STOCHASTIC MODELS FOR ARTIFICIAL CEMENT CONTROL PARAMETERS

The artificial cement characterization requires the measurement of a large number of control parameters. In the present work, twenty four parameters that are monitored on a daily basis during three years of observation are investigated. Due to the large amount of data to be collected, a monthly overage is calculated. This allows the result fluctuations due to managing difficulties to be reduced.The purpose of this study is to show that it is possible, with an acceptable approximation rate, to make a variable reduction, which is to reduce the number of control parameters, as these measurements are identified every hour at the exit of the clinker crusher and daily at the expedition.The aim of the present investigations is to suggest a manufacturing supervision that can give improvements in the characteristics of the finished product and that can reduce, in an objective way, the number of tests. The consequences of such pertinent supervision are a great deal of saving time and lower cost.