Towards a Prediction Analysis in an Industrial Context

The efficiency of a mine’s fixed facilities depends on several parameters, which makes its analysis and improvement very complex. The purpose of this article is to propose an approach to predict the efficiency of the fixed facilities based on data mining algorithms. The input of this approach is the data concerning the parameters influencing the efficiency. The output is the predicted efficiency of the facilities. This approach consists of four steps: domain comprehension, data preprocessing, development of prediction models and finally validation and verification of the proposed models. It has been applied for a Moroccan mining company and the results of this case study are also presented.