Abstract
The analysis of real-time shovel performance monitoring data has been constrained to date by the application of conventional one-dimensional statistics. This article proposes the use of geostatistics to analyse the spatial variance of shovel digging performance. Four key performance indicators are used in the modeling: shovel payload, dig rate, payload frequency and shovel production. The data processing incorporates the implementation of ordinary kriging with specified variogram models. The graphical interpretation of the results obtained provides maps, which are a convenient tool for assessing shovel operating context, operator proficiency and muckpile characteristics. A case study is presented to illustrate the applicability of the new geostatistical approach.
Acknowledgement
The authors would like to acknowledge the support of CRCMining for funding this research under the Smart Mining Systems Research Program – Project S203 “Shovel productivity model”.