BETTER AND BETTER The production environment in a foundry is increasingly changing, demanding several adjustments to maintain the quality of the product
Last month, Swedish computer software company Viking Analytics and automatic pouring systems manufacturer Pour-tech AB announced a partnership to offer artificial intelligence (AI) operators to iron foundries.
Viking Analytics notes that its solution, named EASYpour, uses data analytics and machine learning algorithms developed to automatically adjust the pouring process. This improves product quality, enables faster production, and reduces costs.
EASYpour is one of the first smart-pouring solutions available in the market and more advantageous than traditional methods that require constant intervention from an experienced operator.
According to Pour-tech AB GM Niklas Stigendal, the target was to have an end-to-end automated solution.
“The production environment in a foundry is increasingly changing, demanding several adjustments to maintain the quality of the product, sometimes in a matter of minutes. We wanted to remove this complexity for our clients, so the operators could focus on other tasks.”
Viking Analytics was selected to develop the data-driven predictive model that runs in the solution, he explains.
Based on the expert industry knowledge from Pour-tech AB, Viking Analytics employed its data analytics and AI capabilities to develop machine learning algorithms. This means that, once the algorithm is deployed, it will evaluate information from each pour and adjust the parameters for the next one.
“In practice, it means that the system is always improving itself, by learning from past actions in a feedback loop that maintains long-term quality,” says Viking Analytics co-founder and product manager Arash Toyser.
He adds that this improvement “happens in a fraction of the time that it would take for an operator to make the same decisions”.
By making these adjustments, EASYpour eliminates overpours and scrap, resulting in significant reductions in re-melt costs. It also provides stability in product quality and higher productivity.
“At the foundry where we first implemented the solution, gains of €0.10 a pour have been registered,” added Stigendal.
Currently, two foundries in Denmark and Germany are already successfully using EASYpour.
To Viking Analytics co-founder and CEO Stefan Lagerkvist, the partnership highlights the potential for the adoption of AI in the metals industry.
He adds that automation is already well established in foundries, but in order to survive the increasing competition, companies must take on the digital transformation road, from collecting and analysing data to taking action to improve productivity based on it.
“We are pleased to enable pour-tech to bring digitalisation to their customers,” he concludes.