DigitCorr - Digitalisation in the corrugated board value chain

Important results from the project

A method to analyse long strips of paper, > 100m, with both an Auto-line and a Hyperspectral NIR camera has been developed. This can be used to measure the variations in the constituent materials of corrugated boards. PLS and machine-learning models have been built from corrugated board production data which predicts how the sheets will warp. The foundation has thus been laid for combining measurement technology with data analysis to be able to create a digital twin that can be used to predict how the corrugated board will warp in production. Work is now underway on a follow-up project.

Expected long term effects

In this feasibility study, we have shown that we, in the consortium, work well together and thereby laid the foundation for further projects with the subject. Through the experiments and data collection, we think we have shown that it is possible to build a digital twin of the production that predicts how the corrugated cardboard warps. That is why we are now working on getting a follow-up project

Approach and implementation

Thanks to a good collaboration in the consortium, we have carried out a good project despite external challenges. Early in the project, two of the partners dropped out. Due to Covid-19, we needed to redirect the project to what was feasible. By taking advantage of the expert knowledge within the consortium, we succeeded in building numerical models of corrugated board production that could, roughly speaking, predict the warp of the corrugated boards. An experimental setup could also be developed to characterize the constituent materials