Rapid Classiﬁcation of Petroleum Waxes - A Vis-NIR Spectroscopy and Machine Learning Approach

Aug 4, 2026

Upload by Violet

Petroleum-derived waxes are widely used in the food industry to improve texture and act as protective coatings for products such as fruits and cheeses, requiring strict quality control to meet regulatory demands. This study combines visible and near-infrared (Vis-NIR) spectroscopy with machine learning to characterize macrocrystalline and microcrystalline waxes. Unsupervised HCA and PCA separate samples by chemical composition, while nonparametric supervised models—support vector machines (SVM) and random forest (RF)—classify wax types using five-fold cross-validation. The workflow also identifies discriminative wavelengths to build representative spectral fingerprints, enabling fast, eco-friendly, cost-effective automated quality assessment.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages16

Reading Time40 min

AccessFree to download

Introduction

Materials and Methods

Vis-NIR spectroscopy and data preprocessing

Machine learning approaches

Results and Discussion

Unsupervised clustering (HCA and PCA)

Supervised classification (SVM and RF)

Discriminative wavelengths and spectral fingerprints

Conclusions

foods Article Rapid Classiﬁcation of Petroleum Waxes: A Vis-NIR Spectroscopy and Machine Learning Approach Marta Barea-Sepólveda , Jos² Luis P. Calle, Marta Ferreiro-Gonz¡lez * and Miguel Palma Citation: Barea-Sepúlveda, M.; Calle, J.L.P.; Ferreiro-González, M.; Palma, M. Rapid Classiﬁcation of Petroleum Waxes: A Vis-NIR Spectroscopy and Machine Learning Approach. Foods 2023, 12, 3362. [https://doi.org/](https://doi.org/)[ ](https://doi.org/)[10.3390/foods12183362](10.3390/foods12183362) Academic Editor: Zhengjun Qiu Received: 8 August 2023 Revised: 31 August 2023 Accepted: 5 September 2023 Published: 7 September 2023 Copyright: © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license ([https://](https://)[ ](https://)[creativecommons.org/licenses/by/](creativecommons.org/licenses/by/)[ ](creativecommons.org/licenses/by/)[4.0/](4.0/)) . Department of Analytical Chemistry, Faculty of Sciences, Agri-Food Campus of International Excellence (ceiA3), IVAGRO, University of Cadiz, 11510 Puerto Real, Spain; [marta.barea@uca.es](marta.barea@uca.es) (M.B.-S.); [joseluis.perezcalle@uca.es](joseluis.perezcalle@uca.es) (J.L.P.C.); [miguel.palma@uca.es](miguel.palma@uca.es) (M.P.) * Correspondence: marta.ferreiro@uca.es; Tel.: +34-956-01658 Abstract: Petroleum-derived waxes are used in the food industry as additives to provide texture and as coatings for foodstuffs such as fruits and cheeses. Therefore, food waxes are subject to strict quality controls to comply with regulations. In this research, a combination of visible and nearinfrared (Vis-NIR) spectroscopy with machine learning was employed to effectively characterize two commonly marketed petroleum waxes of food interest: macrocrystalline and microcrystalline. The present study employed unsupervised machine learning algorithms like hierarchical cluster analysis (HCA) and principal component analysis (PCA) to differentiate the wax samples based on their chemical composition. Furthermore, nonparametric supervised machine learning algorithms, such as support vector machines (SVMs) and random forest (RF), were applied to the spectroscopic data for precise classiﬁcation. Results from the HCA and PCA demonstrated a clear trend of grouping the wax samples according to their chemical composition. In combination with ﬁve-fold cross-validation (CV), the SVM models accurately classiﬁed all samples as either macrocrystalline or microcrystalline wax during the test phase. Similar high-performance outcomes were observed with RF models along with ﬁve-fold CV, enabling the identiﬁcation of speciﬁc wavelengths that facilitate discrimination between the wax types, which also made it possible to select the wavelengths that allow discrimination of the samples to build the characteristic spectralprint of each type of petroleum wax. This research underscores the effectiveness of the proposed analytical method in providing fast, environmentally friendly, and cost-effective quality control for waxes. The approach offers a promising alternative to existing techniques, making it a viable option for automated quality assessment of waxes in food industrial applications. Keywords: food waxes; petroleum-derived products; parafﬁns; visible–near-infrared spectroscopy; machine learning; support vector machine; random forest; discrimination; spectralprint 1. Introduction Petroleum waxes are a petroleum-derived product (PDP) with a wide spectrum of industrial applications obtained from lubricating oils. Within the agri-food industry, waxes are commonly used to make fruits, vegetables, and candy look shiny or as a food additive. They provide a protective layer that helps to extend shelf life by reducing the loss of water, thus slowing the dehydration process and keeping the product fresh for a longer period. This application is particularly important f

FAQs

Why is rapid quality control needed for petroleum-derived food waxes?

Petroleum-derived waxes are used as additives and coatings in foods, so they must meet strict regulations. Quality control ensures proper texture, protection, and compliance with safety and purity requirements.

How does the study use spectroscopy and machine learning to classify wax types?

The approach applies visible and near-infrared (Vis-NIR) spectroscopy to generate spectral data, then uses machine learning. Unsupervised HCA and PCA first reveal grouping trends by chemical composition, and supervised SVM and random forest perform accurate classification of macrocrystalline versus microcrystalline waxes.

What is the role of discriminative wavelengths and spectral fingerprints?

Along with five-fold cross-validation, the supervised models yield high-performance classification and highlight specific wavelengths that differentiate wax types. These wavelengths support constructing representative spectral fingerprints for automated assessment.

Aug 4, 2026

Upload by Violet