Estimation of the Edge Crush Resistance of Corrugated Board Using Artificial Intelligence

Abstract

Recently, AI has been used in industry for very precise quality control of various products or in the automation of production processes through the use of trained artificial neural networks (ANNs) which allow us to completely replace a human in often tedious work or in hard-to-reach locations. Although the search for analytical formulas is often desirable and leads to accurate descriptions of various phenomena, when the problem is very complex or when it is impossible to obtain a complete set of data, methods based on artificial intelligence perfectly complement the engineering and scientific workshop. In this article, different AI algorithms were used to build a relationship between the mechanical parameters of papers used for the production of corrugated board, its geometry and the resistance of a cardboard sample to edge crushing. There are many analytical, empirical or advanced numerical models in the literature that are used to estimate the compression resistance of cardboard across the flute. The approach presented here is not only much less demanding in terms of implementation from other models, but is as accurate and precise. In addition, the methodology and example presented in this article show the great potential of using machine learning algorithms in such practical applications.

Managing Circularity of Packaging for Food Products – Carbon Footprint Assessment of Innovative Coated Paper Packaging

“…Similar results were reported in the literature (Guo et al, 2010;Garbowski et al, 2023;Garbowski et al, 2021).…”

Application of Artificial Neural Networks to Numerical Homogenization of the Precast Hollow-Core Concrete Slabs

“…As shown above, for random examples selected, the biggest error was 3.11%, which proves very good agreement between the value computed by trained ANN and the value computed due to homogenization algorithm [10]. Those magnitudes of values of error in homogenization computations are acceptable as shown in [12,18].…”

“…However, it should be properly conditioned to avoid overfitting. Other more advanced ANN architectures such as convolutional, recurrent networks or physics-based neural networks [18,32] were not explored in the study. The application of such architectures or the use of other machine learning regressors could improve accuracy and/or generalizability and could be included in future studies.…”

“…Another example could be the paper of Staszak et al [13], in which it was shown that the precast concrete slabs reinforced with spatial linear trusses may be efficiently homogenized to one effective layer of a representative shell element. On the other hand, the computational time (ii) can be reduced by using soft computing methods such as the artificial neural network (ANN) algorithms [14][15][16][17][18] or metamodeling [19,20]. This usually requires a significant computational effort, which is incurred before the solution of the target computational problem is needed, but it allows one to obtain a solution with an acceptably small error in a fraction of a second when the actual problem occurs.…”

Compressive Strength of Corrugated Paperboard Packages with Low and High Cutout Rates: Numerical Modelling and Experimental Validation

“…The best-known empirical formula for sizing for stack loading is the McKee equation, which attempts to determine the compressive strength of a corrugated box based on the Edge Crush Test (ECT) of the corrugated sheet [ 19 ]. The ECT of a corrugated board can be estimated by artificial intelligence [ 20 ]. Several authors have modified the original equation, and Kellicut and Landt attempted to use the ring crush test (RCT) value [ 21 ].…”