The increasing portability and accessibility of miniaturized NIR spectrometers are promoting the spread of in-field and online applications. Alongside the successful outcomes, there are also several problems related to the acquisition strategies for each instrument and to experimental factors that can influence the collected signals. An insightful investigation of such factors is necessary and could lead to advancements in experimental set-up and data modelling. This work aimed to identify variation sources when using miniaturized NIR sensors and to propose a methodology to investigate such sources based on a multivariate method (ANOVA—Simultaneous Component Analysis) that considers the effects and interactions between them. Five different spectrometers were chosen for their different spectroscopic range and technical characteristics, and samples of worldwide interest were chosen as the case study. Comparing various portable sensors is interesting since results could significantly vary in the same application, justifying the idea that this kind of spectrometer is not to be treated as a general class of instruments.

Alongside the increasing proofs of efficacy of miniaturized NIR instruments in food-related scenarios, it is progressively growing the number of end-users, even incentivized by the low-cost of the sensors. While attention is paid to the analytical protocol–from sampling to data collection, up to the data processing, the importance of error investigation in raw data is generally underestimated. Understanding the sources and the structure of uncertainty related to the raw data improves the quality of measurements and suggests the correct planning of the experiments, as well as helps in chemometric model development. The goal of chemometric modeling is to separate information from noise; therefore, a description of the nature of measurement error structure is necessary. Among the different approaches, we present the study of the Error Covariance Matrices (ECMs) and their decomposition in a bilinear structure as a powerful method to study the main sources of variability when using miniaturized NIR sensors in the actual way of use. Granulated and lump sugar samples were chosen as the case study and analyzed with two miniaturized spectrometers working in the NIR regions around 1350–2550 nm and 900–1750 nm, respectively, in dispersive reflectance mode. Results show that having some insights on multivariate measurement errors associated with spectra could be interesting in paving the way for several applications.

This study focuses on exploring and understanding measurement errors in analytical procedures involving miniaturized near-infrared instruments. Despite recent spreading in different application fields, there remains a lack of emphasis on the accuracy and reliability of these devices, which is a critical concern for accurate scientific outcomes. The study investigates multivariate measurement errors, revealing their complex nature and the influence that preprocessing techniques can have. The research introduces a possible workflow for practical error analysis in experiments involving diverse samples and instruments. Notably, it investigates how sample characteristics impact errors in the case of solid pills and tablets, typical pharmaceutical samples. ASCA was used for understanding critical instrumental factors and the potential and limitations of the method in the current application were discussed. The joint interpretation of multivariate error matrices and their resume through image histograms and K index are discussed in order to evaluate the impact of common preprocessing methods and to assess their influence on signals.