Details:
Summary:
Substrate, in the form of paper, foil, film, or fabric, is the main part of the total manufacturing costs in the printing industry. It is already known that substrate is a major cost in printing & packaging processes (Levenson, Parsons, 2018), and there are many developed ideas on how to avoid unusual wastage of the substrate. Many methods and tools are used to improve the total effectiveness of printing processes, focusing on substrate wastage. Many printing press manufacturers also develop new printing machines that use less substrate for print setup, which is the main part of the used substrate in the printing process. But still many printers everyday fights with issues that have an impact on increasing substrate wastage.
This paper will show a way to predict the total substrate wastage per work order, based on a few vital factors. Prediction analysis and its effects on the example of the heat-shrinkable labels manufacturing process will be shown. The main steps in the manufacturing process of heat-shrinkable labels are divided into four stages, resulting in a finished product of a wound roll of heat-shrinkable labels. These stages are printing, slitting, seaming, and inspection (Kipphan, 2001; Kit, 2009). In the manufacturing practices of most printing companies, there are collected data and records of process parameters. Gathering information from this data in the form of developed models and rules, which uses statistical or computational methods from the area of Artificial Intelligence, like Artificial Neural Networks, Boosted Trees, Support Vector Machines, and others are subjects of interdisciplinary fields of science called Data Mining (Krystosiak, 2019).
There will be shown how developed data mining models can be used for the prediction of substrate wastage levels for every new design of printed products. The manufacturing process of every design of a heat-shrinkable label has a lot of factors and variables. Some of them are more significant, and some of them are not. This paper aims to choose significant factors and compute a model in the learning process by using collected data. In the end, it is important to predict the level of substrate wastage and calculate it in the total cost of each new work order of printed product, to avoid high substrate wastage occurrence during the printing process.