Real-Time Outlier Detection and Removal in Data Collection with LabVIEW using Peirce’s Criteria

Abstract

Testing and experimental processes in industries and research institutes play important roles in understanding systems and developing accurate models. Experimental uncertainties in variables introduce deviations (errors) in measured data, which can stem from various factors, one of which is the presence of outliers. Outliers impact the accuracy of measurements significantly during analysis and lead to inaccurate conclusions. Outliers can be due to environmental changes, drift in measurements, etc., during assessment. Identifying outliers and eliminating them is a crucial step during data analysis. This research aims to investigate real-time outlier detection and removal using LabVIEW. Peirce Criteria are assessed for detecting and eliminating outliers from displacement data obtained through LVDT sensors. The research demonstrates that Peirce’s criteria are particularly well suited for small datasets. By employing LabVIEW and Peirce criteria, this study presents a practical approach to accurately detect and remove outliers in real time to enhance the reliability of data analysis.