A Machine Learning Approach for Predicting and Mitigating Pallet Collapse during Transport - The Case of the Glass Industry

Aug 4, 2026

Upload by Eden

This study explores the prediction and mitigation of pallet collapse during transportation within the glass packaging industry, using a machine learning approach to reduce cargo loss and improve logistics efficiency. Data were collected from a leading glass manufacturer and analyzed under the CRoss-Industry Standard Process for Data Mining (CRISP-DM) framework. A comparative evaluation of Decision Tree and Random Forest models, using F1-score and related metrics, shows Random Forest performs better. The research identifies new geometry- and temperature-related predictors and proposes practical prevention strategies including stack pattern optimization, packaging material improvements, temperature control, and stronger handling protocols.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages23

Reading Time58 min

AccessFree to download

Introduction

Problem of cargo loss and pallet stability

Limitations in prior logistics research

Proposed approach (CRISP-DM and models)

Data collection from the glass manufacturer

Model comparison and evaluation

applied sciences Article A Machine Learning Approach for Predicting and Mitigating Pallet Collapse during Transport: The Case of the Glass Industry Francisco Carvalho 1, João Manuel R. S. Tavares 2 and Marta Campos Ferreira 3, * Citation: Carvalho, F.; Tavares, J.M.R.S.; Campos Ferreira, M. A Machine Learning Approach for Predicting and Mitigating Pallet Collapse during Transport: The Case of the Glass Industry. Appl. Sci. 2024, 14, 8256. [https://doi.org/10.3390/](https://doi.org/10.3390/)[ ](https://doi.org/10.3390/)app14188256 Academic Editor: Arkadiusz Gola Received: 31 July 2024 Revised: 10 September 2024 Accepted: 11 September 2024 Published: 13 September 2024 Copyright: © 2024 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/)) . 1 Faculdade de Engenharia, Universidade do Porto, 4200-465 Porto, Portugal; [up201806858@edu.fe.up.pt](up201806858@edu.fe.up.pt) 2 Instituto de Ciência e Inovação em Engenharia Mecânica e Engenharia Industrial, Departamento de Engenharia Mecânica, Faculdade de Engenharia, Universidade do Porto, 4200-465 Porto, Portugal; [tavares@fe.up.pt](tavares@fe.up.pt) 3 Institute for Systems and Computer Engineering, Technology and Science Faculdade de Engenharia, Universidade do Porto, 4200-465 Porto, Portugal * [Correspondence: mferreira@fe.up.pt](Correspondence: mferreira@fe.up.pt) Abstract: This study explores the prediction and mitigation of pallet collapse during transportation within the glass packaging industry, employing a machine learning approach to reduce cargo loss and enhance logistics efficiency. Using the CRoss-Industry Standard Process for Data Mining (CRISP-DM) framework, data were systematically collected from a leading glass manufacturer and analysed. A comparative analysis between the Decision Tree and Random Forest machine learning algorithms, evaluated using performance metrics such as F1-score, revealed that the latter is more effective at predicting pallet collapse. This study is pioneering in identifying new critical predictive variables, particularly geometry-related and temperature-related features, which significantly influence the stability of pallets. Based on these findings, several strategies to prevent pallet collapse are proposed, including optimizing pallet stacking patterns, enhancing packaging materials, implementing temperature control measures, and developing more robust handling protocols. These insights demonstrate the utility of machine learning in generating actionable recommendations to optimize supply chain operations and offer a foundation for further academic and practical advancements in cargo handling within the glass industry. Keywords: cargo loss; predictive tool; data mining; CRISP-DM framework; glass manufacturer 1. Introduction Cargo loss within logistics systems poses significant financial challenges for businesses, incurring not only the costs associated with replacing damaged or lost goods but also additional expenses such as incident management, increased insurance premiums, missed business opportunities, and potential reputation damage [1] . The urgency of addressing these losses is stressed by their widespread impact across various sectors, particularly in industries where goods are susceptible to damage during transport, such as in the ceramic [2] and the glass industries [3] . However, previous research on logistics has been predominantly focused on broad solutions, such as route optimization and general cargo securing techniques, without sufficiently addressing the specific issue of pallet stability during transportation. This is a critical gap, as the collapse of pallets can lead to significant product loss and safety haza

FAQs

What problem does the study address in the glass industry?

It addresses pallet collapse during transportation, which can cause significant cargo loss and related financial and safety impacts.

Which machine learning models are compared, and how are they evaluated?

The study compares Decision Tree and Random Forest models, evaluated using performance metrics such as F1-score.

What key factors does the study identify as important for predicting pallet collapse?

It identifies new critical predictive variables, especially geometry-related and temperature-related features, that strongly affect pallet stability.

Aug 4, 2026

Upload by Eden