Applied Scrap Precision

Important results from the project

Scrap precision aims to produce benchmarks for mean analysis and variance for the scrap types defined in the "Scrap Book" and then show how the information can be used to replace cleaner (expensive) scrap types with simpler (cheaper) ones and thus more easily handle a situation when the blast furnaces are closed in Sweden and clean scrap becomes scarce. Both of these objectives were achieved but it turned out that the variance varies over time. How to handle this remains to be determined.

Expected long term effects

The results lay a foundation for monitoring the scrap quality in Sweden over time and creating a dialogue for better sorting between steel companies and the scrap trade and product companies that use steel in their products. The calculated values can be used in combination with other methods such as sample melts, sampling with XRF, LIBS and image analysis to build knowledge about what is normal quality, as well as better and worse quality. Digital decision support (AI) can be a way forward.

Approach and implementation

The project participants were six steel mills and a purchasing company as well as a system supplier. The latter carried out calculations using the programs RAWMATSTAT and RAWMATMIX. The steel companies provided process data, conducted test melts, XRF tests and reviewed results together with the purchasing company. The fact that the mean analysis and variance varied over time was unexpected and meant that the scrap analysis part dragged on and less time could be devoted to process improvements.