voestalpine in Donawitz is working on a pioneering pilot project called Smart4Steel in which data science (big data analytics) is generating added value from process data.
Over a project period of three years the process data will be collected, calculated, and interpreted in order to generate added value with the aid of data science (big data analytics).
Collecting and evaluating data sounds pretty simple, but it isn’t. The reason is “Big Data”, i.e. huge quantities of data. Every day during steel manufacturing and processing at the voestalpine site in Donawitz, countless measurement values (raw data) are collected every second by a variety of sensors. Due to the sheer volume of data, manual or conventional statistical approaches to data processing are increasingly reaching their limits.
"Now the time has come for us to look more carefully at the data itself and to use it."
For that reason, for the past few months an interdisciplinary team has been working on the Smart4Steel pilot project. The idea behind the project is to use the data to develop smart solutions, i.e. potential correlations between production data and product quality features.
The “steel and wire” pilot
The smart approach is focused on product optimization and error minimization as a means of creating “smart steel” that voestalpine wishes to produce and process in future. “Big data offers significant potential when the quantities of data can be used to make statements about product quality,” Karner adds. This naturally applies to all the process steps involved in steel manufacturing and processing in the Metal Engineering Division, whether producing wire, rails, or tubes. The ultimate goal of Smart4Steel is to correctly evaluate all the data in order to generate value for the customer. However, many challenges need to be overcome in order to do so. The first step is to develop the idea together with the wire companies.
Right at the start
The continuous casting line—one of the production facilities with a major influence on product quality during steel production—is currently the starting point for recording project data. Afterwards, this data will be linked with the data captured during billet rolling, where, for example, surface imperfections can be detected. The intention is then to incorporate the data captured at the wire-producing companies at the sites in Donawitz and Bruck an der Mur with that from the wire rod mill, the wire drawing plant, and its inspection and testing facilities.