NC State
BioResources
Garbowski, T. (2024). “Revolutionizing corrugated board production and optimization with artificial intelligence,” BioResources 19(2), 2003-2006.

Abstract

In the field of corrugated board production and packaging optimization, the advent of Artificial Intelligence (AI) has initiated a paradigm shift. This paper presents a brief analysis of AI’s role in revolutionizing both the production of corrugated board and the design of corrugated packaging. It explores the integration of AI in the homogenization process of complex corrugated board structures into single-layer, shallow shell-based computational models, aiming to improve and accelerate load-bearing calculations. This work presents also how AI’s predictive and analytical capabilities are pivotal in achieving efficiency, sustainability, and cost-effectiveness in the corrugated board industry.


Download PDF

Full Article

Revolutionizing Corrugated Board Production and Optimization with Artificial Intelligence

Tomasz Garbowski

In the field of corrugated board production and packaging optimization, the advent of Artificial Intelligence (AI) has initiated a paradigm shift. This paper presents a brief analysis of AI’s role in revolutionizing both the production of corrugated board and the design of corrugated packaging. It explores the integration of AI in the homogenization process of complex corrugated board structures into single-layer, shallow shell-based computational models, aiming to improve and accelerate load-bearing calculations. This work presents also how AI’s predictive and analytical capabilities are pivotal in achieving efficiency, sustainability, and cost-effectiveness in the corrugated board industry.

DOI: 10.15376/biores.19.2.2003-2006

Keywords: Corrugated board; Artificial intelligence; Box strength; Production optimization

Contact information: University Center for ECO-materials. Poznan University of Life Sciences; email: tomasz.garbowski@up.poznan.pl

Introduction

In the current market landscape, characterized by growing market and environmental demands, the packaging industry is tasked with the continuous improvement of its processes and products. Corrugated board, being a fundamental material in packaging production, stands at the forefront of these challenges, requiring innovative approaches (Luo et al. 2022; Gaudelas et al. 2023;) to meet evolving requirements (Johst et al. 2023). In recent decades, significant progress has been observed in the digitalization and optimization of corrugated board production, with Artificial Intelligence (AI) being recognized as a potentially transformative force. The possible role of AI could extend from e.g. optimizing the homogenization process, which would simplify complex structures of corrugated board into manageable models (Garbowski 2022, 2023a), to refining the accuracy of safety factor estimations (Garbowski 2023a), crucial for ensuring structural integrity. Moreover, AI algorithms, particularly Artificial Neural Networks, as demonstrated in recent studies (Gajewski et al. 2023; Gu et al. 2023), have already shown significant potential in predicting the crush resistance and load-bearing capacity of various corrugated packaging designs.

The potential application of AI goes beyond the basic aspects of corrugated board production. It could address intricate challenges such as varying environmental conditions during storage and transport, the uncertainty related to material quality, and the complexities introduced by design features like ventilation holes and perforations. These aspects, which are often overlooked in traditional methods, are crucial for ensuring the safety, durability, and efficiency of packaging solutions.

This work seeks to investigate the potential transformative impact of AI on traditional methods of corrugated board production. AI has the potential to not only enhance efficiency but also play a crucial role in optimizing the use of materials and reducing the ecological footprint. A particular emphasis is placed on exploring an innovative AI-based approach that could synergize the strengths of various digitalization pathways. This approach suggests the possibility of using AI to streamline the modeling process by identifying key relationships between the properties of corrugated board and various production parameters such as adhesive quantity, temperature, and humidity. The article discusses how this proposed AI-driven method could potentially combine the advantages of detailed and simplified models, leading to a more efficient, accurate, and adaptable production process. This approach could represent a significant step forward in corrugated board production, potentially meeting the dynamic needs of the packaging industry and paving the way for a sustainable, efficient, and cost-effective future.

Comparative Analysis of Two Digitalization Pathways in Corrugated Board Production

From corrugated board to packaging

This approach takes the finished corrugated board as its starting point and seeks to optimize the packaging production process through digitalization and the application of AI. It revolves around using advanced modeling and simulation techniques to enhance the final product’s structural integrity and efficiency. This method allows for a precise customization of packaging features, such as size, shape, and strength, ensuring that each package is specifically tailored to meet its intended purpose. The main benefit of this approach is the ability to create high-performance packaging solutions that are not only robust but also material-efficient, potentially reducing waste and environmental impact.

However, focusing primarily on the end stage of the production process means that this approach might overlook variations in the raw material properties. Such variations can have significant implications on the final product’s performance, leading to potential discrepancies between the theoretical model predictions and the actual performance of the packaging. This gap could affect the reliability of the packaging solutions, making it crucial for the modeling process to be continuously updated and refined based on real-world data and feedback.

From paper to packaging

In the ‘From Paper to Packaging’ approach, the concept begins with the paper itself, forming the basis for the entire production process. The prospective incorporation of AI early in the corrugated board production process aims to optimize the material properties of the paper, which are crucial for the overall quality and performance of the final product. The potential integration of AI at this initial stage could allow for a comprehensive analysis and enhancement of the paper’s characteristics, such as its strength, elasticity, and moisture resistance. This holistic approach could ensure that the quality improvements made at the paper level cascade through to the final packaging product, potentially leading to superior performance and reliability.

The primary challenge in this approach is the complexity of accurately predicting how changes at the paper stage will impact the performance of the final product. This would require sophisticated AI models and simulations capable of accounting for a multitude of variables and their interdependencies. Despite the advanced capabilities of AI, achieving this level of predictive accuracy could be challenging, necessitating continuous refinement and validation against real-world outcomes. Additionally, the intricate nature of these models means that they could demand significant computational resources and expertise, posing a potential barrier for smaller manufacturers or those with limited technological infrastructure.

Optimizing Corrugated Board Production with AI

Harnessing AI for streamlined modeling in the journey from paper to packaging

The potential integration of Artificial Intelligence (AI) in the journey from paper to packaging offers a transformative opportunity for the industry, especially in terms of simplifying complex computational models. This approach could revolutionize traditional methods by enabling AI to discover intricate relationships between the properties of the produced corrugated board and various production parameters. Harnessing AI, manufacturers might delve deeper into understanding how factors like the amount of adhesive, temperature, and humidity during the production process could influence the final product.

The initial part of this novel approach involves leveraging AI to analyze vast datasets encompassing various production variables. AI algorithms, particularly machine learning models, are adept at identifying patterns and correlations that may be elusive to traditional analysis. For instance, they could predict how slight variations in adhesive application or ambient conditions during production might lead to significant changes in the strength and durability of the corrugated board. This predictive capability could be invaluable, allowing for real-time adjustments and optimization of the production process, potentially leading to consistent quality and reduced material wastage.

Furthermore, this AI-driven approach could simplify the modeling process by focusing on the most impactful parameters. Instead of creating overly complex models trying to account for every conceivable variable, AI could identify which factors have the most significant influence on the quality of the corrugated board. This approach would not only streamline the modeling process but also enhance its practical applicability, making it more accessible to smaller manufacturers or those with limited computational resources. In doing so, AI could democratize the optimization process, enabling a broader range of manufacturers to achieve high-quality production standards.

The adoption of this AI-centric model could represent a significant advancement in the digitalization of the corrugated board production process. It promises not only greater efficiency and accuracy but also a more agile and responsive manufacturing environment. As the technology continues to evolve, it could redefine the boundaries of what is possible in corrugated board production, ushering in an era of smarter, more sustainable packaging solutions.

Bridging the gap between homogenization and real-world variability

In the realm of corrugated board packaging, digital analysis plays a pivotal role, particularly in the ‘From Paper to Packaging’ pathway, where homogenization is a key component. This technique is primarily employed for optimizing the selection of corrugated board in designing specific packaging, focusing on load-bearing capacity calculations. In homogenization, the complex multi-layered structure of corrugated board is potentially simplified into a more manageable single-layer model. The model’s effective parameters are initially derived from the properties of the constituent papers and the geometry of the flute cross-section. While this approach may facilitate theoretical calculations and initial design, it might encounter limitations in predicting the actual performance of the manufactured board. The homogenized model, primarily based on initial material properties and flute geometry, often may not account for the variations introduced during the production process, potentially leading to discrepancies between theoretical predictions and real-world performance.

This deviation could arise not only from the simplifications inherent in the homogenization process but also significantly from the variables introduced during the production process. Factors such as variations in adhesive application, temperature, and humidity during manufacturing can substantially alter the final properties of the corrugated board. Traditional models might struggle to account for these real-world production variables, leading to a gap between theoretical predictions and actual product performance.

The potential introduction of Artificial Intelligence (AI) offers a solution to this challenge. By continuously comparing the properties of the actual corrugated board produced with those of the homogenized model derived from the same constituent papers, AI could identify patterns and relationships that traditional methods overlook. This process would involve feeding production data, including specific details about adhesive quantity, temperature, and humidity, into AI algorithms. Over time, AI could learn and recognize how these production variables impact the final corrugated board properties, allowing for more accurate predictions and optimizations.

Furthermore, by integrating these insights, the production process could be monitored and optimized based on the patterns learned by AI. This approach could not only improve the accuracy of the model but also enhance the efficiency and quality of the corrugated board production. It represents a potential advancement in the field, moving towards a future where production processes are more adaptable and responsive to the complexities of real-world manufacturing conditions.

References Cited

Gajewski, T., Grabski, J. K., Cornaggia, A., and Garbowski, T. (2023). “On the use of artificial intelligence in predicting the compressive strength of various cardboard packaging,” Packaging Technology and Science 37(2) 97-105. DOI:10.1002/pts.2783

Garbowski, T. (2022). “Digitization of corrugated board,” Polish Paper Review 78(12), 683-686. DOI: 10.15199/54.2022.12.2

Garbowski, T. (2023a). “Evaluating safety factors in corrugated packaging for extreme environmental conditions,” Packaging Review 4, 6-15. DOI: 10.15199/42.2023.4.1

Garbowski, T. (2023b). “The use of artificial intelligence for the optimal production of corrugated board,” Polish Paper Review 79(5), 279-288. DOI: 10.15199/54.2023.5.2

Gaudelas, A., Blanchet, P., Gosselin, L., and Cabral, M. R. (2023). “Physical characterization of biobased corrugated panels, an innovative material,” BioResources 18(3), 5838-5858. DOI: 10.15376/biores.18.3.5838-5858

Gu, B. J., Frank, B., and Lee, E. (2023). “A comparative analysis of artificial neural network (ANN) architectures for box compression strength estimation,” Korean Journal of Packaging Science and Technology 29(3), 163-174. DOI: 10.20909/kopast.2023.29.3.163

Johst, P., Kaeppeler, U., Seibert, D., Kucher, M., and Böhm, R. (2023). “Investigation of different cardboard materials under impact loads,” BioResources 18(1), 1933-1947. DOI: 10.15376/biores.18.1.1933-1947

Luo, P., Yang, C., and Wang, T. (2022). “Making ultra-thin high density fiberboard using old corrugated container with kraft lignin,” BioResources 17(2), 2696-2704.