Statistical modeling of charcoal consumption of blast furnaces based on historical data

Abstract

This paper describes the development of statistical models to predict charcoal consumption in blast furnaces based on Response Surface Models (RSM) and Linear Regression Models (LRM). The statistical approach used provides a high level of confidence and allows the company to act preemptively fostering innovative business and in the action plan to reduce hot metal production cost, to improve raw materials processing and other actions in order to provide the blast furnaces with raw materials at minimal cost. It is a special particularity and represents a great step in V & M do Brasil blast furnaces’ operation which no longer uses standard ferrous load and started to operate with greater flexibility and variability concerning the types of ferrous load applied to achieve better economic results.