Data-Driven Evaluation of Near-Infrared Spectroscopy in Grain Quality Control

Abstract

This study highlights the quantitative assessments of near-infrared spectroscopy for the analysis of quality parameters of bread wheat. The focus is on analytical performance. Statistical tools, Cochran’s test and z-score analyses are used for the verification of the homogeneity and stability of the test material. Especially moisture and protein parameters show small deviations and strong agreement with the reference method, which indicates good repeatability and stability. Most parameters are within acceptance criteria, besides hectoliter weight measurements, which show systematic bias. According to the findings, near-infrared (NIR) spectroscopy is a rapid, non-destructive, and reliable approach in terms of quality assessment, and its implications include improving decision-making efficiency and lowering operational costs in the grain supply chain.