir MARKUČIČ, Biserka RUNJE, Vedran MUDRONJA, Sanjin MAHOVIĆ, University of Zagreb, Department of Quality, Croatia Abstract. R&R analysis of results of mine detection trials is presented. Data for R&R analysis has been obtained by means of round robin trials which comprehend different operators, instruments and targets determined by various parameters like type of mine, depth and type of soil in which mine is burried. R&R analysis shows quantitative measure of detection process in certain conditions. Beside information which could be achieved through estimating probability of detetction (PoD) R&R analysis describes to which extent the particular detection process is consistent or uncertain. The aim of R&R analysis of detection is to contribute to the general analysis and assessment of detection reliability. Introduction Repeatability and reproducibility (R&R) studies are commonly utilised in analyses of capability of processes. In this article we apply R&R on the results of pinpointing the buried mines (parts) for the purpose of statistical R&R analysis of mine detection process. Results obtained from R&R analysis are compared with conventional halo size (± 10 cm) representing circle area around true position of the mine in which it is expected to get response (true-positive detection) form the mine utilising mine detectors. In particular, we take the raw data (results of pinpointing in X-Y coordinate system) obtained by operators and the true positions of buried mines. Calculated distance of a pinpointed result from a true mine's position is referred as a radial distance. The results were obtained as a part of trials performed on the test field near Benkovac, Croatia, during May 2005. Measurements of the system capability are based on analysis of repeatability and reproducibility of measuring system in which one utilises The Range & Average Method (also known as Long Method in literature). The Range & Average Method computes the total measurement system variability, and allows the total measurement system variability to be separated into repeatability, reproducibility, and part variation. Repeatability (Equipment variation) is the variability of the measurements obtained by one operator while measuring the same item repeatedly. This is also known as the inherent precision of the measurement equipment (device). Reproducibility (Appraiser variation) is the variability of the measurement system caused by differences in operators’ performances. Mathematically, it is the variability of the average values obtained by several operators while measuring the same item. Total measurement variability, combined R&R, combines repeatability and reproducibility. ECNDT 2006 - We.3.5.3 1