Identifying underlying causes of errors and malfunctions by applying machine learning algorithms – Voortman



Company Name / Department Voortman Steel Machinery
Contact Person Sietse Koenjer
Location Ozonstraat 1, Rijssen
Study programme(s) OML
Community Servitization
Start Date 01-09-2022
Housing arranged by company No



€400,- / € 450,- per Month

Company Description

Seeing ideas become a reality: it sounds nice but it really does happen at Voortman Steel Machinery. And these are ground-breaking ideas. We make the most innovative machinery and product lines for the steel construction and plate-processing industry for customers around the world. Going through the process together with a customer in order to arrive at a production line that integrates seamlessly into his process. Regardless of how often we do it, it is special each time and provides a great deal of satisfaction.

Project Description

Voortman is making the next step into the future by offering our customers numerous cloud services. We’re starting with insight in machine statuses, production statuses and realized production times. But next to these products we are exploring the possibilities of this cloud environment (and the data in it) to support our services department. As an organization we pride ourselves in our capabilities to support our customers throughout the lifespan of the machines we deliver. We’re best in class with regard to after-sales services and we plan on staying the front-runner in our industry. We expect machine learning to be(come) a tool that will make all the difference moving forward. By leveraging the capabilities of machine learning we want to improve the partnership with our customers even more by offering them value in the form of advice or new services. The following assignments are the first concrete steps we intend to make towards using proper machine learning algorithms as we explore this new territory.

At Voortman we develop a wide range of machines that are delivered to our customers as a single unit or as production lines. Once in the field users of these machines can encounter errors or malfunctions. At the moment we’re starting to send the errors that our customers encounter to the cloud and we analyse this data to determine how many times different types of errors occur. Furthermore, to be able to prevent these errors we want to get insight in what factors influence the occurrence of these errors. The errors should therefore be combined with other types of data in a way that we can make inferences about the way these factors influence errors and malfunctions.

Goals of the Project

We want to be able to identify underlying causes of errors and malfunctioning of machines.


An algorithm or tool that can be used to reduce/prevent/predict errors and malfunctioning of machines.

Essential Student Knowledge

Relevant data analysis and machine learning knowledge.

More information: